Subscribe now

TOGETHER WITH NOTADEEPDIVE

Smart shoppers use Checkout so they spend only what they need to.

Become a Smart Shopper today and buy your favourite gadgets from over 500 merchants with Credit Direct Checkout.

Check Your Shopping Limit

Two chances

On Monday, 23 April 1984, Maj. Gen Muhammadu Buhari’s second-in-command, Brigadier Tunde Idiagbon, announced that Nigeria was changing the colors of its naira notes. The exchange of old notes for the new would begin at 8 a.m. on Wednesday and end at 6 p.m. on Sunday, 6 May. During that window, all land borders would be closed.

80 million Nigerians had twelve days to exchange almost every old note they had. With barely 1,240 bank branches nationwide, people abandoned offices and market stalls to join queues; in places without banks, they handed their savings to local government officials operating under armed supervision.

Anyone presenting more than ₦5,000—around forty months’ minimum-wage earnings—in old notes had to have a High Court affidavit explaining where the money came from and three passport photographs.

By October 1984, Buhari credited the exercise with reducing money supply and restoring confidence in the naira. The CBN’s figures showed a temporary victory; currency outside banks fell from ₦4.84 billion in December 1983 to ₦3.60 billion by June, and by September, it was back at ₦4.65 billion, recovering most of the fall within three months.

Thirty-nine years later, Buhari tried again, backing another redesign to pull cash into banks.

Instead, the 2022 cash redesign engineered the worst cash shortage in Nigeria’s peacetime history. Within two months, currency in circulation fell by more than two-thirds, upending trade and forcing anyone who could to switch to transfers and digital payments.

For more than two months, Nigerians slept outside ATMs and paid as much as ₦1,000 in fees to withdraw ₦5,000; traders lost sales, farmers watched produce spoil, commuters could not pay fares, and hospitals struggled to buy supplies or treat patients.

Godwin Emefiele, who proposed and ran the redesign policy, is on trial over how it was approved and executed. He has denied all the charges, and Buhari, who approved the policy and defended it through the worst of the cash scarcity, died in July 2025.

In May 2026, Nigerians held ₦5.19 trillion outside banks, representing 91.3 percent of all currency in circulation. Electronic payments have grown, but cash has returned alongside them.

PoS agents, the villains of the cash crunch, have become an even more crucial part of the financial system.

A number called hoarding

To justify a currency redesign in 2022, Emefiele cited a shortage of clean banknotes, advances in counterfeiting technology and the security situation. Redesigning the naira, he said, would reduce the cash available for ransom payments, reinforce the cashless economy and make monetary policy more effective.

Emefiele said ₦2.73 trillion of Nigeria’s ₦3.23 trillion in circulation was outside commercial banks. He called it a “worrisome trend” and evidence of hoarding.

The figure could not distinguish hoarded cash from money held by households and businesses for daily transactions. Currency in circulation had doubled since 2015, but Nigeria’s money supply had nearly tripled. Cash had fallen from 7.8 percent of it to 7 percent.

The counterfeiting argument for a currency redesign rested on the CBN’s currency operations report, which recorded thirteen counterfeit notes per million banknotes examined in 2020, down from twenty in 2019. Their total face value was ₦56.8 million, or one naira in every fifty-seven thousand in circulation.

Eleven years against cash

The CBN piloted a cashless policy in Lagos in January 2012, imposing processing charges on withdrawals above ₦500,000 for individuals and ₦3 million for companies, while encouraging cards, transfers and other electronic payments. The policy aimed to reduce, not eliminate, the amount of physical cash in the economy.

Cash is expensive to print, transport, sort and secure; it also makes corruption and money laundering easier.

Some parts of this campaign worked because they produced useful alternatives. The Nigeria Instant Payment system enabled near-instant bank transfers while the Bank Verification Number made it easier to identify customers across accounts. The CBN’s 2013 agent-banking framework allowed individuals at corner shops to perform basic banking services in places where opening a branch made little economic sense.

Other alternatives required Nigerians to adopt the CBN’s preferred technology without offering a compelling improvement over products they already had. A year after its launch, the eNaira had yet to move beyond what the IMF described as an initial wave of limited adoption because fintech wallets and mobile-money services already performed many of the functions ordinary users needed.

Cash is a stubborn adversary because it works without electricity or mobile data, and settlement is immediate. It is accepted by a bus conductor or a roadside trader without either party needing to trust each other or a third party.

By 2022, the CBN had spent a decade trying to make electronic payments more attractive and cash more expensive. The currency redesign fused that technocratic campaign with the instrument Buhari’s military government had used in 1984; the CBN would withdraw the money and force Nigerians to adapt.

A controlled shortage

The CBN selected the ₦200, ₦500 and ₦1,000 notes for redesign, which together represented 95.6 percent of the value of all currency in circulation. Nigerians were given from 26 October until 31 January to return them, although the redesigned notes were not scheduled to enter circulation until 15 December.

That left forty-seven days in which old and new notes could circulate together.

On 6 December, before the new notes were released, the CBN added withdrawal limits. Individuals would initially be permitted to withdraw only ₦100,000 a week across counters, ATMs and PoS terminals; companies could take ₦500,000. Following protests, the limits were increased to ₦500,000 and ₦5 million.

The limits showed that the CBN did not intend to just exchange the old stock for an equivalent amount of new cash. It also intended to leave less cash in circulation after the swap.

By 29 January, Emefiele said banks had collected ₦1.9 trillion, with about ₦500 billion still outside the system. The Nigerian Economic Summit Group estimated that only ₦390.3 billion in redesigned notes had been made available by mid-December, against approximately ₦2.9 trillion that were required to replace the withdrawn denominations.

The monetary data showed that currency in circulation fell from ₦3.01 trillion in December 2022 to ₦982.1 billion in February 2023.

TOGETHER WITH AFRICA BUILT PODCAST

Episode 1 of the Africa Built podcast does deep into the history of Africa’s first technology unicorn, how it started as two separate companies, Kasuwa and Sabunta. It also shows how it went from launch to IPO in seven quick years.

Listen to Episode 1

Nigeria’s digital infrastructure had grown quickly, but it had grown around ordinary adoption. It had not been built for the forced migration of an entire cash economy within weeks. Bank applications slowed or failed under the traffic, which was unsurprising given that just about five in ten Nigerian adults had a bank account. Many communities had no nearby bank branch, unreliable mobile coverage and limited access to smartphones. A trader who accepted a transfer might still need cash to pay a supplier, load goods into a vehicle or take a bus home.

As the situation worsened, state governments took the federal government to court. On February 8, the court ordered that the old notes should remain legal tender while it considered the case. Eight days later, Buhari addressed the country and restored only the old ₦200 note. The old ₦500 and ₦1,000 notes, he said, would remain invalid.

On 3 March, the Supreme Court ruled that the policy had been carried out without adequate consultation or reasonable notice and ordered that all three old denominations should circulate with the redesigned notes until the end of the year.

The CBN did not formally direct banks to comply until 13 March. By then, traders refused notes they believed other people might reject and banks were unsure whether to dispense cash the president had declared invalid. A court order did not instantly restore public confidence in accepting them.

The World Bank estimated that non-agricultural, non-oil growth in the first quarter of 2023 could have been two percentage points higher without the redesign. Inflation rose from 21.3 percent in December to 21.9 percent in the first quarter. Bank deposits and broad money continued along their previous trajectory, and banking agents became entrenched in the financial system.

The price of money

A PoS agent has an electronic float in a bank or fintech account, and physical cash.

When a customer wants to withdraw, the customer’s account is debited electronically and the agent hands over cash. When a customer wants to deposit, the agent collects cash and sends its electronic equivalent. The agent’s job is to keep both inventories balanced while earning a fee on each transaction.

Before 2023, the fee was largely for proximity because an agent saved a customer the journey to a bank branch or the queue at an ATM. During the cash crunch, customers had to pay a premium for the cash.

Demand for notes rose at the same time that banks stopped supplying them, forcing agents to source cash from petrol stations, supermarkets, traders and any business that still received it from customers. Some agents paid to get cash and passed the cost to their customers.

The Nigerian Economic Summit Group recorded mark-ups of 20 to 30 percent at the height of the shortage.

PoS agents became easy villains because they were the people visibly collecting the fee, and some understandably exploited the shortage. Yet, the underlying cause of scarcity was the CBN, and the banks that had once distributed cash at no direct price could no longer do so.

The accidental migration

Nigerians made 5.2 billion instant-payment transactions in 2022. In 2023, they made 9.7 billion. The value processed through electronic channels rose from ₦387 trillion to about ₦600 trillion.

Some of this growth would have happened without the redesign because Nigeria’s digital-payment volumes had been rising for years. The cash crunch compressed several years of customer acquisition into a few months and forced millions of transactions onto platforms that had previously been optional.

During the crisis, Lagos phone-accessories merchant Oberry Agamah began using Moniepoint terminals because transfers into her regular bank accounts were slow and unreliable. Customers would show that they had been debited and leave with goods before the money reached her.

Payment companies did not create the shortage, but the shortage delivered them user volume and a national demonstration of their usefulness.

The PoS terminals fintechs supply could be used to pay a merchant electronically, but it was also commonly used to withdraw cash.

A customer transfers money through digital rails so that an agent can hand over banknotes. One half of the transaction is cashless, but the purpose of the transaction is to obtain cash.

The naira redesign made electronic payments indispensable without making cash dispensable, and fintechs succeeded by connecting the two forms of money more reliably than banks did.

What stayed

The PoS terminals banks had subsidised so merchants could accept card payments were the machinery of the cash business.

The terminals have become important enough for the CBN to prescribe where they may operate, how they connect to payment switches and how their owners are identified. The regulator that once treated cash outside bank vaults as evidence of a policy failure now regulates millions of businesses whose central service is distributing that cash.

The CBN has also returned to its older instrument. Since January 2026, individuals withdrawing more than ₦500,000 a week pay a 3 percent charge on the excess while companies pay 5 percent above ₦5 million. Forty percent of the charges go to the CBN and 60 percent to the financial institution processing the withdrawal.

Payment systems that gave Nigerians something better than cash grew; instant transfers were faster than cheques and agents were closer than bank branches.

Policies that relied on making cash unavailable worked only while the pressure remained. Cash outside banks fell after the 1984 swap and recovered within months. It fell more dramatically in 2023 and returned as soon as the old notes were allowed back into circulation.

There is still no public accounting of how much illicit money either exercise exposed. There is no figure showing how much counterfeit currency was prevented from returning or how much unexplained wealth was identified because its owner tried to exchange old notes.

Buhari’s two currency swaps demonstrated the reach of the Nigerian state. In both cases, the government could compel citizens to surrender the money in their possession. It could reduce currency outside banks for as long as it restricted the replacement.

It could not make the need for cash disappear.

Send you next week!

Subscribe now

TOGETHER WITH CREDIT DIRECT

That phone. That watch. The gadget you keep removing from your cart.

With Credit Direct Checkout, 25% down is all it takes to go from browsing to owning.

Shop With Checkout

Let’s start a tech publication

Everyone and their daddy believes they should start a tech media publication.

The thinking is that reporters at existing publications don’t have sufficient expertise. Stories miss important context, and most of the analysis is shallow.

So every year or so, people who work in the technology ecosystem toy with the idea of assembling a newsroom of top reporters to produce better work and eventually become the publication.

I have written about the problems with that argument, so I will not rehash them. The more interesting question is what happens after someone assembles that newsroom and has to pay those reporters monthly.

I’ve been thinking about that since reading Techpoint Africa in the last two weeks.

Techpoint was founded in 2015 to explain Nigeria’s nascent technology industry to a wide audience, using office tours, founder profiles and accessible reporting. It has published important stories and built one of the category’s most recognisable brands.

Yet, since at least November 2024, one of its most prolific reporters has been Artificial Intelligence.

PointAI, Techpoint’s AI author, produces news using a mixture of public sources and Techpoint’s data, after which a human reviewer checks the work. At the time of writing, PointAI has 338 articles to its name. That’s almost certainly more stories over the period than any other reporter at the publication.

Techpoint’s masthead lists a five-person editorial team. Chimgozirim Nwokoma, listed as a senior reporter, is now a senior reporter at Condia. Bolu Abiodun, another senior reporter on the masthead, last published on Techpoint in May. Oluwanifemi Kolawole, one of the publication’s longstanding bylines, last published there in December 2024. There are still active human reporters at Techpoint, but it’s likely a tinier team than what an already small masthead suggests.

If the newsroom is small but the story count has not noticeably slipped, then PointAI is worth discussing. AI can write adequate news updates, so that question is settled. But why does a newsroom’s output have to keep the same pace when a smaller newsroom could simply publish less?

In an interview with Communique, Techpoint CEO Muyiwa Mátùlúkò committed to quadrupling Techpoint’s startup coverage over five years. That commitment is a window into how Nigerian tech media works.

An AI byline is an interesting choice for a media publication because credibility is the asset a publication sells or extends to publishers. It builds credibility through the reporters whose names repeatedly appear in bylines and the important stories they produce.

Readers learn which journalists understand a beat and a difficult story carries weight partly because of a reporter’s reputation.

But credibility is expensive, and finding ways to pay for it is work in progress. Big Cabal Media, TechCabal’s parent company, cut 19 per cent of its workforce in 2023.

In a tenth-anniversary account Techpoint published in 2025, a dramatic revenue decline in October 2023 led to salary cuts of up to 70 percent instead of layoffs. The company was still operating at 60 percent capacity well into 2024. In October 2025, its parent company, Businessfront, laid off an undisclosed number of employees, citing long-term sustainability and strategic focus.

The original bargain

Most Nigerian tech publications begin with roughly the same model. Hire a small newsroom, publish every day, build an audience, and then convert the audience into pageviews. They sell those pageviews to advertisers and use the advertising revenue to pay the newsroom, which—stay with me here—produces more stories that attract even more readers.

The model always runs into a set of problems, which are amplified when a publication is niche.

The Nigerian technology audience is valuable but not especially large, so the number of companies with a reason and a budget to advertise to that audience is modest. Programmatic advertising rewards enormous scale, which niche business publications do not have.

Direct advertising pays better, but costs more to sell. You’ll hire salespeople and maintain relationships. Crucially, you’ll need to find a steady supply of companies willing to spend money.

The way out of this problem is expanding the surface area of things a publication can sell.

The publication already knows that website advertising cannot support the newsroom it wants. What it does not know is which adjacent product will. So it launches a newsletter, a podcast, native social video, a research arm, and an events business.

TechCabal’s Moonshot has become an important commercial product. At Semafor, events now produce roughly half of revenue. The company reported $40 million in revenue and $2 million in EBITDA in 2025, its first full year of profitability.

Finding a vertical that works

The problem with expanding the surface area of your publication is that there’s no guarantee you will find something that works quickly.

A podcast needs hosts, production, distribution and a sales proposition before it has an advertiser. Video needs equipment, editors and enough consistent output to learn what audiences want. An event is worse, requiring venue deposits, vendors, programming, sponsor sales and months of staff time before anyone knows whether the room will be filled.

Every new surface for revenue is also a new surface for losses.

Techpoint began hosting events in 2017. ‘Techpoint Inspired’ was followed by ‘Techpoint Build’, town halls, a fintech summit and a blockchain conference. Some became meaningful parts of the brand and attracted sponsorship.

In 2023, the company tried to extend the formula into human resources with Modern Workplace Africa. Mátùlúkò told Communiqué it failed to reach the intended audience and was not profitable. It put its events business on hold the following year.

That experience shows that “events” is not really a business model. A recurring conference with a recognised audience, returning sponsors and an operating team that knows how to deliver it is a business model. The first edition of an event in a new vertical is a bet.

Media is also a herd business. Once one publication proves that a conference, research arm or specialised newsletter can make money, others cluster around the same model. They approach the same sponsors, invite many of the same speakers and compete for the same attendees’ limited time.

What looks like a proven revenue line may still be a bad business for the fifth publication that attempts it.

This herd effect is intensified by how concentrated the pool of buyers is. A publication may have six different products while selling all six to much the same group of banks, telcos, fintechs, venture firms and global technology companies. Diversification on the product side can conceal concentration on the customer side.

Success with an experiment or vertical ironically produces more problems. When one experiment finally succeeds, its profits subsidise the newsroom. It also finances the podcast that has not found an audience, the video unit that has not found a sponsor and the second event that management insists needs one more year.

It’s difficult to know if a failed product is fundamentally bad, mistimed, or poorly executed. Abandoning an experiment is great financial discipline, but it may also mean abandoning the next Moonshot just before it clicks.

Once a company has one profitable product, it can afford to postpone that decision to kill a product. The winner extends the runway of every other loser.

The accounting of newsrooms only compounds the complexity. An event may appear profitable because the reporters who developed its programme, recruited its speakers and attracted its audience are already counted as newsroom costs.

A video unit loses money while making a larger sponsorship package possible. The newsroom acts as marketing for the event, and the event pays for the newsroom. Management is not choosing among a collection of self-contained profit-and-loss statements.

Execution complicates the decision further. Editorial credibility may persuade people to attend the first conference or take a meeting about a research project, but it does not automatically give a publication expertise in venue operations, enterprise sales, sponsor fulfilment, video production or product management. A media company can have the correct strategy and still lose money because it has built five businesses it does not yet know how to operate.

These experiments also draw on the same management attention, sales relationships, and institutional reputation. A poorly executed conference can weaken a sponsor relationship that the newsletter depends on. A podcast without an audience can still consume the editor who should be improving the newsletter. The loss centres do not always sit harmlessly beside the winner; they can make the winner worse.

The challenge, then, is not simply to expand the surface area of things that can be sold. It is to find the money-spinner before the search for it consumes the company and then to decide how many other experiments that money-spinner should be required to carry.

Three jobs, one newsroom

A small editorial team must publish frequently enough to maintain a habit, quickly enough to remain relevant and deeply enough to earn respect. Those are three different jobs.

Daily news has become a commodity that supplies search traffic. But since every publication receives the same press release and watches the same regulator, there is a sameness to daily news.

Analysis is slower. It requires a reporter who has covered a beat long enough to know what is unusual, sources who will explain what is missing from the announcement, and an editor willing to let the article take several days. Investigations are slower and more uncertain. They can consume weeks and end with nothing publishable.

A large publication solves this by having enough reporters to do several things at once. Nigerian tech publications typically ask the same small cast of reporters to perform all these roles.

The daily news obligation wins because tomorrow’s homepage cannot be filled with the investigation that might be ready next month. The newsroom produces the work that keeps the machine alive, even though the work that gives the publication its reputation is the work the machine leaves least time to produce.

Two demand curves

Articles face at least two demand curves: the demand for attention and the demand for payment.

A recent working paper by Gregory Martin, Shoshana Vasserman and Cameron Pfiffer examined four years of activity at a large American metropolitan newspaper: 605 million article visits, 1.2 billion user sessions and more than 55 million paywall encounters.

Soft news produced traffic, but Hard news—local politics, public health and the local economy—produced a greater willingness to pay.

The estimated willingness to pay for the average Local News article was roughly twice that of the average Entertainment article. A newsroom optimised for advertising would allocate staff differently from one optimised for subscriptions.

Yet, the paper found that even the strongest subscription-producing sections could not cover their staffing costs with digital subscription revenue alone.

In Nigeria, the advertising market is thinner, and reader revenue is constrained by incomes, payment friction and the abundance of free alternatives. A niche publication can have a deeply influential audience without having enough people willing and able to pay a recurring subscription that supports a large newsroom.

As we’ve seen with a handful of publications, paywalls can capture more money from loyal readers while shrinking the reach that makes sponsorship attractive. Keeping everything free preserves reach while leaving the publication dependent on advertisers.

That is why arguments about whether Nigerian readers “value journalism” are pointless. Readers can value a product, advertisers can value its audience, and neither will add up to the cost of producing the product at the desired quality and frequency.

Even willingness to pay is not the same thing as willingness to pay enough.

This is why the hunt for a money-spinner is existential and why it often changes what the company sells.

The business behind the publication

Consider an event. On paper, an event is another format for editorial content: interviews happen on a stage rather than in an article. Economically, it is a different product because a sponsor is paying for a room full of potential customers, speaking time for an executive, private access to founders, recruitment leads, product demonstrations and the status of appearing at the centre of an industry conversation.

Research and consulting follow a similar path. A publicly available article gives many readers a general understanding of a market. A commissioned report gives a client an answer to a specific question. The client pays far more because the information is tailored, scarce, and useful to a decision.

Branded content sells production and credibility; executive profiles signal status while awards sell recognition. Each of these surfaces sells a seat near the industry or some version of it. A bank that pays for a keynote slot is not buying an article.

Techpoint’s monthly Pitch Friday meetup began as a community product, but a NAMIP case study says founders who attended later returned as paying sponsors. The event also fed an investor newsletter and the flagship Techpoint Digest. Direct revenue from subscriptions or event tickets was limited; the relationships produced sponsorships and other business opportunities.

TechCabal has built an insights business that offers research and consulting to development institutions, financial institutions, technology companies, startups, governments and investors. Its Moonshot conference sells sponsorship, exhibitions and proximity to the African technology ecosystem. Technext has built conferences around cryptocurrency and financial technology. Across the category, the pattern is that a newsroom creates an audience and a position in the industry; the rest of the company tries to turn that position into something a client can put in a budget.

A good investigation may change how founders behave, help an investor avoid a bad company, give a regulator a clearer picture of an industry, and make thousands of readers better informed. The publication cannot send each beneficiary an invoice but an event can charge a sponsor for a booth.

Journalism creates the asset and access captures the value.

This explains why a profitable Nigerian tech publication often becomes, in economic terms, a market-access company with a newsroom attached.

The website is the public face and underneath it sits a mixture of sponsorship sales, event production, research, data, branded work and corporate relationships. It creates a peculiar internal arrangement where the part of the company with the greatest prestige is not necessarily the part with the highest revenue.

A peculiar tension follows from that arrangement. Editors want reporters to produce work the industry cannot ignore. Commercial teams want products clients will buy this quarter. Management wants the newsroom to publish every day, build the brand, support events, moderate panels, appear in sponsored videos and somehow find time for the investigation that will prove the publication still matters. Even success can feel like losing the argument about what the company was built to do.

The subsidiser has a cycle

I’ve argued before that the industry’s thinness changes editorial relationships. Reporters, founders, investors and communications people know one another, meet in the same rooms and often disagree over whether a story has enough “context.” Context becomes a negotiation over meaning.

The company being covered may be an advertiser, an event sponsor, a research client, a source, and the employer of a future speaker. The publication does not need to promise favourable coverage for the structure to matter. Commercial dependence works through softer channels: which relationship can survive a difficult headline, which sponsor is considered “strategic,” how much senior management time goes into repairing offence, and how quickly an editor learns that a technically defensible story can still create a commercial problem.

In a large market, the loss of one advertiser is painful, but in a thin market, the same loss can remove a meaningful share of an event budget.

It also means technology publications inherit the funding cycle of the industry they cover. When venture money is abundant, startups hire aggressively, launch products, sponsor conferences, buy employer-brand campaigns and pay to be associated with growth. Investors and professional-services firms want access to the same founders. Banks and telcos want to look innovative. The ecosystem’s optimism becomes media revenue.

When funding slows, those budgets are among the easiest to cut. Startups stop buying visibility and start extending runway. Investors need fewer large events to meet founders. Global companies trim local marketing. The publication still has salaries, rent, and the obligation to publish tomorrow morning.

Techpoint’s 2023 experience, a failed vertical and 70 percent salary cuts, shows how these risks converge. Its response was a broader parent company and a new business-media strategy: Finance in Africa, Energy in Africa and Intelpoint alongside Techpoint, modelled in part on the specialist-publication logic of Industry Dive.

A portfolio of publications is only diversified if the customers and revenue cycles are meaningfully different. Otherwise, it is a larger production system resting on adjacent versions of the same corporate budget.

See you next week!

And to OGs, last week’s Notadeepdive was our most-read newsletter ever, but you might have missed it. Check your Promotions or spam folder and, if it landed there, drag it back into Primary so the next one reaches you properly.

As always, if this email was forwarded to you, subscribe for free:

Subscribe now

TOGETHER WITH CREDIT DIRECT

The conference about money just got a sponsor that understands it.

Yield by Credit Direct is headlining Naira Life 2026, Zikoko’s annual gathering for Nigerians who take their financial lives seriously.

August 22nd. See you there!

GET YOUR TICKETS

Everyone wants to know about Jumia’s timeline

If I were into “prediction markets” a.k.a betting apps with extra steps, one sure-fire wager is that every time a journalist asks me for a comment, the company they want answers about is Jumia.

“When will Jumia become profitable?” is a favourite question.

“Can it beat Chinese competitors like Temu and Shein?” is another.

These are useful questions, but they are too focused on what Jumia is doing now when much of the answer is buried in its past.

Francis Dufay took over from co-CEOs Jeremy Hodara and Sacha Poignonnec in November 2022. And like several Jumia leaders before him, he arrived promising greater discipline and a credible path to profitability.

By the end of 2020, Hodara and Poignonnec said Jumia had been put “firmly” on track towards breakeven. Marketing spend had fallen, costs were being cut, and the company’s adjusted EBITDA loss was shrinking.

Then Jumia’s share price climbed to an all-time closing high of $65.51 in February 2021. The company sensibly took advantage of the excitement, raising $341.2 million from investors in March. This came only four months after another share sale had raised $231.4 million. In total, Jumia raised $573 million in new money.

Now flush with cash, sales and advertising spending more than doubled, rising from $37.1 million in 2020 to $81.9 million in 2021. Annual active consumers grew by 18%, from 6.8 million to 8 million. But operating loss widened 41% to $240.9 million.

The spending continued into 2022 with another $75.7 million on sales and advertising, while its adjusted EBITDA loss reached $207 million. Hodara and Poignonnec left that November.

There is a pattern here of Jumia becoming disciplined when its back is against the wall. When money is scarce, it cuts employees, closes businesses, exits markets and discovers efficiency. When money is abundant, it has historically found new ways to spend it.

That is the context in which Dufay’s latest profitability promise should be understood.

For now, Jumia is walking the talk. Its headcount has fallen from 4,318 employees at the end of 2022 to just over 1,980 in March 2026. In the first quarter of 2026, GMV grew by 32% after adjusting for exited markets, physical-goods orders grew by 31% and adjusted EBITDA loss fell by 32% to $10.7 million.

Jumia now says it will reach adjusted EBITDA breakeven and positive cash flow in the fourth quarter of 2026, before delivering full-year profitability and positive cash flow in 2027.

The real test is not simply whether Dufay can enforce discipline while Jumia has limited cash. It is whether the discipline survives success. What happens if the share price rises again and management is given room to become ambitious?

There is also a more fundamental question: has Jumia actually expanded the market for online shopping, or has it mostly bought temporary bursts of activity?

In a November 2014 TechCabal article (one of the many gems I found in my research), Feyi Fawehinmi described African ecommerce as “a tough business that is not for the faint of heart.” At the time, only 360,000 people had bought something from Jumia. Nigeria had roughly 170 million people, internet adoption was growing, and yet the appeal of shopping online remained limited.

Jumia has obviously grown since then, partly powered by an extraordinary amount of investor money. But customer growth peaked in 2021 and has trended downwards even as more Nigerians shop online.

Of course it’s worth mentioning that Jumia has exited countries, closed businesses and changed the mix of activity counted on its platform.

Still, the broader question remains. Historically, whenever Jumia placed discipline ahead of expansion, growth slowed. Dufay must prove that discipline can produce efficient growth rather than simply manage decline.

This is why Jumia’s story needs more than a short article or a quick quote to a journalist. Like the story we published last week about Nigeria’s 2008 stock-market crash, it rewards going backwards, following the money and examining the gap between what people said and what they actually did.

But it also demands a different format. So we made a podcast.

A few weeks ago, I shared the first episode of our Jumia series. About 200 people clicked through and listened. Episode two is now out, and I genuinely think it is worth every minute of your time.

You can listen on YouTube, Spotify, Apple Podcasts or Deezer.

We are going to do much more long-form work over the next year, and your feedback will be essential to what this becomes. It would be fantastic to get this episode to 2,000 listens.

Please listen. And if you enjoy it, give the show a rating, like the episode, or share it with someone who has ever wondered how Jumia managed to spend so much money without becoming profitable.

And if there’s something you hate, your feedback will help make this much better.

It is really entertaining stuff, if I say so myself.

Share

Subscribe now

TOGETHER WITH ZEDCREST WEALTH

The Nigerian equities market has evolved significantly over the past two decades. Today, the market is more transparent, better regulated, and increasingly supported by technology that makes investing more accessible than ever before. The next phase of market growth will be driven by greater retail participation, particularly from a new generation of investors seeking long-term wealth creation. The goal, however, isn’t participation at any cost. It’s informed participation.

Get Free Stocks On Us

A money-making opportunity for a generation

In 2007, *Adewole was bent over the day’s newspaper, tracking the closing prices of several publicly listed companies and doing some maths that made him smile. According to the day’s prices, his portfolio was now worth ₦8 million. Two years earlier, he had taken a cautious slice of his pension, the reward for four decades at a public corporation, and put ₦2 million into the market.

He started investing in the stock market almost by accident. His bank, one of the most prominent in the country, was selling shares, and the bank branch had become part of the sales machine. Customers who came in to deposit or withdraw money would be nudged by tellers, account officers, or whoever else was on the banking floor that day to buy shares.

Adewole did not know what a price-to-earnings ratio was or how to assess the strength of a balance sheet. But he trusted his bank, and everyone agreed that stocks were the next big thing.

That Adewole and millions of other Nigerians trusted their banks was quite an achievement.

In the early 90s, “wonder banks” collapsed, taking depositors’ savings with them, and licensed banks had not behaved much better. In January 1998, the Central Bank revoked the licences of 26 banks, including 12 commercial banks because of their “grave financial conditions.”

In January 2006, eighteen months before Adewole sat smiling over his newspaper, the CBN closed 14 more banks that had failed to raise a ₦25 billion capital requirement. The banks that met the new capital requirements emerged with glass towers, new logos, and full-page newspaper ads inviting Nigerians to own a piece of the financial future. For a retiree like Adewole, the pitch was hard to resist, especially once bank stocks began their bull run.

Seasoned investors were more reluctant. They knew a different market where you bought stocks you recognised, kept the certificates somewhere safe, and waited for dividends. This new market would take some getting used to.

First, the banks needed money

On July 6, 2004, Charles Soludo, one month into his term as Central Bank Governor, told the Bankers’ Committee in Abuja that Nigeria’s 89 banks had 18 months to have at least ₦25 billion of their own money as shareholders’ funds (the cushion that protects depositors’ savings from bad loans). The old requirement was ₦2 billion, and most banks did not even have that.

To survive, 89 banks would need to raise ₦2.2 trillion. The only way out for the laggards who couldn’t dream of raising such sums was a merger.

Soludo believed that Nigeria didn’t have banks big enough for its economy.

The oil boom of the 2000s created billion-dollar energy projects, and Nigerian banks were too small to lead the financing, leaving foreign lenders like BNP Paribas, Citibank, and WestLB to profit from the economic expansion.

Yet, Soludo also understood what happens when you order 89 banks to find ₦25 billion each in 18 months. Some would raise real money, but others would try to cut corners. So he promised a “risk-focused, rule-based regulatory framework” in which arbitrariness would be “reduced to the barest minimum.” The Central Bank would publish the names of qualifying banks on December 31, 2005, verify every naira raised, and punish operators who broke the rules.

The recapitalisation had a second goal of burying the memory of the 90s bank failures. Big banks, the thinking went, were safe banks.

By the December 2005 deadline, only 25 banks were left standing. Getting there required the biggest fundraising Nigeria’s capital market had ever seen.

Euromoney reported ₦406 billion raised domestically in 2005, with more than $650 million from international investors. The Nigerian Stock Exchange was busy with new listings, rights issues, and follow-on offerings through 2005 and 2006, and the brokers underwriting it all got rich.

Many of these capital raises were exactly what they said they were, with real money from institutional and retail investors who wanted to own a piece of bigger, safer banks. Others were something stranger.

Afribank had used depositors’ money to buy 80% of its own public offer, paying ₦25 per share for stock that was trading at ₦11 on the Exchange. The shares later fell below ₦3.

Intercontinental Bank had raised 30% of its share capital with depositor funds. Oceanic Bank’s CEO ended up controlling more than 35% of the bank through Special Purpose Vehicles financed the same way. A significant part of the consolidation had been a sham.

Sanusi Lamido Sanusi, who became Central Bank Governor in 2009, shared the findings in a 2010 convocation lecture at Bayero University in Kano after CBN auditors examined the wreckage.

The verification process Soludo publicly committed to in 2004 had, in Sanusi’s words, been stopped “for some unexplained reason.” Central Bank officials had raised concerns with their own senior management at the time, and the Nigeria Deposit Insurance Corporation (NDIC) had documented its worries and pressed for action. Both were ignored.

By the end of 2007, the market capitalisation of the 25 surviving banks had grown ninefold, and they accounted for 60% of the entire Nigerian Stock Exchange. Banks took almost all the money raised from the market that year, as new issues soared from ₦552.78 billion in 2005 to ₦2.4 trillion in 2007.

It was a rising tide that lifted all boats. Banks and brokers booked interest income, the issuing houses collected their fees, and Soludo could point to all the activity as proof the reforms had worked. By April 2007, Nigeria had 246 registered brokerage firms, up from the small club that existed before consolidation.

The success then traveled overseas when Soludo was named African central banker of the year and the Nigerian Stock Exchange was called the world’s best-performing bourse.

Every time he opened a newspaper, Adewole saw good news. At that point, there was nothing else to see.

Then came the loans

By 2007, people like Adewole who had bought bank shares could take loans to buy even more shares. So you would put up your ₦2 million in shares as collateral, and the bank would lend you another ₦2 million to buy more. When share prices rose, you could sell some of the shares, repay the loan, and keep the difference.

It was hinged on the presumption that prices would keep rising, and for a while they did. Every price increase made the collateral worth more, which justified more lending, which put more money into the same shares, which pushed the price up again. It was roundtripping of epic proportions.

By the end of 2008, the banks were carrying ₦1.2 trillion in share-backed loans, of which roughly ₦400 billion were direct margin loans to individuals and stockbrokers.

The exposures ran through the biggest names in the industry: Intercontinental, GTBank, Ecobank, First Bank, UBA, Diamond, Union, Stanbic IBTC. Some of it was margin lending to retail customers, while the rest was to corporate clients. The banks and their customers would be fine as long as the share prices stayed up.

Olutola Mobolurin of Capital Bancorp described the mechanism perfectly. Stocks were sold to people who were made to believe that prices would never come down. Margin loans were given to buyers who did not know that a falling price would wipe them out.

Adewole was one of them. He held shares he had bought from a bank alongside a margin loan the bank had given him to buy more of those shares. Because the shares themselves were the collateral for the loan, when the price fell, the value of the collateral fell with it, and the bank could call the loan, which would force him to sell the shares to repay it, pushing the price further, and causing the bank to call in even more of the loan.

He could never have seen this coming from just reading the newspapers.

The people behind the scenes

Peter Ukuoritsemofe Ololo was a chartered accountant and licensed stockbroker who studied accounting at Ahmadu Bello University and spent more than a decade at Union Bank and ICON Limited before starting Falcon Securities in 1993. The Wall Street Journal described him as a once high-flying stockbroker for Nigeria’s biggest banks and wealthy individuals, a man who drove himself in a Peugeot 307 while, on some days, his firm accounted for 40% of trading volume on the Nigerian Stock Exchange.

By the boom years, Falcon was the brokerage, and Ololo was the market maker, and being the biggest buyer and borrower gave him leverage nobody else had.

Companies in which Ololo was a director or major shareholder borrowed more than $700 million from five banks, an incredibly large exposure in a fragile system. Separately, the CBN’s debtor list put the exposure incurred through Falcon Securities at ₦88.3 billion, making Ololo the largest individual debtor to three failed banks.

According to EFCC testimony at a Lagos High Court in 2014, a Falcon Securities account was opened with a zero balance to receive a ₦15 billion credit line; another ₦15 billion went to Falcon for the purchase of blue-chip stocks on the floor of the Exchange. The case was still being tried in 2015, and no final public judgment is available.

In a separate matter, Ololo, Adigwe and six others pleaded not guilty to a 36-count charge of conspiracy, stealing and receiving stolen property; that case, too, had reached no verdict by 2015.

According to the Wall Street Journal, when a bank wanted its share price moved, an executive would call Ololo and tell him to buy or sell, and other traders would follow his lead. The EFCC named Ololo in two cases, alleging he had conspired with Sebastian Adigwe of Afribank and Bartholomew Ebong of Union Bank, “by fraudulent means to manipulate the market price” of both banks’ shares.

If Ololo was the broker the banks used, Erastus Akingbola was the bank. He had founded Intercontinental, built it into one of Nigeria’s five biggest, and given it the face of a modern institution with hundreds of thousands of shareholders.

In one EFCC account, Intercontinental issued three cheques totalling ₦10 billion in May 2009 to Tropics Securities and related entities, for shares said to have been bought on the bank’s behalf. Investigators said Akingbola held substantial interests in those same firms. The EFCC alleged he had used ₦179 billion to buy Intercontinental’s shares and prop up its price.

This was a world of dealings Adewole knew nothing about.

In yet another world, there were private placements that allowed you to buy company shares before they listed on the Exchange. The belief was that once the shares were listed, the public would chase them, the prices would rise, and people who got in early would get rich.

Make your first trade

Between 2004 and 2008, roughly ₦650 billion was raised across some 300 private placements, and many of the companies involved never publicly listed.

More than a decade later, Mary Uduk, then the SEC’s acting director-general, told a conference at the University of Lagos that many private companies had exploited gaps in the law during the boom, promising investors public listings they never intended to pursue. She singled out BGL Plc, formerly Banc Garanti, a capital market operator that raised money and never brought its shares to the Exchange.

Other companies gave reasons for not listing.

ARM Properties raised ₦5.675 billion in August 2008, with a listing planned for the fourth quarter. By 2009, its board had postponed the listing, saying the market could not give investors fair value, and the company eventually surfaced on the NASD over-the-counter market as Mixta Real Estate.

Geo-Fluids opened a ₦12 billion placement in May 2008, with shareholders approving a listing the following year, but management found the market too tepid, and the shares did not trade until 2013. Aquitane Oil & Gas sought another ₦15 billion in June 2008, and there’s no account of what became of the offer.

Whether a company never meant to list or didn’t manage to, investors were stuck with shares that could not be sold.

Starcomms’ private placement opened on June 3, 2008, and closed the same day, offering 4.95 billion shares at ₦13 each, with Stanbic IBTC and Chapel Hill Denham as joint issuing houses.

It listed on the stock exchange in July at ₦13.56, fell to ₦7.46 within two months, and ended the year at ₦3.86. Chapel Hill later said that several investors never even read the private placement memorandum. Many bought through brokers and friends who were among the invited investors, without seeing the documents or understanding that Starcomms had been sold as a growth stock expected to lose money in 2008, turn profitable in 2009, and pay dividends in 2010.

The crash

In April 2008, Meristem Securities published a report in BusinessDay: The Nigerian Stock Market Crash: Coming Soon?

That same month, stockbrokers started calling the clients who had borrowed against their shares to let them know the banks wanted their money back. If share prices were still this high, the customers asked, why couldn’t the banks wait? No one had a clear answer. The brokers’ instruction was to repay or sell, usually within days.

Most people had no cash because everything was in the shares, so they were forced to sell. Those sales caused prices to fall, and by the end of April the All-Share Index had declined from 66,121 to 60,339, nine percent in four weeks. Many interpreted this as a market correction.

So when the index edged back to 60,570 in May, it seemed like the worst had passed, so some clients bought more. The bounce lasted three weeks.

June closed at 54,905, and July slid further to 50,422 even as CBN governor Soludo assured the public that with crude at $147 a barrel and reserves at $64 billion, there was nothing to worry about.

Throughout public addresses in July and early August, Soludo claimed Nigerian banks were insulated from the global financial crisis because their fundamentals were great. The market wasn’t buying it, and prices continued to slide.

On August 28, the Federal Government called a meeting in Abuja and invited the Central Bank, the Securities and Exchange Commission, the Nigerian Stock Exchange, and the major brokerages.

The meeting resolved on a range of measures to arrest the stock market slide. The SEC would halve its fees, and the CBN would lower its monetary policy rate, its cash reserve requirement, and its liquidity ratio. Banks would restructure margin loans through 2009, giving them time before recognising bad loans.

Crucially, the Nigerian Stock Exchange changed how much a stock could move in a day. Up, five percent. Down, one percent.

It would take months for a stock price to halve given the one percent rule, and until it did, buyers weren’t interested in paying these artificial prices.

Through September and October 2008, investors who tried to sell shares could not because there were no bids.

In October, the Exchange bowed to common sense and removed the one percent rule. The official explanation was that it didn’t align with international practice, but the backlog of unfilled sell orders had become unmanageable.

With the rule gone, the air went out of the balloon at once. The All-Share Index opened in October near 46,000 and closed at 36,325, a decline of more than twenty percent. By November 28th, the index was at 33,025. By December 16th, 28,085.

Over sixty per cent of the three hundred-odd securities listed on the Exchange were on constant offer. At every price, sellers outnumbered buyers. No matter the price they were looking to exit at, shareholders could not get out.

Adewole was one of them. By the end of 2008, he had stopped opening the newspaper. His shares were worth a fraction of the loan they secured; the bank wanted its money, and there was no cash and no buyer. So the bank took the shares, millions of them, as unsellable in their hands as they had been in his.

In June 2009, President Yar’Adua replaced Soludo with a new central bank governor, Sanusi Lamido Sanusi, who ordered the Central Bank and the Deposit Insurance Corporation to open the banks’ books together.

On August 14 2009, Sanusi announced that five banks: Afribank, Intercontinental, Oceanic, Union Bank, and Finbank, were insolvent. The five banks had ₦2.8 trillion in loans. Almost half of the loans (₦1.14 trillion) were non-performing, with another ₦456 billion in margin loans.

Sanusi sacked all five chief executives. Erastus Akingbola of Intercontinental, Cecilia Ibru of Oceanic, Bartholomew Ebong of Union, Okey Nwosu of Finbank, and Sebastian Adigwe of Afribank.

In October, three more chief executives at Bank PHB, Spring Bank, and Equatorial Trust Bank were sacked. The CBN then injected ₦620 billion into the rescued banks, some two and a half percent of Nigeria’s 2010 GDP.

Eight days before Akingbola was sacked, Aliko Dangote had been elected the seventeenth president of the Council of the Nigerian Stock Exchange.

At the time of his election, Intercontinental Bank alleged in a petition that Dangote’s companies owed it ₦35.8 billion—Dangote denied owing the bank. Either way, a bank listed on the stock exchange was publicly calling the Exchange’s President a debtor. In August 2010, the SEC removed him, along with the Exchange’s director-general.

Of the eight sacked bank chiefs, only Cecilia Ibru was initially convicted in 2010 after she was charged with twenty-five counts of bank and securities fraud. She entered a plea bargain, and the charges were reduced to three. Francis Atuche of Bank PHB was later convicted in 2021, with the conviction upheld by the Court of Appeal in 2022.

On October 8, 2010, Justice Dan Abutu of the Federal High Court in Lagos sentenced her to six months on each count, running concurrently. She served a prison term of six months and forfeited 199 assets, including properties in Maryland (USA), Lagos, and Dubai, shares, and over ₦190 billion in cash. About $7 million was realised from the sale of sixty-one of those properties, and $3.27 million was transferred to the Asset Management Corporation of Nigeria.

Erastus Akingbola fled to London on the day Sanusi sacked him and stayed there for about a year. In 2010, under threat of extradition, he returned to Nigeria and pleaded not guilty. The case was reassigned to a judge named Charles Archibong, who dismissed the charges and barred prosecutors from trying Akingbola again.

The National Judicial Council later compulsorily retired Archibong, primarily for what they termed his mishandling of the case. The Supreme Court would overturn Archibong’s judgment six years later.

Sixteen years after Akingbola was first charged, his case has yet to produce a verdict. At least fifteen judges have handled it, and one of his lawyers has died of old age.

Peter Ololo, the broker whom other brokers called “Market,” was arrested twice in 2009 and released on bail. His lawyer, Femi Falana, told the Wall Street Journal in December that year: “He was simply managing the stocks. When the shares were going up, everyone was happy. Now that the shares are falling, he can’t be blamed for that.” As of 2014, the trial was still being adjourned, and there’s no public record of the courts reaching a verdict.

In May 2019, a reporter from the Punch found a 69-year-old man named Abraham Owoeye selling plumbing materials, roofing sheets, and cement out of a small shop in Lagos. In 2004, he had been talked into buying ₦3 million in shares. By 2009, the shares were worth a fraction of that. Eleven years later, his dividends still sat unclaimed at the registrar.

By December 2018, unclaimed dividends in the Nigerian capital market had soared to ₦129.62 billion.

Some shareholders had died, others had bought shares in different names. Some could not be bothered to fight through registrars, indemnities and paperwork for dividends that now felt too small to justify the trouble.

Many, like Owoeye, had simply moved on from the market.

“I just decided not to think about it because at 69, I have realised that there is more to this life,” Owoeye said.

In March 2026, the All-Share Index of the Nigerian Stock Exchange crossed 200,000 points for the first time in its history. For anyone who had stopped checking in 2008, it was just another number.

Adewole Peter wouldn’t have opened the paper for it. Abraham Owoeye, who was selling plumbing materials in Lagos eleven years after the crash and refusing to read anything with numbers in it, wouldn’t have cared. A chart can go back up, but a person may not.

The market has had two strong years, with pension funds looking at equities again and foreign investors buying Nigerian assets. A new generation of retail investors is investing in a market that is no longer as mysterious as it used to be.

It is a better market than the one that found Adewole at his bank branch and sold him shares he didn’t fully understand.

The Central Bank of Nigeria is not where it was in 2008. In the last year, it has pushed banks to recognise the cost of old regulatory forbearance, restricted dividends for affected banks, and asked lenders to stress-test their books.

It’s not that the CBN has become incapable of failure; after all, regulators always look strongest between crises. Rather, it is more that some of the things that were hidden in 2008 are harder to hide now.

The Securities and Exchange Commission is also less sleepy. In 2026, a handful of stockbrokers were penalised for infractions that included false trading and trades capable of creating artificial prices.

When UBA and Access did not pay final dividends for 2025, investors sold. The share prices moved the way share prices should move when shareholders realise that reported profit and cash in their accounts are not the same thing. This is the healthier version of a market that punishes disappointment and reads regulatory circulars. It asks whether profit can survive provisioning, and notices when a bank cannot turn earnings into dividends.

But memory does not move with the index. The people who were burnt in 2008 weren’t sitting at home waiting for better disclosure standards. They still aren’t.

Their market thesis is simply “never again.” They lost the belief that the market was a fair place to take risks as they watched banks sell shares, lend against those shares, use those shares as collateral, then seize the shares when the prices fell. They watched brokers talk confidently until the phones stopped ringing. They watched private placements promise future listings that never came.

And they watched bank chiefs walk in and out of court years later, while ordinary investors adjusted their lives around the hole left in their savings.

So today, the question is not whether the stock market can create wealth or not. It can. The question is whether the next Adewole enters it differently and with more information.

About Sponsored Notadeepdive editions

Sponsored Notadeepdive essays are commissioned by ambitious companies we carefully choose to work with. They support the kind of long-form business stories that are researched, thoughtful and useful to readers.

The essay is held to the same standard as every other Notadeepdive edition.

For questions or feedback about sponsored editions, email me at olumuyiwa@notadeepdive.com.

Today’s Notadeepdive was brought to you by Zedcrest Wealth.

Alongside access to the market, Zedcrest Wealth invests in investor education through the Zedcrest Wealth Academy, expert market insights, and stock recommendations that help people make decisions based on understanding rather than excitement or FOMO.

Take Your Position is our invitation to actively participate in the market with clarity, confidence, and conviction.

Edited by: Yahaya Hassan, Ikenna ‘BFG’ Sam-Ejehu

Graphic design and direction: Adeyemi Bamtefa

Newspaper screenshots: Archivi.ng

Subscribe now

TOGETHER WITH CREDIT DIRECT

Few things feel better than seeing your returns come in month after month.

That’s the feeling Yield gives you, with one of the best rates in the market today.

Start with Yield today.

Sign Up On Yield

Jumia (Don’t groan!)

If you’ve read this newsletter for any length of time, you’ve watched me obsess over Jumia for years. I’ve talked about their dark stores, their cloud bill, how they’re no longer a doorstep delivery company, nd the magic they'll need to pull off to be profitable by 2028.

What I've never done is tell the Jumia story from the beginning. Like how Jumia started life in 2012 as two separate companies called Kasuwa and Sabunta, or how the company you know as Jumia was legally Africa Internet Group until 2016.

I realised quickly that a story like this doesn't fit in a newsletter. Fourteen years of filings, court documents, and investor presentations, and the world Jumia was formed in, needed audio, not another 3,000 words.

So Nnanna (of Fintech is Easy, and comfortably this newsletter's favourite guest writer) and I made Africa Built, a podcast that reconstructs how Africa's biggest companies were actually built.

Episode 1 is Jumia: The Amazon of Africa, and it's full of details I didn't know until I started digging. Konga's eventual buyer, Leo Stan Ekeh, insists he was building e-commerce in Nigeria before Jumia or Konga existed.

The Jumia–Konga war that invented Nigerian Black Friday was funded on both sides, in part, by the same Swedish money.

▶ [Listen to Episode 1: The Amazon of Africa] on Spotify, YouTube, and Apple.

We look forward to your feedback, shares, and five-star ratings.

Share

And if this newsletter was forwarded to you, subscribe for free here:

Subscribe now

TOGETHER WITH CREDIT DIRECT

Few things feel better than seeing your returns come in month after month.

That’s the feeling Yield gives you, with one of the best rates in the market today.

Start with Yield today.

Sign Up On Yield

Kitchen in search of work

Your favourite restaurant might be two hours away. That is nice to know on an ordinary day and no help at all when you want a hot meal now. So you cook, walk to the buka five minutes from the house, stop at Chicken Republic on the way home, or you open an app and let it bring you the best thing nearby.

Restaurants, delivery apps and cloud kitchens are three answers to the problem of needing good food right away.

A Quick Service Restaurant (QSR) pays for nearness in advance. It takes a site on a busy road, pays rent, puts cooks, freezers, tables and waits for people to walk in and buy good. The rent pays for a kitchen, but it also pays for a place in front of customers. The people walking into the restaurant help cover the staff and equipment before anyone a delivery app brings more customers.

That is the cheapest version of nearness but is still not an easy business. Food Concepts reported revenue of ₦107.5 billion in 2025 and kept about ₦7.2 billion after tax which is 6.6 kobo for every naira of revenue. And that’s with almost 300 outlets, a large procurement operation, and all the walk-in demand its locations can generate.

The delivery app does not need to rent a kitchen or buy ingredients. It aggregates restaurants near you, displays them on one screen, and delivers. The restaurant keeps the food margin, the rider gets paid for the trip, and the app takes a commission and service fee. From its share of the transaction, it pays for refunds, support, and technology. DoorDash completed 903 million orders worth $29.7 billion in the last quarter of 2025 and reported $213 million in profit.

A cloud kitchen’s goal is to escape the constraints of the first two models. It takes the expensive outlet on a busy road out of the equation, along with the dining room and front-of-house staff.

Instead, it rents one central kitchen from which it sells a bevy of food brands. Asian food, Nigerian breakfast options, spicy lunch…you get the picture. It uses the same rent, cold room, prep team, and equipment to serve customers these options.

That is FoodCourt’s model. In 2024, it had about 15 brands in two Lagos kitchens. It said it handled 30,000 orders a month, at an average order value of ₦15,000.

A family could order Chinese food, native dishes and burgers in one basket. A customer who did not want noodles could order jollof. Someone who did not want either could order shawarma. Diverse brands gave FoodCourt more chances to turn roughly the same inputs into more orders.

That model works when there are enough orders to begin with. A kitchen’s costs are roughly the same whether it makes 500 meals a day or 50. An army of brands does not automatically make a kitchen busy.

FoodCourt never published how its 30,000 monthly orders were split between its two locations. If we assume an equal spread, the kitchens were doing about 15,000 orders each month, or roughly 500 a day.

Then it set out to grow with by expanding to Abuja, and a plan for smaller locations reaching deeper Lagos neighbourhoods, with the central kitchens as hubs. Each new kitchen is its own bet on the demand around it.

An Abuja kitchen cannot use a busy evening in Lekki to pay its cooks. A third Lagos kitchen cannot use an order from Gbagada to keep its lights on. Each kitchen needs enough people inside a workable delivery radius, ordering often enough to costs.

FoodCourt would have needed another 15,000 monthly orders just to keep three kitchens working at the same average level as two. It’s unclear whether it found those orders.

The cloud model saves money by concentrating production but a growth plan requires spreading production out.

What we know is that two former employees told TechCabal that growth did not keep pace with the cost of the second Lagos site and Abuja. FoodCourt’s chief executive, Henry Nneji, said the company had believed the Abuja launch would strengthen its long-term position, but that circumstances evolved differently than expected.

The startup couldn’t wait around for diners to walk in so it would have relied on its app. It also listed some brands on Chowdeck for visibility and ran corporate meal services. It made bread, juices and sauces for retailers. It made food for other businesses and even experimented with grocery delivery.

A kitchen that has rent, ovens, cold storage, prep staff and inventory cannot sit idle between lunch and dinner. Corporate meals can fill lunch. Chowdeck can bring customers FoodCourt’s app does not reach. Retail can keep a bakery working. Contract manufacturing can turn spare capacity into someone else’s product.

At a certain point, FoodCourt was a kitchen looking for work.

The cloud kitchen removes the dining room, but it also removes the walk-in customer who helps pay for the kitchen while delivery demand is still growing. FoodCourt had to create that demand itself, then keep finding more of it each time it added another site.

The app gives FoodCourt the full food margin, but it has to acquire the customer. Chowdeck brings visibility, but takes a commission from the business FoodCourt was built to keep. Corporate meals, retail and contract manufacturing can make a kitchen work harder, but they also turn the company into a caterer, retailer and manufacturer alongside its original job as a restaurant and delivery business.

FoodCourt had said it was profitable in 2024 just after it rebuilt parts of its operation. It got larger equipment, pre-prepped ingredients, a push to get meal preparation down to 20 minutes, and made nearly 100 people redundant after it said it had identified theft, wastage and underpricing.

A company can be profitable while raising money to expand so no questions there. A company can also be profitable at one kitchen location and still be unable to finance the next one from its cash pile.

Food Concepts has the scale FoodCourt hoped to achieve with a dense store network, big procurement volumes and customers already walking through the door. Even then, there is not much room in the business for a new location to open badly, for a kitchen to sit underused, or for costs to rise faster than orders.

TechCabal reports that salaries were delayed across FoodCourt’s three branches and Nneji told staff that a funding facility in its final stages had taken longer than expected.

We do not know the size of the facility, who was providing it or whether it was meant to fund expansion, unpaid operating bills or both. FoodCourt is a private company so we may never get answers to those questions.

But taken together, we can piece together from public record a coherent theory for what happened here.

FoodCourt had built more kitchen capacity which it then tried to fill that capacity through direct delivery, aggregators, corporate meals, retail and contract manufacturing. The new locations did not grow quickly enough to cover their added costs, putting a business with thin enough margins on skates.

Share

See you on Sunday!

Subscribe now

TOGETHER WITH CREDIT DIRECT

Spend it or grow it. Most people choose without realising they’re choosing.

Yield by Credit Direct grows your money at up to 21% per annum. The decision is yours. Make it the right one.

Earn up to 21% per annum

A supply-side problem

In America, roughly 400,000 independent restaurants compete for visibility, which makes an aggregator useful. The same oversupply is why the platforms can charge high commissions; with no single restaurant being indispensable, food delivery platforms take 15 to 30% on each order, on restaurant margins of 3 to 5%.

In Nigeria, bukkas and independent kitchens feed most of the country. Only four Quick Service Restaurant franchises have more than thirty outlets each, and below them, a few large single-brand chains.

The restaurants that can supply delivery platforms at scale have the kitchens, footprint, and discipline to produce at volume. What they lack—software, last-mile logistics—they can buy. Yet, for now, the specialisation of restaurants and delivery platforms suits everyone. The restaurants get demand they didn’t generate, and the platforms get users and retention. If the dynamics broke tomorrow, major restaurant chains would lose a sales channel, and the platform may lose its reason to exist.

In changing the balance, a restaurant may decide to go directly to the customer; after all, software has never been cheaper. The hard (and valuable) half is the messy business of logistics. All that to say: Chicken Republic is pushing its own delivery app (again!) while continuing a partnership with Glovo and Chowdeck.

Chicken Republic

Chicken Republic has been one of the food delivery apps’ biggest suppliers for years, Chowdeck named it one of its fastest-growing partners in 2025. For most of that stretch, the channel was close to being invisible on Food Concepts (Chicken Republic’s parent company)books until 2025. The trade receivables the company booked from its online partners over the year, i.e, Glovo and Chowdeck, grew from under ₦800 million to ₦12.6 billion.

In August 2024, Food Concepts signed an exclusive delivery deal with Chowdeck in Lagos and Ibadan, pulling the chain off rivals like Glovo and HeyFood in those cities. This came a year after Chowdeck onboarded the chain and saw a surge in its order volume. By one analyst’s estimate, Chicken Republic may account for thirty percent of Nigeria’s food-delivery order volume, with multi-brand restaurant chains completing the pie.

It may be one reason Food Concepts launched Republic Dash, a first-party ordering app carrying Chicken Republic, PieXpress, and Chopbox, the three brands under its roof in 2025. (Was the name inspired by DoorDash?)

Republic Dash is in select cities across Nigeria and Ghana, and it has crossed ten thousand installs on the Play Store. It has a companion driver app, though it is still unclear if Chicken Republic hires riders or uses another company’s fleet.

In its 2025 report, Food Concepts describes Republic Dash as a way of “building direct relationships with our most valuable customers.” It filed the app under omnichannel alongside its current delivery partners.

Chicken Republic began taking website orders as early as 2015 and ran phone-and-web experiments through the pandemic. All of it amounted to nothing, so it is easy to file Republic Dash alongside those forgotten experiments.

But the earlier attempts failed due to timing. The market education that makes a direct channel viable and the slow work of educating customers did not exist until recently. The platforms that invested in customer education are now gone: OFood, Jumia Food, Bolt Food.

Also, building a proprietary stack historically meant expensive custom software and idle courier fleets. But cheap engineering is now abundant, last-mile logistics has become something an asset-heavy chain can buy, and the sales numbers suggest there’s something to fight for.

And Food Concepts is not the company it was years ago, and neither is the food delivery market it operates in a semblance of its former self. The restaurant chains have taken a long, unpaid education in the economics of selling food online, courtesy of the very apps it fed. What it brings now is a central kitchen, distribution fleet, centralised procurement, warehousing and dedicated production units. It is the infrastructure that makes the chain indispensable to the aggregators, and the same infrastructure that app-native food delivery apps with their in-house kitchens, like FoodCourt, have been trying to assemble.

The Illusion of Dominance?

The truly scarce input in Nigerian food delivery is standardized, branded, delivery-ready food. Very few kitchens can produce a consistent product at scale under a trusted brand name.

When supply is this thin, durable value accrues to the producer, not the courier. Aggregators look dominant because they currently control demand generation and last-mile fulfilment. To fill out their catalogues, they have onboarded informal vendors diligently. But the chains drive the volume. When Jumia Food launched as HelloFood in 2012, and when later aggregators entered, their growth engines were the same institutional names. Those brands were part of why users downloaded the apps in the first place.

The platforms have rarely turned that volume into success. Jumia Food lost money for a decade before shutting down its continental footprint in late 2023. Bolt Food had exited Nigeria weeks earlier, reportedly having lost primary quick-service franchises to competitors. The platforms absorb the heavy capital losses of the logistics layer, while the franchises keep the customer and outlive the tech companies built to serve them.

Part of what makes the chains defensible is economies of scale. Buying ingredients by the metric ton lets the chains absorb currency shocks and pass some of the savings to the consumer. A nascent food delivery operator juggling delivery and its own kitchen operations buys raw inputs closer to retail prices.

The squeeze runs down into the courier network. Aggregators are caught in a multi-sided bind: they cannot raise delivery fees on a price-sensitive consumer, and they cannot cut rider payouts without losing riders to competitors.

Can the Chicken go the last mile?

Yet the fast-food business is also a struggle. Even Nigeria’s best-run QSR groups keep less than seven naira of every hundred they generate. Food Concepts recorded a net profit margin of 6.7% in 2025, up from 3.5% the previous year. This thinness leaves little to fund an expensive tech operation.

Cloud kitchens, food delivery apps that produce their own food, across the world have demonstrated that cooking and packing food, maintaining logistics operations, generating demand and managing customer-relationship software are distinct and capital-intensive businesses.

Being cost-efficient and successful at one says nothing about the other three.

Chicken Republic’s decade of failed logistics experiments may also suggest a structural inability to manage delivery. Republic Dash has 10,000 downloads; Chowdeck claims over 1.5 million users and says it is profitable.

There is little evidence that a legacy fast-food operator has the institutional muscle to pull off this adventure. If Republic Dash tries to compete as a discovery engine for its own brands, it may struggle to outspend platforms for the digital attention it used to receive for a small commission.

But there is a narrower, more plausible version of the bet. City-wide rider availability is essential for a marketplace connecting thousands of restaurants to millions of homes. But a single chain delivering its own food from local outlets to repeat customers within a fixed radius faces a simpler operational problem.

Aggregator Defenses

The platforms are not waiting to see how this plays out. In the past two years, they have been spending to get ahead of suppliers who may route around their marketplaces.

Much of the investment is going into groceries, where supply is abundant and substitutable. Grocery delivery is a category nobody loves on margin, about 1-3%, and everybody wants on frequency. Glovo is running micro-fulfilment dark stores in select locations. Delivery Hero, its parent company, has found that customers who order food and groceries spend more than five times what food-only customers do. Chowdeck is reportedly opening two dark stores weekly, aiming for five hundred by the end of 2026, stocked through an exclusive partnership with the bulk e-commerce platform GoLemon, which handles sourcing and quality control while Chowdeck runs fulfillment.

Another sharp move is to get inside the merchant so that leaving the platform costs something. Last June, Chowdeck bought Mira, a point-of-sale system for restaurants. This backend integration has the potential to create an operational dependency that discourages vendors from switching from the platform, and even if they do, convert them to customers of a different product within its ecosystem.

Glovo is chasing the same supplier lock-in with loans. Through a 2025 partnership with the fintech Salad Africa, Glovo extends working-capital loans to its restaurants, underwritten by their real-time sales on the app and repaid as a slice of future orders processed through Glovo. A restaurant carrying the platform’s working capital may find leaving complicated.

Chowdeck built Relay, an on-demand, white-label logistics utility, reachable by API or dashboard and paid strictly per drop. If a chain like Chicken Republic migrates customers to a first-party app to escape marketplace commissions, Food delivery apps can still earn on the volume by serving as the underlying courier, without carrying the cost of acquiring those customers. By selling last-mile to the vendors expanding beyond its app, the biggest supplier can become a customer of a different product.

Both major platforms are building defenses around the concentration risk of depending on a few corporate kitchens. Read together, Chicken Republic’s app and the aggregators’ countermoves are the same strategy executed from opposite ends of the value chain. The supplier is integrating forward into delivery; the aggregator is integrating backward into supply.

If platforms keep their margins by owning discovery and the grocery run, while the chains find that going direct costs more per order than the commission they already pay, concentration is a feature the market can live with, and Republic Dash will remain a marketing curiosity.

See you on Sunday!

Subscribe now

TOGETHER WITH CREDIT DIRECT

Get the most powerful vivo phone with just a 20% down payment. Use Credit Direct Checkout, the smarter way to shop.

Get Started

The proximity business of food delivery

In Lagos, food is the cheap part (relative to inflation!); there is white rice or jollof rice on every street. Shawarma, mediocre spaghetti, and turkey are usually two stalls away on busy streets. The economy may have thinned out bukkas because rent is insane, but a hungry person in Lagos is never really far from a meal.

So when you use a food delivery app, what you’re paying a premium for is the nearest good meal the moment you decide to eat, brought straight to your door.

Inevitably, everything in food delivery is a contest over who supplies nearness to the customer, and in the fine print are questions like, “At what cost?” or “Who paid to create the demand in the first place?”

Restaurants, delivery apps, and cloud kitchens are three answers to the same question.

What prompted these thoughts around food delivery is FoodCourt. Since March, FoodCourt has been unavailable on Lagos Island. When customers asked whether it had shut down, the company said its Lekki hub was briefly offline for “operational adjustments.”

That adjustment appears to have now spread to the mainland. On the FoodCourt app, pretty much all items are unavailable, and on Chowdeck, where it was listed for ages, it’s no longer there.

This is not the YC-backed startup’s first wobble. Foodcourt undertook an operational overhaul and cut nearly 100 staff in 2024, and if you think about it, the startup was built to win the market in a specific way.

How do you win the battle for nearness to the customer?

The restaurant has the cheapest answer. A Chicken Republic on a busy road sells meals for rent it will pay anyway. The staff works in shifts, and people walk in for breakfast, lunch, and dinner. The physical space is for cooking and dining, but it also advertises the space, generating demand from existing foot traffic. Customers don’t have to be acquired in that sense.

Food Concepts, the company behind Chicken Republic, PieXpress, and Chop Box, makes most of its sales from walk-ins. In 2025, its restaurants brought in ₦106.8 billion of the group’s ₦107.5 billion in revenue.

It made ₦7.2 billion of profit after tax which meant it kept seven kobo kept for every naira earned. And that’s the entity with the “cheapest” nearness to the customer. In 2023 the company kept ₦377 million on ₦66 billion — six-tenths of one kobo — after a falling naira and a ₦5.5 billion diesel bill.

The delivery app is the expensive answer to proximity to customers. It doesn’t have kitchens, so to offer proximity, it invests in infrastructure like logistics, inventory management, technology, etc. After the restaurant takes the food margin and the rider takes the trip, what reaches the platform is a sliver, and out of that sliver come refunds, promotions, and technology costs.

How thin is that sliver? DoorDash, with tens of millions of repeat users and a thriving subscription business, completed 903 million orders in Q4 2025 worth $29.7 billion and eked out a profit of $213 million.

Measured against the food that moved through the platform, that’s seven-tenths of a cent on the dollar, after a decade of operations and real scale. In Nigeria, Jumia, which had every reason to make it work, shut Jumia Food across seven markets in 2023 and said it had never been profitable. Bolt Food left the same year. Engineering proximity to the customer is a worse business than owning it.

So here’s an idea. Why not be both and double your odds? Own a dozen virtual brands (FoodCourt’s Chinese menu is good), cook your food in centralized kitchens, sell directly to customers, and skip paying commission to food delivery companies.

To be both, the cloud kitchen has to remove the dining space, because that’s the expensive part. But the space doubled as a billboard of sorts. If you take it away, the rent for premium space disappears, but in its place, you’ll pay to acquire every customer who would otherwise have walked in. Every order will now begin with someone installing an app.

Realistically, alongside the app, you’ll likely need to list on a bigger platform like Chowdeck anyway and pay commission. You might even have to pay some advertising fees for premium placement and then pay a commission anyway.

Cooking a dozen menus to order from one space is slow, and orders will run late (in my experience, you should eat two small noodles before you order). It forced FoodCourt to rebuild the kitchen in 2024 and start cooking base ingredients ahead of demand. It also cut staff. The shared kitchen that saved on rent was not immediately as efficient as thought.

This is not a Nigerian failure of nerve. FoodCourt’s closest analogue is Rebel Foods in India — the same model, multiple virtual brands from shared kitchens ordered through its EatSure app, scaled past 45 brands across 80 cities. In the year to March 2025, Rebel made about $191 million in operating revenue and lost nearly $40 million, spending $1.23 for every dollar it earned. It now runs 19 offline stores and pitches franchisees a 1,500-square-foot outlet with a dine-in Oven Story pizza store out front and cloud kitchens behind it. Elsewhere, the same retreat: Kitchen United fell back to selling software, and Kitopi rebranded as a hospitality company with more than 200 physical outlets across the Gulf. Nobody stayed with the pure cloud kitchen for long.

This is not a uniquely Nigerian problem. FoodCourt’s comparable is Rebel Foods in India. They have the same model of multiple virtual brands from shared kitchens ordered through its EatSure app, scaled past 45 brands across 80 cities.

In the year to March 2025, Rebel made about $191 million in operating revenue and lost nearly $40 million, spending $1.23 for every dollar it earned. It now runs 19 offline stores and pitches franchisees a 1,500-square-foot outlet with a dine-in Oven Story pizza store out front and cloud kitchens behind it.

Elsewhere, the same retreat: Kitchen United pivoted to selling software, and Kitopi rebranded as a hospitality company with more than 200 physical outlets across the Gulf. Nobody stayed with the pure cloud kitchen for long.

Let’s turn back to Chicken Republic, which offers the cheap version of proximity at 190 locations, where delivery is a layer on a kitchen a walk-in crowd already pays for. Food Concepts’ receivables note shows ₦12.632 billion of additions and ₦12.228 billion of collections during 2025. The company says those movements included e-commerce sales through Glovo and Chowdeck.

Its size means it gets to choose. It made Glovo its key online platform in 2023, launched its own app, Republic Dash, in 2025, and still has both aggregators on speed dial. An order through Republic Dash saves the commission it would pay Glovo or Chowdeck and keeps the customer. The aggregators bring orders it would not otherwise see. A brand with 190 kitchens and standing demand plays those options against each other. FoodCourt, with one shared kitchen and no walk-ins, has nothing to play.

Step back, and the venture-funded delivery boom is an expensive way of teaching Nigerians to want a hot meal brought to the door, and whoever owns the most kitchens in the most places, paid for by the most walk-in diners, will collect on the habit once it forms. The disruptors did the R&D, but the incumbent still owns the demand.

Cloud kitchens raise money to engineer the one thing a Chicken Republic outlet gets thrown in for free once it’s on a busy road. It might work, but so far, the assumptions aren’t as neat as first expected.

Share

Subscribe now

TOGETHER WITH CREDIT DIRECT

If it’s growth for you and your loved ones, choose Yield.

There’s a perfect plan for everyone and every situation.

Sign Up On Yield

Nollywood’s abundance problem

In March 2025, a smooth-talking, free-spending businessman hires a dance troupe to get a woman’s attention and, for the rest of the movie’s runtime, spends the budget of a small Nigerian state on her until they live happily ever after.

The film was Love in Every Word, Omoni Oboli’s romantic drama about the enduring Nollywood belief that there’s no romance without finance. Oboli bypassed the cinema (the destination of choice for Nollywood’s royalty) and the streamers, taking the film straight to YouTube.

Within twenty-four hours, it had crossed a million views. It ended the year with more than thirty million views and a top-five place in Google’s Year in Search, while “Achalugo” became the most-searched word in Nigeria.

Tens of millions will watch a Nigerian film the week it drops, so we know there’s no demand problem. What they will not do is pay a subscription to watch it. So, who pays for these films?

At the end of the Love in Every Word sequel, the credits list UBA, Coca-Cola, Vaseline, Close-Up, GIG Logistics, and MTN. Those were Oboli’s customers. The thirty million viewers were the inventory she sold to her customers.

The popularity of Nollywood films on YouTube is prompting questions on the future of paid video streaming in Africa. In a country where people love films, why can’t paid streamers find a critical mass to pay subscription fees?

The accepted answer is that Nigerians are too poor. It is the one Jason Njoku reached for after a decade of IrokoTV and a hundred million dollars, and it is partly true. He has long doubted whether Nigeria has 250,000 people who can consistently afford a five-dollar-a-month product.

But poverty cannot explain why these same viewers are buying data and watching two-hour-long shows by the millions. What the public is rejecting are subscriptions, not Nollywood.

Consider that the audience is willing to pay for live football. If Arsenal play Liverpool at 5:30, there is no substitute. The match happens once, and you either watch it or miss it.

Nigerian movies aren’t like that. While Hollywood will spend millions of dollars on crafting specific titles and building franchises, the predominant Nollywood method is volume. Nigeria produces well over 1,000 films annually, with skitmakers and YouTube film producers adding to this abundance. Only a handful of filmmakers have the time and resources to build recognisable title franchises.

A subscription business assumes that a library creates power, but a library’s power comes when the audience believes the library contains something scarce. If what you have is a large collection of highly substitutable stories, your catalogue can become nice to have, but not worth throwing scarce disposable income at every month.

This is also why piracy is not merely a legal problem in Nollywood. Piracy shows that the audience wants the content, but not badly enough to accept the platform’s preferred terms. It tells you that the demand is elastic. The viewer is not saying, “I do not want this.” He is saying, “I do not want it like that.”

Does this abundance problem predate YouTube?

One interesting roundabout answer is that IrokoTV began on YouTube. Before the subscription app, Jason Njoku and Bastian Gotter launched “Nollywood Love” on YouTube. Njoku would later fly to Lagos, buy online rights from producers, and strike a deal with YouTube, to put Nollywood films online for free. These full films, which were fully licensed, were free to watch in 2011. This shows the YouTube model is not new to Nollywood.

Iroko’s YouTube network generated hundreds of millions of views, proving that the audience was there. It was time to turn all that watching into enough money to build the kind of company investors had funded.

It built an app and charged subscriptions. The pitch would have been that this was a home for Nigerian films, organised for an audience that had previously been scattered across DVD sellers, pirate sites, YouTube channels and memory cards.

However, Nollywood’s natural state has always been overflow. DVDs, VCDs, Alaba marketers, Africa Magic reruns, roadside sellers, pirated discs, hard-drive transfers, Bluetooth, memory cards, illegal uploads and that one neighbour whose laptop is somehow always full of films.

The domestic Nigerian viewer had to choose between IrokoTV and a messy, informal distribution system that had already taught him that Nollywood was replaceable and cheap.

The diaspora audience was different. A Nigerian in London in 2011 had a scarcity problem, and if she wanted Nollywood, she needed someone travelling from Nigeria with discs, a pirate website, or a service that organised the chaos. Iroko solved that problem well.

A Nigerian in Lagos did not lack Nollywood. He lacked convenience, quality, payment simplicity and sometimes legality. Those are real problems, but they are not the same as scarcity. You can charge more easily for scarcity than for tidiness.

Nollywood was already producing at industrial speed when Iroko arrived. The stories were already formulaic, and substitution was already high.

In 2011, free video could work for a special aggregator with a licensed back-catalogue, a YouTube deal and a global audience. It could not yet work for everyone because the ad market was too thin and brand integration had not matured. Mobile video had not become an everyday Nigerian habit, so a producer could not reliably say: I will give this two-hour film away, collect tens of millions of views, and make the money back from brands.

By 2025, data got cheaper, and smartphones became more accessible. The audience grew comfortable watching long Nigerian videos on phones, and brands learned that creator-led video was not a novelty. Then, skitmakers also joined the party.

Mark Angel led the way, opening the gates to a broader wave of creators like Broda Shaggi, Taaooma, Mr Macaroni, Sabinus, Kiekie, Sydney Talker, among others. The model was taking shape by giving the audience the video for free and selling the audience to advertisers.

The skitmakers showed that production could be funded without a subscription. They showed that brands would pay to sit inside culture and that an audience that refused to subscribe could still be valuable.

Omoni Oboli and other filmmakers are basically running the same model for feature-length films. Give the audience the film for free and let social media create virality with memes and short clips. The last leg is allowing brands buy proximity to the attention.

Nigeria’s music industry confronted and arguably solved a similar problem years ago. Used to downloading music from bootleg websites, Nigerians will not pay for recorded music any more than they will pay for film.

Music streaming services also struggle to find paying subscribers, so the freemium model of apps like Audiomack, Boomplay, and YouTube, which is supported by ads, has become the solution.

Artists make their money from shows, endorsements and image rights, but streaming income for musicians with mostly Nigerian listeners is notoriously paltry.

There have been some useful but brief periods of profitability. In 2008, MTN launched caller tunes for fifty naira per song, billed from a customer’s airtime, and it became the largest distributor of Nigerian music in the country.

The recording was everywhere and was worth nothing, so the money came from selling the audience a novelty through a telco. When Callertunes lost its allure, music found more ways out of the trap than film has, because it kept things that a copy cannot reproduce.

A Nollywood plot is replaceable; rich-man-poor-girl arrives in twenty versions before dinner. The star is scarce by identity, the concert is scarce the way the Arsenal match is, and the fan who streams from London pays in pounds.

Music sells three scarce things. Film, so far, sells one: a brand’s proximity to attention. That is why a touring musician out-earns a filmmaker with thirty million views while both give the recording away. Nollywood has solved distribution but it has not yet found its stage.

See you on Sunday!

Share

Subscribe now

TOGETHER WITH CREDIT DIRECT

These are the most crucial numbers from Nigeria’s Credit Landscape Report.

On their own, the numbers say a lot, but the report explains why they matter and how they’re shaping the future of credit in Nigeria.

Read The Report

Understanding the Latest CBN Memos on Holdco Structure

On June 10, 2026, the Central Bank of Nigeria published consequential policy directives as part of its drive to sanitise the financial services sector. One policy updates the rulebook for financial holding companies while the other is called “ring-fencing operations of closely linked entities,” which sounds like nothing.

Yet, these are the biggest shake-ups the CBN has handed the industry since the recapitalisation directive, and most of the people affected haven’t read them yet.

Here is what it means.

For about fifteen years, Nigerian finance has used a quiet trick. You don’t build one company; you build one brand and collect licences under it. The bank sits in one company, the payments business sits in another, and the lender sits in a third, sometimes licensed by a different regulator entirely.

Then you stitch them together with a single app, and most customers have no idea (or have no reason to care) that the wallet, the loan and the transfer they just made were legally facilitated by three different businesses.

OPay’s app takes your deposits through one licence and gives you loans through a microfinance bank you’ve never heard of. Moniepoint’s business app sits on top of a microfinance bank and a separate payments company. Sycamore, a bootstrapped Lagos lender, runs a moneylending business, an asset manager and a microfinance bank through one app and is structured as a group. Paystack also initially tried to do it with Zap before having to retreat and comply.

The big banks played the same game at a grander scale: Access Bank owns banks in more than twenty countries, all hanging off the Nigerian bank’s balance sheet. And VFD built a listed company on the premise that if you call yourself an investment company rather than a holding company, you can own microfinance banks, a mortgage bank, an insurance broker, hotels and restaurants in the same portfolio and answer to nobody in particular.

The CBN has looked at all of this and decided that a lot of regulatory arbitrage is at play. Companies are doing things their licences don’t allow, through sister companies whose licences do, while the customer thinks he’s dealing with one business.

So the new rules attack the trick at its root, which is the connection between the companies. By inventing a category called “closely linked entities,” the draft is deliberately ambiguous in its definition.

You’re closely linked if one company controls the other, obviously. But also if you share directors, systems or branding. You’re also closely linked if one of you depends on the other by contract. With that definition, almost every financial group in Nigeria is one organism, no matter how many CAC registrations it’s spread across.

And once you’re one Holding company, three things follow.

Your app can no longer do everything. Each company’s app may only offer what that company’s licence permits. If your customer wants a loan from your sister company, the sister company must onboard them, which means doing its own KYC, wallet, getting explicit consent, and a disclosure telling customers that they’re free to get the loan elsewhere. The loan money lands in the sister company’s wallet, not yours. The seamless super-app experience where customers tap for savings, credit, and insurance, all inside one interface, goes from being a delightful innovative product decision to a compliance violation.

Customer funds can’t be lent across the group, can’t secure the group’s borrowing, and can’t pay a sister company’s salaries. Customer data also can’t be shared between linked companies without explicit consent. The CBN can(in theory) even order you to run separate data centres.

This unfortunately eliminates the whole point of an ecosystem as the data flywheel, cross-subsidy, and cheap internal funding become unplugged piece by piece.

On the financial side, any loan a banking subsidiary sends up to its Holdco gets treated as a return of capital and struck off the bank’s capital base when its capital adequacy ratio is computed. Loans to a sister company carry a 100% risk weight if secured, a full capital deduction if not.

Every naira a bank lends to its own group costs the bank a naira of regulatory capital. The cheap internal funding that made the ecosystem worth building is now the most expensive money in the group, and the Holdco is left to run on dividends, treasury-bill income, and whatever it can raise from outside. Expect to see more public responses on this from listed financial services entities, as this changes their reporting guidelines.

Every set of closely linked entities must put a non-operating holding company on top: a company that does nothing except own the others, licensed by the CBN, holding at least 51% of each subsidiary, and capitalised at the sum of all its subsidiaries’ minimum capital plus 20% — in real paid-in money, not reserves. If you’d rather not build that, the alternative is to merge everything into one company, hand back every licence except one, and stop doing whatever the surrendered licences allowed.

How does this affect today’s Players?

The groups that already became holding companies like GTCO, Stanbic IBTC, FCMB, FirstHoldco, Access Holdings, and Sterling Holdings have to redraw their org charts again.

Their Nigerian subsidiaries must hang directly off the Holdco, and foreign subsidiaries must sit under a separate intermediate holding company. Access Bank’s twenty-something foreign banks currently sit under the Nigerian bank itself. Moving them means change-of-control approval from every regulator from Nairobi to London, with tax and capital consequences. Access was already under shareholder pressure to sell its weaker foreign units; this turns that debate into simple arithmetic, because every subsidiary you keep is a subsidiary you must lift out of the bank and re-approve in its host country. The cuts and divestments will be brutal and devoid of sentiment.

Zenith just received the CBN approval for its holding company, barely a month before the rules for being a holding company changed.

The rules also reach down to the board table, and that is where even the most compliant Holdco of them all gets a reorganisation. A Holdco director may now sit on only one subsidiary board; Holdco directors together can’t hold more than a fifth of any subsidiary’s seats; Holdco staff can’t be non-executive directors anywhere in the group; and the old habit of subsidiary directors sitting in on the parent’s board meetings is banned outright.

Take GTCO. Segun Agbaje, the group chief executive, sits on the boards of GTBank Ghana, GTBank UK, GTBank Kenya and GTBank Tanzania and chairs the UK bank. Under the draft, that’s three boards too many.

Every group of any size typically runs on the same overlapping cast of executives, which is precisely how the Holdcos run its subsidiaries with the same uniform goal, and cross-pollination of ideas. Once that synergy is cut, these groups suddenly need a far deeper bench of fit-and-proper independent directors than Nigeria has lying around.

UBA, despite not being a Holdo, is affected. As a big bank with multiple subsidiaries, the draft mandates that closely linked entities form a non-operating Holdco. And if the group contains a commercial bank, the full holding company regime applies. UBA, its African banks, and pension custodian are a closely linked group by every part of the definition. This means that the same Holdco conversion Tony Elumelu has spent fifteen years avoiding for the UBA group now needs to be had.

There is also the monetary consideration of all these changes. By asking for a 20% buffer on top of every subsidiary’s minimum capital requirement, the CBN is asking some of these non-compliant entities to return to their shareholders for fresh capital to restructure equity distributions across companies.

Aspiring Conglomerates are also affected here as the Holdco rules ban holding companies from owning anything outside financial services. This is defined narrowly as businesses regulated by the CBN, SEC, NAICOM or PENCOM.

Sterling has spent years telling anyone who’d listen that its future runs through health, education, agriculture, energy and transport. As a lender to those sectors, nothing changes; banks can lend to anyone. But the version of the story where the holding company “allegedly” has ownership stakes in clinics, EVs, Energy grids and farms it finances is now illegal by design.

A financial holding company, the CBN is saying, is a filing cabinet for financial licences and nothing more. Zenith’s newly announced “beyond banking” ambitions have also hit the same wall.

VFD ran the opposite play of this by becoming a conglomerate that engineered its way around ever becoming a financial holding company. My unconfirmed theory is that they and Sterling were the ones this draft had in mind at inception, as their “proprietary investment company” label worked because microfinance and mortgage banks did not technically count as banks under the old Holdco rules.

The new rule is no longer about what you own, but about how connected these entities are. VFD’s forty-plus subsidiaries share the same owners, directors and a brand name. The regulated half of that portfolio must now go into a licensed Holdco that cannot also own the real estate portfolio.

This means the likely ending is a split: a clean financial group on one side, everything else on the other and even then, if both halves keep calling themselves VFD, the shared-branding clause may follow them across the fence. (Luckily, the founder honed his craft at the feet of Tony Elumelu, so a Heirs Holdings - UBA group type split should be on the horizon for them)

For fintechs, however, this becomes a template for how the industry should be built in the future. Paystack, for example, seems to have already built the template the CBN wants with their Stack Group restructuring with a switching licence on one side, Zap(powered by Fidelity Bank) and the Brass + Ladder microfinance banks on the other.

However, even this has a potential landmine on the horizon, as the fine print mandates the holding companies to be Nigerian entities. Stack’s ultimate parent is Stripe. Moniepoint is a foreign Inc. Nearly every venture-backed fintech in the country runs its Nigerian licences out of a Delaware, UK or Mauritius parent, because that’s what foreign investors require. According to the draft, this structure is already non-compliant.

Finally, these rules also masquerade as death sentences against ambition. An example is the shared-services section. It says a Holdco may centralise office space, legal and IT for its subsidiaries. Risk management, compliance, internal audit, the company secretariat must now live inside each subsidiary, independently staffed and independently funded. That means the Group Chief Compliance Officer function that covers three companies from the Holdco floor is no longer allowed to exist.

That means a smaller holding company must either build a holding company capitalised above the combined minimums of its individual licensed businesses, with multiple boards, multiple compliance functions, recovery plans, twice-yearly continuity drills and an annual independent audit of every shared service, or merge down to one licence and divest the other business lines.

This does not matter if the group is profitable and bootstrapped, or if these entities have little chance of survival as standalone side projects. There are hundreds of two- and three-licence companies in Lagos doing the same maths this week. Most of them will merge, sell, or hand licences back. The Cardoso-led CBN has been adamant about the sanctity and capital adequacy of the industry, and these new rules will produce exactly that.

So who wins?

The lawyers and consultants who bill the restructuring, the senior executives drafted in to fill all those new independent boards and standalone compliance functions, and the few groups already built the way the CBN wants.

The CBN’s document admits the 2014 framework caused a surge in overheads, and its fix is more boards, more audits, more duplicated infrastructure across more companies. Everybody in this market knows compliance is the moat nobody wanted to pay for. The CBN has now named the price, and it is paid in structure — more boards, more audits, more standalone entities — with no opting out.

These are still exposure drafts, and affected institutions have until July 9 to engage the CBN on this.

*The Opinions shared here are the independent analysis of the author and do not represent the views of any organisation they may be affiliated with. Nothing in this opinion piece should be regarded as Tax, Legal, Financial, or Regulatory advice.

Please do your own research and engage licensed practitioners wherever necessary.

Subscribe now

TOGETHER WITH CREDIT DIRECT

Fewer loans. Better repayment. Higher amounts. Women in Nigeria’s credit market are quietly outperforming the averages, and the full picture is even more interesting.

Nigeria’s Credit Landscape Report breaks down who borrows, why, and what it means for the future of credit in this country.

Download The Report

MTN Tries Video Streaming (Again)

Profitable companies can spend decades throwing good money after bad, chasing an idea their executives fell in love with. Nine times out of ten, that idea was pitched by a very well-paid consultant and money must not waste.

For outsiders watching said company, our instinct is to scoff and call them stupid. This is emotionally satisfying, but usually wrong. Many of the people who fall in love with these ideas are very smart, so now and then, it is worth figuring out how they think. There was a room where the idea made sense; the work is to enter that room, engage with the idea and then work backwards.

This may or may not be related to this week’s announcement that MTN is launching yet another video streaming service.

This week, MTN Group announced MTN One TV, “a new entertainment proposition” designed to make digital video more accessible across African markets. The service is tied to MTN’s Ambition 2030 strategy and promises the usual mix of local storytelling, live channels, international programming, free-to-view content, advertising-funded viewing, pay-as-you-watch access and subscriptions.

This is MTN’s nth attempt at streaming. One TV belongs to a family that includes FrontRow, VU, the eVOD partnership, the media layer inside ayoba, and now this new pan-African entertainment proposition. It’s safe to call streaming a company obsession.

One TV arrives at a moment when “video streaming is a dead business” has hardened into a continental cliché.

Canal+ shut down Showmax in April, presumably worried about the streamer’s losses and business model fit. Amazon and Netflix famously stopped funding African originals barely two years after they made splashy starts. And Jason Njoku’s IrokoTV, the most documented of the lot, spent $100 million over a decade and concluded that there is no Nigerian market for paid streaming. Yet MTN will not be deterred.

So let’s get into the mind of the execs. They’re running a company that spent twenty years making billions from setting up networks and connecting calls, a.k.a revenue from voice calls. As internet penetration started to deepen, in came Whatsapp/Telegram/iMessage offering the ability to make endless calls with very little data.

Nigerian telcos made peace with this shift and rebranded around data. Airtel called itself the Smartphone network, and Glo crowned itself the Grandmaster of Data.

But data is a crueler business than voice. To sell more data, you have to lay fibre cables, build network towers, buy spectrum, acquire customers, and now and then, hold a townhall swearing you’re not stealing customer data.

After spending all that money, you will enter price wars with competitors while regulators take ten years to approve price increases. Your customers will also have two SIM cards and buy data from whoever offers the better deal.

Investors understand this. They see a capital-hungry, price-pressured, churn-prone business. They also see the apps riding on top of that business like YouTube, TikTok, Netflix, WhatsApp, Instagram, DStv Stream. These companies keep customers glued to their phones, shape culture, collect attention, and, in the best cases, trade at the kind of multiples that make telecom executives stare into the middle distance during trading hours.

A savvy consultant does not need to work too hard here. Video is the single biggest thing on the network, and people buy data to watch things. The apps that make people watch things are building valuable consumer relationships on infrastructure telcos pay for.

So, MTN already has the customer, the SIM card, the billing relationship, the distribution, the call centre, the wallet, and the network. Why remain a landlord collecting regulated rent while the tenants get rich?

Capture the streaming layer, the consultant says, and you capture the customer. Capture the customer, and you can sell more data, reduce churn, build an advertising business, and give investors a platform story instead of a utility story.

Every sentence in that pitch is individually true. The trouble starts when you try to assemble them into a business.

Streaming drives data demand, and MTN sells data, so MTN already earns money when the customer watches a video, regardless of who owns the app. A customer watching YouTube on MTN is still an MTN data customer. A customer watching Netflix on MTN is still an MTN data customer. Owning the streaming service does not create extra gigabytes, because nobody in recorded history has watched more television out of loyalty to a network operator.

Ownership only matters if it captures something other than data revenue. Also, every time operators tried to make their own content services attractive, they all arrived at the same strategy: subsidise the data.

FrontRow launched with a paid subscription, but its quickly leaned on MTN data bundles and free-data promotions. The eVOD partnership gave MTN customers 4GB of free data monthly to watch. Ayoba grew by offering free data to MTN customers using its features. The products that are supposed to sell more data keep needing free data to be attractive.

If video is the great engine of data demand, why are you giving away the fuel?

If you argue that content bundles reduce customer churn, that mostly works where a customer signs a 24-month contract and the free streaming tips the decision. If a family is choosing between two 24-month mobile plans, free Netflix or Apple TV can tip the decision. That is a model in America and parts of Europe. It works because the contract is the wall, and the content is decoration on the wall.

But the Nigerian prepaid customer renews her relationship with MTN every few days, often in increments of ₦500, with another SIM card as an option. There is no 24-month contract to defend, so there is no wall.

So the old telco streaming thesis made sense in the boardroom but broke upon contact with the market. Telcos saw video eating their networks and assumed the video layer was where the money had gone. A customer using your network to watch a film is not secretly asking you to become a film company.

Daily Investor, after speaking to senior executives at South African operators, put the diagnosis bluntly: consultants kept recommending streaming because it was what operators wanted to hear. The publication reported that “consultants and consulting firms had a big role” in these streaming strategies. Voice revenue was declining, and Data was pressured while Over-The-Top companies were growing. The operator wanted a way to climb up the value chain, and the consultants sold one.

Yet, if streaming was truly strategic, the budgets didn’t match the conviction. Most telco streaming ventures were funded like side bets, small enough that failure would not hurt, which also guaranteed they were never large enough to matter. If you compete for attention with Netflix, MultiChoice, YouTube, TikTok, and pirate Telegram channels, you either bring frightening amounts of money, a distribution advantage that changes the economics, or a product so cheap and useful that people forgive its limitations.

Netflix’s content budget in 2022 was roughly $17 billion, more than the market capitalisations of MTN and Vodacom combined at the time. That comparison is unfair in the strict sense because MTN was not literally trying to outspend Netflix globally. But it is useful in showing the scale of the war that telco executives were entering.

What if the side bet was never meant to fight?

A streaming service too small to compete is exactly the right size for slide 14 of a capital markets presentation, under a heading about ecosystems. The CEO needs a story in which his utility is secretly a platform. The app becomes the physical evidence the story requires, and the funding is calibrated to the story’s needs, which are modest, rather than the market’s needs, which are not.

When you think about it this way, chronic underfunding stops being only the reason these services failed. It paints a picture of the apps as props, and props are not supposed to be expensive.

MTN’s Ambition 2025 promised that ayoba, its WhatsApp challenger and super-app experiment, would reach 100 million monthly active users. Ayoba reached roughly 35 million monthly users, helped by free-data incentives, before MTN began pulling it from app stores this March. Ambition 2030 has now arrived, and within months, it has produced One TV.

To be fair, the 2026 pitch is not the 2014 pitch because nobody can walk at the telco believes it should build the Netflix of Africa. Too many people have tried and too many expensive bodies are already on the floor.

The new consultant would sell the inverted thesis. Everyone who tried to be Netflix in Africa is dead, which is precisely why MTN should enter. MTN can stop pretending that streaming is the business; the business is billing, advertising, discovery, data, and distribution. The video is just the merchandise.

One TV is mostly being framed as a flexible layer rather than a premium subscription streamer. Free-to-view content where users will not pay. Advertising-funded viewing where attention can be monetised. Pay-per-view where the event is strong enough. Subscriptions where the market can bear them. Airtime and Mobile Money payments where bank cards are rare. Technology rented from Synamedia rather than painfully built in-house. Content assembled through partnerships rather than a suicidal originals budget.

The asset every dead streamer lacked, and burned fortunes trying to build, is the ability to collect small amounts of money from customers with no appetite for monthly commitments. MTN can collect ₦100 from a customer several times a day. Across its footprint, it serves more than 300 million people. It has Mobile Money, airtime billing and customer data. It has an advertising business that wants inventory. It has a network that benefits when video consumption grows, even if the streaming product itself is not glamorous.

Understood as a merchandising layer for carrier billing and ad inventory, One TV is not obviously stupid. It might never be loved and it will never be cool. It will not produce a single show anyone fights about on Twitter or Instagram. But it could still be useful if it gives advertisers more surfaces, creates more reasons for customers to transact through MTN, and turns the streaming product into one component of a larger system.

That is the charitable case.

The difficulty is that MTN has already tested most of the components separately, with a lot of evidence.

VU had the network relationship, the billing proximity, a subscription subsidy, and the MTN brand. It shut down in 2017.

FrontRow had MTN data bundles and later free-data promotions. It became VU, then disappeared.

eVOD outsourced much of the content risk to a broadcaster and still needed free data to make the proposition attractive. The three-year exclusivity came and went without becoming a continent-changing model. MusicTime had sachet pricing and data-inclusive listening. It was folded into ayoba. Ayoba had messaging, music, channels, mini-apps, free data, and the full weight of a super-app ambition. It is now being phased out.

Every pillar of the new thesis has an individual obituary somewhere in MTN’s archive.

The remaining claim is that the combination works where the components did not. Billing plus ads plus free content plus pay-per-view plus MoMo plus local partnerships plus rented technology. Maybe the old attempts failed because they were too early, too subscription-heavy, too South African, too content-led, and too eager to imitate Netflix.

One TV appears to be a third bet that streaming is a thin commercial layer on top of assets MTN already owns.

Can MTN create habitual viewing without spending like a media company? Can it sell enough advertising without becoming just another low-value ad surface? Can airtime billing turn curiosity into recurring revenue, or does it merely make it easier for customers to spend tiny amounts once and disappear? Can a telco build an entertainment product people choose, rather than one they tolerate because it comes with free data?

The answer may still be no. But at least this time, the no has to work harder.

Share

See you on Sunday!

Subscribe now

This is a Notadeepdive x Fintech Is Easy crossover episode, which means two newsletters get to annoy you at once. If you missed the earlier comps — Afrobeats, Nollywood, the Premier League table — catch up.

The World Cup kicks off today, and the Super Eagles have saved my “2026 World Cup” group chat tens of thousands of dollars by not qualifying. No flights to book, no jerseys to buy, so we now have to contend with being disappointed by the serial chokers: the Netherlands, Portugal, England, Belgium. Pick your poison.

Even the Pope got in on the fun this week, reminding star players to pass the ball. So while we wait for Ronaldo’s first knockout-round goal, let’s name a squad we can actually trust: Nigeria’s fintech starting XI.

The rules are simple. Position matters, not just fame. A striker who can’t defend doesn’t go in defence, no matter how big the valuation. And as always, this is very unserious analysis.

We’re playing a 4-3-3.

GK — NIBSS

Every save in Nigerian fintech is actually a NIBSS save. The keeper is the one player who is never allowed an off day, because when this one goes down, the entire league stops. You have lived through those afternoons when transfers hang in the air like a Pickford clearance. One of the best real-time payment systems in the world is still waiting for its flowers. Keepers are used to that.

RB — Paga

The veteran fullback. Was patrolling the agent-banking flank before half this squad was incorporated. Doesn’t bomb forward much these days, but positionally flawless; you only ever notice a good fullback when he’s missing.

CB — Interswitch (captain)

The armband was never up for debate. Charting since 2002, profitable, and reads the game easily. The forwards are louder, but when the ball has to go through, it goes through Interswitch.

CB — GTBank

The defensive veteran. Was doing digital banking before half this squad had a CAC registration, and Squad shows the legs haven’t gone. Doesn’t chase the ball, doesn’t dive in; just stands where the danger is going to be and lets the danger embarrass itself. Banks aren’t supposed to make this team, but try taking GTBank off the pitch and see how the defence holds.

LB — Kuda

The modern attacking fullback: electric going forward, occasionally caught upfield with its lending play and routine downtime notifications. The UK move means it’s technically playing in Europe now. The talent is obvious; the defensive discipline is a work in progress.

CM — Moniepoint

The box-to-box engine. Covers every blade of grass — POS, payroll, lending, SME banking — wins every duel, and still posts big numbers at the end of the season. The unicorn valuation makes it the first name on the team sheet.

CM — Paystack

The deep-lying playmaker. Doesn’t run much, doesn’t shout, but every attack in this ecosystem flows through its passing range. Signed by a bigger club (Stripe) yet still moves like it’s independent. Zap is the attempt to add goals to all those assists but the jury’s still out on that one.

AM — Flutterwave

The mercurial number 10. On its day, the most gifted player on the pitch with global reach, big-stage pedigree, and highlight reels for days. Also leads the league in yellow cards, and the disciplinary record follows it from country to country. You start it anyway, because the ceiling is too high to leave on the bench.

RW — LemFi

The pacey winger is doing its best football abroad. Hugs the diaspora touchline, beats established cross-border remittance companies at their own game, and whips remittances into the box at rates and speeds that have everyone else suddenly demanding to play wide and do remittances.

ST — OPay

The poacher nobody admits to rating. “I don’t even use OPay,” you say, as it scores its 30th of the season. It’s in every box, on every street, topping every download and NIBBS NIP chart.

LW — Grey

The Lamine Yamal of the squad. Young enough that the veterans remember its debut, already confident enough to take on the adults — Stablecoins, the banks, multicurrency accounts, business banking, cross-border banking itself — and win in multiple markets. Moves money across the world like it’s a street game: tricks, nutmegs, features pulled out of nowhere. The scary part is the ceiling. It hasn’t even peaked.

The Super Sub

PiggyVest (the closer — comes on in the 70th minute to lock down whatever you’ve saved)

MANAGER — CBN

Tactical instructions arrive by circular, occasionally at half-time, sometimes reversing the instructions from the first half. Nobody knows the formation until the press conference, and even then. But say what you want: every player on this pitch is here because the gaffer licensed them to play.

I’ve named my XI. Now do your worst in the comments: who got snubbed, who’s playing out of position, and who deserves the armband instead. If your favourite didn’t make the squad, that’s between you and the manager.

This belongs in the group chat. You know the one.

Share

Subscribe now

Around here, we talk about the idea that lending is easy. Getting customers to repay on time is usually where things get a little sticky. Since most customers are subprime borrowers who have no collateral, digital lenders must have their wits about them. At some point, a part of the industry decided that fear was a good solution to this willingness-to-pay problem.

That means when you install a digital lending app, it asks permission to scrape your contacts, photos, and call logs. And when you fall behind on repayment, you’ll learn what that permission was for. Prepare to have messages sent to your mother, boss, or some other unfortunate random contact announcing that you’re a thief and a debtor.

None of what I’ve just described is a secret. The Federal Competition and Consumer Protection Commission (FCCPC) has spent years documenting it. Between 2023 and 2024, it delisted 47 unregistered loan apps with Google’s help and placed dozens more on a watchlist, eventually convening a joint task force with NITDA, the police, the ICPC, the EFCC and the National Human Rights Commission to deal with apps that took customers’ contacts and images and used them to harass, intimidate and shame them.

By the time the FCCPC sat down to write a comprehensive rulebook, every serious person agreed it was about time.

The rule was the Digital, Electronic, Online and Non-Traditional Consumer Lending Regulations (DEON), and it took effect on 21 July 2025, with full enforcement from January 2026. The rules had the obvious right things like mandatory registration of digital lenders, clear disclosure of rates before disbursement, a ban on contact-harvesting, and a ban on shaming borrowers to force repayment. If you have read a loan-app harassment story, you’ll get the sense that DEON should have arrived years earlier.

Which makes what happened next peculiar.

Buried in a rule written for digital lenders was a decision to classify airtime lending as consumer lending.

And airtime lending is a different animal from payday loans. When you run out of airtime and still need to make a call, your network provider offers you ₦100 and you repay at your next recharge, plus a fee. There is no app harvesting your contacts, no recovery agent, no shame.

The telco takes its money back from your next top-up. Like most good technology, it works so smoothly that the customer never thinks about the machinery underneath. I wrote about that machinery last August, because it is a genuinely good business that pulled an entire telecoms subsector — Value Added Services, or VAS — back from the dead.

If VAS was on the ropes before airtime lending, they had it coming. These were the people behind the caller tunes, the horoscope SMS, the “win a car” promos. For years, their trade was the unsolicited subscription and services you never asked for but were signed up for anyway, debiting your airtime.

In 2016, the Nigerian Communications Commission (NCC) gave the networks and their VAS partners until June to stop sending unsolicited messages or face a ₦5 million fine plus ₦500,000 for every additional day, and gave subscribers an option to opt out with a Do-Not-Disturb code.

In December 2018, after a two-year audit of call records and subscription logs across every network, the NCC concluded that a huge share of VAS subscriptions had never been voluntary, and that some providers had built deliberate mechanisms to force-subscribe people and drain their airtime without consent. The regulator put the illegal charges at around ₦36 billion in just the five months to December 2018, ordered refunds, and threatened to suspend or decommission the offending platforms outright.

So when airtime lending came along and offered the VAS industry an honest, enormous line of business, the people in it did the sensible thing and kept their heads down.

The sector is not yet spotless, and one fair criticism of airtime lending is that it operates entirely outside the formal credit system, so a Nigerian who borrows and repays faithfully for years builds no credit history a bank can see, and the data on his reliability sits with the platform rather than a bureau. But a business that advances small sums and gets paid back is a long way from one that forcibly debited millions of people for horoscopes.

With DEON due to come into force in early 2026, the airtime lenders carried on as though it had nothing to do with them. The assumption was that airtime is a telecoms service, telecoms is the NCC’s house, and a consumer-protection regulator surely couldn’t redraw it. But the ground was shifting, and almost nobody in the sector was looking down.

They looked down on April 2nd, when the FCCPC directed MTN, Airtel, Globacom, and 9mobile to comply with DEON immediately. Rather than risk the penalties, the operators switched off airtime lending. A rule written to stop loan apps from terrorising defaulters and sanitise digital lending had, as its first visible act, taken emergency airtime away.

I wrote in April that a cynical person might notice Nigeria has never met a healthy pool of cash without wondering how to insert a middleman into it. I’m looking at you, ARCON.

The FCCPC’s case for pulling airtime into DEON was not, in fact, harassment; its argument was competition. The Commission says its findings showed that some operators ran “exclusionary third-party technical arrangements” in breach of the Federal Competition and Consumer Protection Act, exclusive deals that wired a single technical partner into a network’s airtime-lending pipe and shut rivals out.

DEON’s remedy followed from that diagnosis. Any operator lending airtime must use at least two service-activation providers, one of them fully Nigerian-owned, cannot strike a lending deal with a dominant intermediary without the Commission’s approval, and may not enter exclusive arrangements that foreclose competitors. The framing was free-market, and the rules, the Commission said, sought to open the market to local participants alongside foreign partners.

The trouble with this remedy is that the competition FCCPC claimed to be angling for actually arrived while the Commission was busy engineering it.

Commercial banks like GTCO launched a Quick Airtime Loan, lending ₦100 to ₦10,000 over your *737# code at 2.95% a month against the telcos’ 15%, running it on its own USSD rails and HabariPay’s payment infrastructure, with repayment pulled automatically from your account on the next inflow.

It asked no one’s permission, partnered with no telco, and sought no airtime-lending licence. It was a bank; advancing small sums and recovering them from an account it already holds is the most ordinary thing a bank does.

When a vacuum opened in April with the suspension of airtime lending, more banks jumped in; Access, Zenith, UBA, FirstBank, and FCMB all pushed their own airtime-loan codes within weeks. A market the FCCPC was trying to legislate into competition was, at the same moment, being competed for at a fifth of the price by firms its regulation never had to touch.

The banks aren’t a perfect substitute, of course, because you need a funded bank account to use them, so they don’t reach the unbanked subscriber the telco model serves at ₦0 balance. But the direction is that the thing the FCCPC said it wanted — more players, a cheaper product, the customer better off — happened from a class of competitor the rules didn’t anticipate, the moment there was a gap to fill.

A competition regulator’s job is to lower the barriers and punish the abuse — break the exclusive contract, ban the tying arrangement, open the field — and then let whoever wants to compete arrive on their own terms.

Clearing the field enables competition. Deciding that an airtime lender must use two intermediaries, that one of them must hold a particular nationality, and that the Commission must approve each arrangement in the chain, is the regulator specifying the number and identity of the players rather than the conditions of play. The first produces a GTBank at 2.95%. The second produces a queue for a licence.

You could grant the FCCPC its entire diagnosis — exclusive arrangements existed, concentration was real, foreign firms held data on millions of Nigerians that never touched a local credit bureau — and still say the cure was wrong. An enabling environment doesn’t hand-pick its beneficiaries. The surest sign that a rule is about competition rather than capture is that it tells you what conduct is forbidden and then falls silent on who gets to show up.

So, a rule almost everyone wanted went to court over the part almost no one needed. WASPAN, the association of Nigerian NCC-licensed firms that run this business, argued that consumer protection does not stretch to redesigning a telecoms market, which is the NCC’s remit.

Airtel and Glo restored airtime lending soon in late May, but MTN has not rejoined the party.

The FCCPC had a real problem, a good rule, and a public that wanted it enforced. It had, in the loan-app provisions, exactly the tools to clear the field: ban the abuse, force disclosure, open the market, and let the banks and the fintechs and the aggregators arrive on their own.

Instead, the Commission reached past all of it for a clause that specified who could stand in the airtime pipe and what passport they had to hold, lost the rule it had taken years to build, and stalled a service millions of people use to stay connected. The loan apps it was actually written to stop are, presumably, still out there, texting somebody’s mother and accusing them of being thieves.

Share

Subscribe now

TOGETHER WITH CREDIT DIRECT

The answer might surprise you.

The Nigeria Consumer Credit Insights Report breaks down who borrows, how much, for what, and what it reveals about financial behaviour across the country.

The answer to that question is just the beginning.

Read It Here

Nigeria’s BNPL flavour

The point of Buy Now, Pay Later is that you want something. A coat, concert tickets, an iPhone 17 Promax. You'd like to have it now and deal with the cost later, in some abstract future. Klarna built an empire on this at the checkout pages of Nike, H&M, and, lately, Walmart. The things Americans most often finance this way are televisions, appliances, and the odd sofa. It's a way to buy what you can't afford and tell people that it works for your cash flow.

In Nigeria, people are using Buy Now Pay Later (BNPL) in counterintuitive ways. Most of the people using BNPL don’t have salaries.

Almost half of all the BNPL transactions in Credit Direct’s consumer credit report are from self-employed people, against under a third for people who draw a wage. The self-employed people even borrow bigger, which sounds backwards.

The premise of lending to a consumer is that the consumer has a payslip you can point at. Here, the people with none of that are the main customers, and you should see what they’re buying.

Top of the list are mid-range Android phones that cost ₦130,000 to ₦280,000, which the report calls the sweet spot. These are the most expensive phones you can get without putting up cash. (iPhones turn up too, albeit as a smaller slice, bought by people who could mostly have paid outright and chose to pay in bits anyway. That's the one corner of the Nigerian market doing the American thing.)

Picture the customer. You sell things like phone cases, frozen chicken, ankara, or work hours as a dispatch rider. While you don’t earn a salary, there’s a trickle of small payments that arrive when they arrive. So the phone is the shop where the orders come in, and where you check whether the transfer actually went through. But it costs more than you clear in a month.

When a factory borrows to buy a machine, we call it asset finance and give it a banker. When the woman selling provisions on a small scale borrows to buy a phone, that’s the Nigerian flavour of Buy Now, Pay Later.

M-KOPA has put more than $1.6 billion of credit into the hands of everyday earners — hairdressers, boda riders, carpenters, people with steady work and no payslip — mostly to buy smartphones, repaid in daily instalments that map how money reaches people like that. It says most of its customers use the phone to earn.

If you miss your payments, the phone switches off. It feels a little sinister, but it’s effective.

Asked how long they want to repay, most borrowers choose the longest term on offer. The monthly figure on anything shorter is one their cash flow can't survive, which is, again, why they're paying in instalments at all. The product assumes someone spreading out a treat. The person on the other end is rationing a business expense.

Maybe the self-employed borrow this way only because nobody else will lend to them, a phone is just a phone, and I'm reading a small-business epic into people buying Androids. Maybe. The Nigerian data can't see why anyone buys anything; it sees who, what, and for how long.

In America, Buy Now, Pay Later is sliding downhill, and people have started putting groceries on it. In Nigeria, BNPL was always a small-business loan for the millions who'll never be offered one.

Read Credit Direct’s Consumer Insights Report here.

Subscribe now

Ngozi Chukwu, the author of this newsletter edition, will begin publishing “Footnote” on Notadeepdive every Thursday. She’ll write about food delivery, fintech and whatever else captures her imagination. That’s one more reason to subscribe:

Subscribe now

TOGETHER WITH CREDIT DIRECT

The answer might surprise you.

The Nigeria Consumer Credit Insights Report breaks down who borrows, how much, for what, and what it reveals about financial behaviour across the country.

The answer to that question is just the beginning.

Read It Here

Restaurant catfishing

Every time you use a food delivery app like Chowdeck, Glovo, or Swoop, the technology works so well that you don’t have to think about it. You want food, you know the restaurant exists, and you cannot be bothered to leave your chair/bed.

Somewhere in this arrangement is the assumption that the food delivery platform has an active relationship with the restaurant and that the food you receive comes from the restaurant you selected. It’s reinforced every time you walk into that same restaurant and see dispatch riders with their branded t-shirts picking up orders.

That arrangement has recently become a matter of public debate.

Three times in the last six months, Nigerians on social media have experienced collective panic over what can only be described as “restaurant catfishing,” the strange idea that the restaurant you’re ordering from may not be the restaurant.

The first two times, there was an impostor. In December, Corporate Ewa alleged someone had registered its name on Glovo and was buying food from roadside kitchens near Ojuelegba to fill orders, with one customer reporting food poisoning. Glovo delisted the account and promised an investigation.

In May, Techpoint built a fictitious restaurant, registered it on Glovo and Chowdeck, and got an order out the door before either platform flagged it. Both incidents are fraud. Fraud is easy to get upset about, and because it has a villain, the obvious fix is to check who you are onboarding/fix your KYC process.

Norma was the third time, and here, there was no impostor.

On May 28, the Lagos suya spot asked Chowdeck to remove its name, logo, and menu because it had never agreed to be on Chowdeck. How did customers receive orders from a restaurant that hadn’t consented to listing on Chowdeck?

The explanation came this week with Chowdeck clarifying that an agent it calls a shopper stands near popular restaurants, buys the food a customer ordered through the app, and hands it to a rider, no agreement with the restaurant required.

To clear up the confusion, it will now add badges — Verified, Awaiting Verification, Shopper — so you can tell a partner from a shop Chowdeck is merely buying from. Chowdeck framed it as a vulnerability in a system built to support small businesses.

The most obvious question, which many publications have asked (not a bad thing!), is “Is it fair to make a business carry the reputational risk of a listing it never approved?”

But assisted fulfillment, a.k.a. using a shopper, is not really a question of fairness to restaurants because, at face value, the restaurant loses no money.

But money isn’t everything. Restaurants whose items get delivered without their consent have no control over delivery temperature, food handling, or timing. There’s also the reality of customers blaming the restaurant for a bad delivery it never authorized. It’s entirely possible to lose significant brand equity because of a delivery process it has no control over.

Every order Norma didn’t consent to fulfill builds Chowdeck’s ratings and retention. Fast-growing restaurants, like Norma, which launched its fifth location last year, that are worth taking these customers from are exactly the ones with enough pull to refuse to list on Chowdeck.

But assisted fulfilment is cheap customer acquisition for a wannabe super-app because the brand that’s pulling the customer’s attention belongs to someone else. So, the Shopper badge isn’t protection for the restaurant, and it’s not because anyone cares about small businesses. It’s Chowdeck’s claim on a Norma customer that Norma will never meet.

Norma keeps the full counter price of every catfished order because it pays no commission and runs no promotion. A restaurant that actually signs with a food delivery service will pay for the privilege of demand generation.

When Lucky Boy, a Pasadena burger stand, sued Postmates in 2021, it set the commission at roughly 30% of every order.

On a per-order basis, the shopper model is bad: Chowdeck collects no commission, pays a shopper’s time, and floats cash to buy items at full counter price. It only works when you accept that the food is the cost of acquisition and the customer is the asset.

This makes the list of catfished restaurants a proxy for Chowdeck’s leverage. The unconsented restaurants are a map of the supply Chowdeck has demand for but cannot get under contract. While conventional wisdom about delivery apps is that they aggregate demand and restaurants compete to supply it, the Nigerian restaurant market is small enough that the demand sits with the brands, and the food delivery apps end up competing. Case in point, all delivery apps have a rotating cast of deals with Chicken Republic and The Place, suggesting the leverage these big franchises have.

American food delivery startups ran this assisted play on the way up, and the way it played out is a cautionary tale.

In 2015, In-N-Out sued DoorDash for delivering its burgers under an imitation of its logo after being told to stop. Lucky Boy said Postmates posted wrong prices under its name, told customers it was closed when it was open, and steered people searching for Lucky Boy toward restaurants that paid. Grubhub built much of its catalogue this way; the FTC found last year that as many as half its listed restaurants, up to 325,000, were there without consent, that it tended to remove them only when threatened, and that it treated the listing as a lever to sell a partnership. The penalty was $140 million, but Grubhub eventually paid $25 million because that’s all it could afford. Is there a Nigerian In-N-Out with the standing and appetite to sue, or an FCCPC with the appetite to act as the FTC did?

The delivery platforms had compelling incentives to “steal” the customer. DoorDash did not post a profit as a public company until late 2024, and it got there on scale built largely in these grey years of assisted fulfilment. Delivering food is a bad business.

“The economics are tough in this market because the costs are very high and there is plenty of competition, so there is downward pressure on the commissions that we make and upward pressure on marketing costs because everyone is fighting for customers.” - Francis Dufay

Which is why you see a lot of these platforms talk about building super apps. They open dark stores, buy a startup that sells inventory software and hardware, all to one end: owning the relationship with the customer.

You could argue the relationship isn’t worth much yet and that in a market this thin, assisted fulfilment is scrappy catalogue-filling rather than a land grab.

But that objection answers itself. Assisted fulfilment is only worth the legal and reputational exposure to someone who already believes the customer is the prize, which is pretty much the super-app thesis. Chowdeck is wagering that the relationship it is building on Norma’s name will be worth more than the strain of the brand pushing back or going to court.

Share

If this newsletter was forwarded to you, subscribe for free:

Subscribe now

TOGETHER WITH CREDIT DIRECT

You don’t need an occasion to tell them they matter. Gift an investment because they deserve to know it today.

Gift A Yield

The three-body theory

Sometimes the corporate world aligns the incentives of three otherwise unconnected entities.

Imagine the first entity as a late-stage startup that needs cash but doesn’t want to raise money at a down round. The second entity is a government that needs a very big and public technology-sector win. The third entity is really a hanger-on; a stock exchange that needs an Initial Public Offering (IPO) to prove its startup board has not been a complete failure.

The startup’s problem is that its old and rich valuation was possible when it still had a stronger story. But no matter what story it tells, it’ll need new buyers of that story sooner or later. The new buyers will poke and prod the numbers until it’s a third of what it once was. It could also list abroad and discover that public-market investors are less sentimental than private backers. A sale is almost out of the question because of said rich valuation.

So the startup needs money, but not just any money. It needs money that takes its word at it, without all the aforementioned poking and prodding.

The government has a different problem: it cannot take pictures of reforms. A factory opening, or a bell-ringing ceremony with a founder and a stock exchange official, can deliver hard evidence of efforts.

With technology advertised as the future of many emerging economies for fifteen years, a listing of a recognisable name on the local exchange, even in a dual listing, is premium photo-op material. The president will shake the founder’s hand, the exchange gets its listing, and the startup gets its money. Incentives are perfectly aligned.

The trouble is that IPOs are complex. The companies large enough to matter have spent the last decade structuring themselves to be incorporated abroad, dollar-denominated, and bound for foreign exchanges. The government cannot summon a listable champion out of thin air; it can only endorse one that already exists.

If the big win from this kind of endorsement is public approval, then you need to sequence this properly. Maybe a meeting at the presidential office, a trip abroad, and something with ministers.

Yet, each of these has to be a step the company wants to take, because this has to be a company large enough to make the endorsement worth giving, yet in a position to welcome friendly capital, and flexible enough to accept the ceremony that comes with it. Remember, what the government wants out of this is a recognisable win.

The exchange, which is the smallest party in the room, has a rather straightforward need: a splashy IPO. While the exchange has no leverage, it has patience and a willingness to accept a dual listing. The alternative is no technology IPO.

Welcome to the entirely hypothetical Three-Body Theory (please do not take capital raising advice from a newsletter!).

The Three-Body Theory is a situation in which three parties with different problems need the same transaction to happen. The startup needs capital without a down round; the government needs a public technology-sector win, and the exchange needs one credible listing.

Most people think of that transaction as the IPO, but the transaction is the round before it. It is the money that gets the company from its current problem to the bell-ringing ceremony everyone believes is coming. In one sense, this is just a bridge round.

For the startup, the round reduces pressure because cash buys time, which is a better alternative to raising money at unfavorable terms. If it can raise enough money before the IPO, it does not have to rush into a listing window that may be unkind or accept a private round that tells everyone the last valuation was a little too generous.

Also, the useful thing about strategic money is that it does not have to be all the money. It only has to arrive first.

Once the cheque is in, other money has a reason to gather. Nobody has to say they are joining a round designed to postpone a harder conversation about valuation. That is the elegance of the bridge.

The startup says strategic capital; the government says innovation, while the exchange says market deepening. Existing investors will say continued support.

The problem is that the three bodies do not share a calendar. The startup wants process, controlled disclosure, careful language, and investor alignment. No serious company wants its funding announcement treated like a ribbon-cutting. But the government wants credit and wants it now, so there’s bound to be friction there.

The existing investors, in this theory, are watching with the patience of people who know the best outcome needs compromise. You can hold on to your ideals, but forcing a company into a bad IPO is not a good outcome.

A bridge round may dilute investors, but that’s hardly fatal.

If the round keeps an IPO alive, buys time, and allows everyone believing the old valuation is still within reach, dilution is an acceptable tradeoff.

Nobody gets everything, but everybody gets enough to keep them happy.

Share

And if this email was forwarded to you, subscribe here for free:

Subscribe now

TOGETHER WITH CREDIT DIRECT

A thoughtful gift lasts a moment.

A smart one grows with them.

Gift an investment.

Gift A Yield Today

Ships vs Speedboats

There are two ways to think about banks and fintechs.

The first is the ship. It’s large, visible, regulated, full of people, impossible to ignore, and not built for sharp turns. Ships are slow, not because their captains are stupid, but because they have the responsibility of balancing people, history, compliance departments, regulators, risk committees, and millions of customers who will bust a nerve if that ₦10,000 transfer hangs.

The second is the speedboat. Small, noisy, exciting, occasionally reckless. It can turn fast, crash into the side of the market, repaint itself…it can move fast and break things.

If you run a ship and you see speedboats cutting through the water, your first instinct could be irritation or envy. Your instinct, if you’re Segun Agbaje, is to build one.

A few days ago, Agbaje told Nairametrics that GTCO is no longer worried about fintech competition.

“Everybody was really nervous about fintech, so we built our own speedboat. That’s Habari.”

“Habari is competing very, very effectively and it means we are not scared about the threat of fintechs any longer. We have a very strong engine to compete with them.”

Agbaje has been talking about this speedboat years before HabariPay became an enviable revenue line in GTCO’s annual report. At some point, this speedboat was a theory about what banking would become, and at other points, it represented theories about fintech valuations, payment income, and why banks could no longer judge their competition by looking at other banks.

For a while, it was also a theory about music, shopping, videos, books, friends, bill-splitting, utility payments, and 10,000 small businesses in a super app built by one of Nigeria’s most profitable banks.

Those theories, a.k.a the first version of Habari, did not win.

By 2020, Agbaje’s fintech anxiety was public. In an interview, he said competitor analysis could no longer stop at banks. Banks had to look at fintechs, telcos, FMCG companies, and even betting companies because anyone with mobile wallets and customer data had become a potential competitor in financial services.

He was especially blunt about the pace of change. “Three years ago, we were talking about this like it was way, way in the future. It’s here today.”

The reason for urgency was the great age of fintech unbundling. PiggyVest took savings and made it social and savvy. Carbon, FairMoney and others took consumer lending and made it instant and easy. Paystack and Flutterwave made online payments stop feeling like painful negotiations with your bank’s e-business team. OPay came with distribution and a frightening willingness to spend their way to scale.

At Social Media Week Lagos in 2020, Agbaje pointed out that GTBank, one of Nigeria’s best-valued banks, was trading at around one time book value, while payment companies were valued at 30 times earnings. “What is there not to like about this space?” he asked.

He also cited Frost & Sullivan’s projection that Nigerian fintech revenues would grow from $153.1 million in 2017 to $543.3 million by 2022, and drew the obvious conclusion that 30% of banking income was at risk.

If the market was going to reward payments like software and punish banking like banking, GTBank needed to move some of its economics into the right bucket.

The bank already operated GTPay, GTCollections, QuickCredit and Habari. It was number one in NIP inflows and outflows for a period in 2020, with more than 18% of outflows and almost 17% of inflows. Agbaje told analysts Paystack might start bigger and know more on day one, but GTBank would “bridge that gap very quickly.” GTBank was going to win payments “very easily.”

TechCabal’s 2020 framing was that GTBank’s fintech ambitions could lead to a future billion-dollar IPO. Worldpay, an example Agbaje has cited, started inside a bank and became a huge payments company.

The public mood on banks venturing into fintech was split.

Nigeria’s Tech founders generally believed banks were ships; too slow to beat startups at their own game. Bank people believed startups were underestimating distribution and the violence of cheap deposits. Agbaje believed GTBank could have the advantages of the ship with a speedboat alongside.

When GTBank launched Habari in November 2018, Agbaje went out of his way to say it was not a bank. Habari was Nigeria’s largest platform for music, shopping, lifestyle content and more. It had music, videos, books, shopping from over 10,000 small businesses, friend discovery, bill splitting, transfers and utility payments. GTBank said it was “less about us as a bank and more about our customers and everything they need to enable their lifestyle.”

People might not want a bank, but they still needed banking. The winning company would not necessarily be the one with the best branch network or the largest loan book. It would be the one embedded in the customer’s daily life, catching payments, commerce, content, credit and identity as they moved through the same interface.

It was a very 2018 idea; a time when everyone wanted to be a platform/superapp. The broad idea was to take the boring licensed activity and wrap it in something shinier and more social (it sounds like the kind of pitch old people believe as a way of keeping up with the times).

If Habari worked, GTBank would own the customer context around the payment. Habari did not become the Nigerian WeChat.

The problem with bank-built lifestyle platforms is that customers can already smell the “trying to be cool” by committee setup. Nobody wakes up looking for a bank-owned place to stream music, read books, shop and find friends.

By 2021, GTBank had become GTCO, a holding company built to house banking and non-banking subsidiaries. The fintech ambition morphed into HabariPay, and the consumer super-app dream gave way to HabariPay, with a practical flagship product called Squad that would compete in payments.

HabariPay’s business today is payments infrastructure. It runs a payment gateway that processes transactions through virtual accounts, USSD, cards and bank transfers. It also has switching services and value-added services like bills, airtime and bulk SMS through licensed aggregators. Squad gives merchants payment links, storefronts, POS terminals and other tools for getting paid. It is a much better business than the original Habari.

In 2025, HabariPay made ₦12.9 billion in operating income, up from ₦5.8 billion the year before.

Profit after tax rose from ₦3.8 billion to ₦9.7 billion. Operating expenses doubled to ₦3.2 billion.

HabariPay had ₦12.9 billion in operating income and spent ₦3.2 billion to run the business. If it was a standalone venture-backed fintech trying to hire aggressively, subsidise customers, fight fraud, pay cloud bills, buy POS terminals, build a field force and advertise its way into market share, that cost profile would be impossible.

GTCO has found a way to make money in fintech without behaving like a venture-backed fintech. It did not need to win every consumer’s wallet or become the main app on a trader’s phone. Most importantly, it did not need to subsidise a generation into changing their behaviour.

It took the route of merchants, payment acceptance, group relationships, operational discipline, and a cost base that still looks like GTBank.

So this victory speech needs a footnote.

HabariPay is profitable, but it is small. Its ₦9.7 billion profit was about 1.1% of GTCO’s ₦865 billion group profit in 2025. Its ₦12.9 billion operating income is nice, but it is still a rounding error beside the mothership.

It’s a good holding-company story but it is not yet a “we beat fintech” story.

The consumer wallet war feels fought and won. OPay reportedly had over 50 million registered users in Nigeria by mid-2025, with around 10 million daily active users. PalmPay claims 35 million users. Moniepoint, which started from the merchant side and has been moving into personal banking, recorded $264.51 million in 2023 revenue, processes more than one billion transactions monthly and serves ten million businesses and individuals across Nigeria.

GTCO has not built OPay, PalmPay or Moniepoint. It has built a profitable payments subsidiary.

GTCO never needed to win the entire fintech market. It needed to stop fintechs from eating profitable pieces of its business without a response. On that narrow thesis, HabariPay is progress.

The less generous reading is that Agbaje has moved the goalpost from “we will win payments very easily” to “we are no longer scared.” Those are different claims.

There might be a lesson in there that banks rarely beat startups by becoming startups. They beat them by copying the part of the startup model that makes money, then attaching it to a cheaper balance sheet, a trusted brand and existing customer relationships.

Ships do not need to win every race against speedboats. Sometimes they just need one fast enough to stop passengers from asking why everyone else looks more alive.

Share

See you on Sunday

Subscribe now

TOGETHER WITH CREDIT DIRECT

You can wait another year, or you can start earning on Yield today.

Explore Yield

Jumia is not a doorstep delivery company

The general idea when you think of Jumia is of an e-commerce platform you visit when you want to buy, say, a new phone. You sift through a bunch of really expensive options, pay online, and it gets delivered to your doorstep in a major city like Lagos or Nairobi.

But one part of that picture isn’t completely accurate because Jumia isn’t really a doorstep delivery company. At least not anymore.

In Q1 2026, 74% of Jumia packages were fulfilled through pickup stations rather than door delivery, up from 67% a year ago. So the correct picture is: you go to Jumia, order a phone, and pick it up the same day or a few days later at a pickup station.

Skipping the last mile is a strategic masterstroke for Jumia’s unit economics. They still move products through warehouses and logistics partners, but they no longer have to send a rider to a customer’s address, call three times, and return with the package when the person isn’t home.

It’s a key reason Jumia’s fulfilment cost per physical goods order stayed flat at $2.06 even as volumes grew sharply and fuel prices increased in Nigeria and other African countries.

The mental picture of ordering from Jumia is wrong on geography, too. This is not an e-commerce company selling phones to people in Lekki, Kilimani and Zamalek. In Q1 2026, 62% of Jumia’s orders came from upcountry regions. (By “upcountry,” Jumia means regions outside its densest e-commerce markets. Places like Gaya, Okene, Yenagoa, Owerri, Wurukum, Ekpoma-Eguare and Ogijo.)

In Lagos, walking to a pickup station may be stressful because you have a ton of options. In a secondary city like Okene, where the item you want may be unavailable, overpriced or dependent on a trader who is also waiting for someone else to bring it from Lagos, walking to a pickup station is part of the bargain.

In Nigeria, Jumia added more than 80 pickup stations in Q1 2026, and its distribution network still covers only about one-third of the Nigerian population. Côte d’Ivoire, one of Jumia’s more mature markets, is closer to 60%.

Jumia is narrower, more physical, and more local.

Since taking over in November 2022, Francis Dufay has methodically killed the distractions. Jumia Food and Jumia Prime are gone. JumiaPay is no longer the center of the story; the company will stop reporting Total Payment Volume as a primary metric by Q2..

Physical goods are the business and the path to the promised land of adjusted EBIDATA breakeven by Q4 2026 and profitability by 2027.

The path to breakeven is to make the platform bigger, take more from each transaction, and stop the cost base from growing wildly.

Asked about the path to cash-flow positivity, Dufay said there are no mysterious blockers; “It’s mostly an execution game.” The company has to keep scaling the top line, improve unit economics, and reduce fixed costs in absolute terms.

As part of increasing its top line, Jumia increased commissions and take rates across most of its countries in January 2026.

This is an important shift because, less than two years ago, Jumia was still talking about keeping commissions stable so merchants would keep supplying the platform. Now, Jumia believes it has earned the right to take more because the platform is delivering more volume, better service and better access to customers.

Yet, even commissions and take rates have a ceiling, so it’s good to see that the company is increasing revenue from other business lines.

Marketing and advertising revenue grew 44% to $2.2 million, and value-added services revenue grew to $1.7 million, helped by warehousing fees from higher volumes, especially from Chinese sellers. Advertising revenue is still only about 1% of Gross Merchandise Value (GMV), and management believes there is room for improvement.

The second thing Jumia has to execute on is supply.

A pickup station does not mean much if the things people want are unavailable, expensive or slow to arrive. It’s challenging to have a catalogue that’s good enough with compelling prices.

This is why the international sellers from China and Turkey using Jumia are crucial; gross items sold from international sellers grew 87% in Q1 2026. According to Dufay, the current growth from international sellers is the result of three to four years of work. A new Chinese vendor can take more than a year or two before making a meaningful GMV and margin contribution.

So Jumia has had to convince merchants who don’t know Lagos, Nairobi, Abidjan or Accra to send inventory here. It is slow, operational work but the payoff is that these international sellers help Jumia widen its assortment. They often operate in categories like fashion, accessories and home and living, where gross profit ratios are higher. They also use more advertising and storage services, which means Jumia can monetise them beyond the basic commission.

This version of Jumia, a marketplace where merchants pay for access, storage, visibility and distribution, is one shareholders should care about.

Jumia also has to execute on cost, and because this is 2026, that involves saying AI a lot. On the Q1 earnings call, AI is mentioned 14 times.

Management says AI is being used in cybersecurity, code quality, accounting, HR reporting, logistics, customer service and seller management. Some of this is the usual public-company AI signalling, and shareholders will probably enjoy it for that reason alone.

There’s also more talk of layoffs, with a plan to let 200 people go this year. When Dufay took over at the end of 2022, Jumia had 4,318 employees. By March 2026, it had just over 1,980 employees. Nothing like talking about AI while cutting headcount to excite shareholders.

Costs rose sharply in sales and advertising to $5.1 million in Q1, and management says it views that positively because the spend is going to customer acquisition and engagement on the back of stronger product fundamentals.

There is a delicate balance in play. If you spend too little, GMV growth will slow, but spend too much, and the road to breakeven will continue to stretch ceaselessly into the horizon. Spend badly, and Jumia returns to the old problem of acquiring customers who never come back.

Dufay told analysts that Jumia’s marketing spend ratio is still lower than that of much larger emerging-market e-commerce players, and that most of the budget goes into online channels where spend can be moved daily or weekly if a market is not responding.

So the model is not simply “cut your way to profitability.” That would be easier to understand and less interesting. Jumia is trying to spend more on the parts of the business that drive GMV while cutting the parts that do not need to grow with GMV.

The business that now makes the most sense is not the one with the grandest internet story. It is the one that gets a phone case from a seller in China to a pickup station in Gaya, charges the merchant for visibility, charges for storage, takes a commission, and keeps the fulfillment cost flat.

After years of big promises, the Jumia story has become smaller and therein lies the path to profitability.

Share

If this newsletter was sent to you, subscribe here:

Subscribe now

TOGETHER WITH CREDIT DIRECT

Owning your dream gadget is much easier when you don’t have to pay the full price up front. 

With Credit Direct Checkout, you only need a 25% down payment to get that iPad, laptop, or more. 

Know Your Limit

Super app dreams

If you go to a VC today and tell them you’re looking to raise money for a food delivery startup in Nigeria, you might get walked out of the room. They have seen and heard it before and may be understandably jaded.

While Chowdeck and Glovo are giving the sector some hope, you cannot miss the issues. Restaurants do not want to pay commissions, so they pass the costs to customers who also underestimate what it truly costs to get a meal to their doorstep. A recent Substack essay about Chowdeck went viral because it captured customer frustration. Delivery fees and service fees are trending upwards, while conditions around subscription fees are changed without notice.

If a startup is constantly adding costs and changing terms under the feet of its customers, it is unlikely that unit economics are great.

Francis Dufay, Jumia’s CEO, gave the best soundbites about this business when Jumia shut down its food delivery arm in 2023. At the time Jumia killed food delivery, the business accounted for about 11% of Jumia’s GMV in the first nine months. It had also been Jumia’s fastest-growing category for years.

Dufay’s argument was that food delivery was difficult everywhere, with “challenging economics and big losses.” In Africa, the business has high costs, heavy competition, downward pressure on commissions, and rising marketing costs because everyone is fighting for the same customers.

So “we are building food delivery in Nigeria” is probably not the story you want to tell investors, and as I’ve said before, fundraising is storytelling. What’s your alternative story?

You can say food delivery is the first wedge into a much larger consumer platform, and that restaurants are really merchant acquisition, riders are logistics infrastructure, and the payment flow is really the beginning of a wallet. Repeat lunch orders are frequency and customer data is credit underwriting.

Basically, you could tell a story about a superapp.

Super apps are seductive because they turn a difficult first business into a larger constellation of businesses. The first business may have thin margins, difficult customers, operational stress, and unclear profitability. The other businesses improve the story with a theoretically larger addressable market.

Some start from payments, banking, logistics, messaging or even e-commerce. The broad ambition is to own one high-frequency relationship, then stretch it into many others.

There’s nothing better than a look at some of Nigeria’s super app dreams to understand where the next thing is.

GTCO (Habari)

GTBank launched Habari in 2018 as Nigeria’s largest platform for music, shopping, lifestyle content, and more. Segun Agbaje said the app was “less about us as a bank and more about our customers and everything they need to enable their lifestyle.”

GTBank had one of the strongest consumer banking brands in the country, and the thinking was that if fintechs were going to attack the bank from the customer experience point, GTBank could play that game too.

A customer can trust you with their salary and still not want it to be their music app, shopping mall, and cultural companion.

OPay (ORide, OFood, OEverything)

OPay had ORide, OCar, OBus, OFood, OExpress, OMall, OTrade, OKash, and the wallet at various points. OPay was building a Nigerian super app with restaurant delivery, logistics, B2C and B2B commerce, and ride-hailing, all centered around the payment product.

When Lagos restricted motorcycle ride-hailing, OPay got smaller, and well, it’s on its way to an IPO, so we know how that went.

Jumia (Marketplace)

If any African internet company could argue that one activity would reinforce the others, it was Jumia. You could buy an item, pay with JumiaPay, order food, and at some points, book hotels or even find flights.

Yet, all the verticals were spun off until we ended up with this new lean version of Jumia, which has exited, cut, and narrowed its way to this entity that now expects to break even in 2027.

MTN (Ayoba)

MTN had SIM cards, airtime, billing relationships, data, agents, and presence across African markets. It wanted to be an African answer to WhatsApp and WeChat.

Free data can make people try an app, but can it make people move their church groups, school chats, family arguments, vendor relationships, office gossip, and football banter from WhatsApp? Ayoba has now shut down.

Swoop

Swoop is a food delivery business that does not see itself as a food delivery business. The founder, Niederhoffer, references Kaspi in Kazakhstan and WeChat in China, platforms that turned a single high-frequency consumer behaviour into the rails for everything else. Food delivery is the start, with payments coming next.

“In Africa, there’s no legacy banking infrastructure. Essentially, you’re not competing with credit cards. Those are not popular, and there’s a huge opportunity.” - Niederhoffer

I don’t fault whatever you have to say to raise money.

Each of the four companies above with superapp ambitions started from a stronger position than food delivery.

Swoop’s thesis is that the destination (a payments and financial services layer) is a vacuum waiting to be filled. Of course, startups change their destination and pathways all the time, but this is certainly worth watching.

In any case, Swoop has $7.3 million and a really good story. Game on.

Share

See you next week!

If this newsletter was shared with you, subscribe for free:

Subscribe now

TOGETHER WITH CREDIT DIRECT

MoMo’s promise vs hard reality

When ideas fail or companies struggle, it’s very easy to say, “Ah, I called it, always knew that wouldn’t work!” Yet some of those sentiments may not have been accurate when the product or company launched.

When MTN launched its mobile money service MoMo, for instance, the consensus was that it would crush existing fintechs and be a threat to banks. The believers weren’t crazy, although their frames of reference were mostly M-PESA or Mobile Money in Ghana.

The broad theory of MoMo’s inevitable domination was that with MTN’s subscribers and retail touchpoints, it would be easy to spread the gospel of their wallet. It had worked in other African countries; at the end of 2021, MTN Mobile Money was operational in about 15 countries, with 56.8 million customers completing 10 billion transactions worth roughly $240 billion.

In 2019, GTBank called telco-led payment services “a more compelling threat.” Nairametrics ran with the headline “Mobile money war: Telecoms threaten banks’ future in Nigeria” and argued that it would be unwise to bet against telcos.

By the time MTN and Airtel received approval in principle for mobile money in 2021, TechCabal argued that telcos were even better positioned than banks and fintechs to deepen mobile money penetration given their subscriber base, infrastructure, and agent networks.

MoMo Payment Service Bank (PSB) launched with a promise that customers could dial *671# on any network to open a wallet and send money to any phone number in Nigeria. It also started with more than 166,000 active agents and promised to reach unbanked and underserved Nigerians. Four years on, that promise is far from fulfilled.

This week, Techpoint reported that MTN Group plans to acquire a 60% stake in MoMo Payment Service Bank and Y’ello Digital Financial Services for a combined ₦95.5 billion. “It’s not just a sale; it’s a restructuring designed to shift control of these fintech units up to the group level while easing pressure on MTN Nigeria’s books,” Techpoint wrote.

The transaction will bring a ₦152.06 billion cash injection into the fintech subsidiaries while valuing them at ₦95.5 billion, more than twice their December 2025 carrying value. MTN Nigeria is moving more of the fintech risk, funding, and upside to MTN Group, while keeping a minority stake in case those subsidiaries improve.

A few weeks ago, we looked at MTN Nigeria’s FY 2025 numbers and saw MTN had written down the value of its fintech subsidiaries by ₦62.56 billion. MoMo had 3.7 million active wallets at the end of 2025, miles from its 30 million to 40 million target. The company had discussed “Fintech Structural Separation” and “Fintech Capital Injection” and even held an ad hoc committee meeting on “Fintech Strategy.”

As I wrote then, those were not “the agenda items of a business gliding elegantly toward destiny.”

Whatever happened with MoMo, given that its destiny seemed to be that of an inevitable segment winner?

The problem starts with the market MoMo entered. Instant payments in Nigeria were already normal, and NIBSS Instant Payment transactions had reached 5.2 billion in 2022.

In some markets, the telco wallet became the main rail. In Nigeria, NIBSS, banks, POS agents, and fintech wallets had already done much of that work. So MoMo was entering a market where money movement was commoditised, so what was its edge going to be?

It was certainly not the price. MoMo’s flat ₦10 transfer fee was only competitive for amounts above ₦5,000, meaning it was not enough to change behaviour in a market where many everyday payments were small, frequent, and already moving through bank apps, POS agents, OPay and PalmPay.

A payments product needs a wedge. It must make money movement radically easier, be accepted in more places, be trusted more than alternatives, be cheaper or more reliable, give users a second product they badly want, or own a high-frequency habit before asking people to change financial behaviour.

MTN employs a lot of smart people who understand the constraints. In 2023, Eli Hini, then MoMo PSB CEO, explained the fintech’s strategy as using MTN’s “wide distribution network” and the mobile phone as the key channel, whether a basic feature phone or smartphone. In another interview, he suggested that the idea was to take money from under people’s pillows and beds and bring it into the financial ecosystem, a.k.a financial inclusion.

Compare that to what the fintechs it was supposed to dominate were doing.

OPay and PalmPay focused on being physically present when banks failed and ATMs ran out of cash. During the 2023 cash crunch, Semafor reported that vendors and customers turned to OPay and PalmPay because traditional banks could not handle the surge in online transactions. By March 24, 2023, both apps were ranked first and second among Nigeria’s most downloaded finance apps on Google Play.

In hindsight, there was still a huge cash-to-digital opportunity in Nigeria, but the fight was about who could make a market woman, student, dispatch rider, and salary earner trust that the wallet/product would work right now.

OPay answered that question with free transfers, ridiculous subsidies, agents, aggressive customer support, a consumer app that felt lighter than most bank apps, and the good fortune of a cash crunch that made everyone reconsider their banking habits. PalmPay was bundled on Transsion phones even before you made a payment choice.

Moniepoint focused on merchants and in-person payments, then expanded from the business owner to the customer. It now claims to process more than 800 million transactions monthly, with over $17 billion in monthly value, and that two out of three adults in Nigeria make payments through a Moniepoint terminal.

MoMo’s MTN Problem

A telco is excellent at reach, billing, SIM registration, airtime, USSD, and regulatory negotiation. But consumer finance is not only reach; it is product obsession, fraud control, merchant acceptance, customer service, pricing discipline, incentives, dispute resolution, and thousands of tiny moments where users decide whether to keep trusting you.

Then there is the Payment Service Bank licence that can accept deposits, hold wallets, provide payments, issue cards, run agents, and offer remittances. But it cannot lend or accept foreign currency deposits.

That matters because payments alone is a marginal business. Payments will attract users, but credit, merchant tools, working capital, savings yield, and commerce deepen the relationship.

A wallet that only moves money is useful, but in Nigeria, money already moves, so your wallet must either be more reliable than the banks, more present than cash, more rewarding than alternatives, or attached to another product people urgently want.

Again, MTN has smart people, and they also know this. Its executives talked about financial education, mobile device financing, SME exposure, online payments, remittances, and collaboration with banks and fintechs.

“Agents determine the fees they want to charge, so fees are not constant or controlled. That also doesn’t help customer adoption.” - Eli Hini (2023)

Again, what is MoMo’s sharp wedge? If it’s rural inclusion, why does it look like many PSBs are focusing on cities and competing for already-banked customers?

If it is agents, then why has the agent network shrunk dramatically?

If it were merchants, why have OPay, PalmPay, Moniepoint, and the banks become the names people associate with everyday payments?

MoMo had 3.5 million active wallets by Q1 2023, according to Hini. By the end of 2023, TechCabal reported that active wallets had risen to 5.3 million, supported by an agent network that had grown from 103,000 to 327,000 and a merchant ecosystem that had reached 324,000 merchants. By 2024, those numbers declined.

MTN’s interpretation was that it was improving the quality of the ecosystem, and that is possible. Not all wallets are equal, and not all agents are useful; some customers exist only because of incentives, and some agents transact only because commissions exist, so cleaning that up can make the business healthier.

But when a fintech business loses that much surface area in a market where competitors are becoming more visible, they have to prove that what remains is a smaller, better machine.

We can’t talk about MoMo without that security incident that cost the company roughly ₦10 billion. Hini called it a technical glitch, said no customers lost money, and that MoMo had reviewed protocols and improved controls around integrations and partnerships. For a payments business trying to build trust in its first year, a major loss involving integrations impacts product confidence.

MoMo came with a distribution thesis into a market where distribution had already become a knife fight. It expected its reach to do more of the work than it could with a constrained licence in a market where the most valuable fintech loops required either lending, merchant economics, stronger consumer incentives, or brutal last-mile execution. It launched into a Nigeria where instant payments were already massive, bank transfers were already normal, and the remaining prize was not “teach Nigerians that digital money exists.” The prize was “be the financial product they trust when cash fails, banks break, and every failed transfer has immediate consequences.”

OPay and PalmPay met that moment from the consumer side. Moniepoint met it from the merchant side. The banks, for all their failures, still held salaries, corporate accounts, cards, credit relationships, and decades of customer inertia. MoMo had MTN’s brand and distribution, but it was never really clear why a Nigerian who already had a bank app, OPay, PalmPay, a POS agent nearby, and a Moniepoint terminal in front of them needed MoMo as their primary wallet.

Anyway, as it stands, MTN Nigeria has a telecom business to protect, especially after a brutal 2024 and a much better 2025. Moving MoMo and Y’ello Digital into a better-capitalised HoldCo could reduce pressure on MTN Nigeria’s books, bring fintech expertise closer to the asset, and create room for partnerships or future regulatory changes. MTN Group has the broader fintech ambition, the continental playbook, and more reasons to keep funding the experiment.

Yet a separate MoMo still has to answer the same old question. What does it do that Nigerians do not already get from OPay, PalmPay, Moniepoint, their bank, or the POS agent beside their street?

P:S: Will OPay IPO before the Butterfly Fintech? We’ll talk about it only on Sunday.

See you on Sunday!