M-Pesa was launched in 2007 by Safaricom as an SMS-based money transfer service, aimed at Kenya's unbanked majority. It succeeded because of an improbable combination: a dominant telco, a flexible central bank, a massive physical agent network and a brutally simple use case ("send money home"). Today, over half of Kenya's GDP flows through M-Pesa, and the model has been replicated in seven countries. It is the largest financial inclusion case ever built.
What is M-Pesa, exactly?
M-Pesa (from "M" for mobile and "Pesa", which means "money" in Swahili) is a money transfer and storage service operated through mobile phones. It was launched in March 2007 by Safaricom, Kenya's largest telecommunications operator, in partnership with UK-based Vodafone.
The mechanics are simple: a user walks into a registered agent (a corner shop, a pharmacy, a gas station), hands over cash, and receives the equivalent as "M-Pesa credit" on their mobile phone. From there, they can send that balance via SMS to any other phone number, pay bills, receive salaries, or cash out at another agent. No bank account required. No smartphone either — M-Pesa works on any phone, including the most basic models, through USSD menus.
In practice, M-Pesa turned the mobile phone into a digital wallet accessible to anyone with an ID. In a country where, in 2006, fewer than 20% of adults had a bank account, that was revolutionary.
The context: why Kenya, and why 2007?
To understand the magnitude of what happened, you need to understand Kenya in 2006.
The country had about 37 million people, of whom nearly 80% lived outside major urban centers. The economy was largely informal. Banking infrastructure was thin: there were roughly 450 bank branches in the entire country, concentrated in Nairobi and Mombasa. For someone living upcountry, going to a bank could mean a full day of travel.
At the same time, something peculiar was happening. Safaricom, originally a subsidiary of Kenya Posts and Telecommunications and later partially acquired by Vodafone, held roughly 80% of the mobile phone market. Kenyans had adopted mobile phones before they had adopted credit cards, checking accounts, or even reliable electricity at home.
And there was a behavior that Safaricom executives noticed: people were using airtime credit as currency. A worker in Nairobi would buy airtime credit, transfer it to their father's phone upcountry, and the father would sell the airtime to a neighbor for cash. It was a workaround, but it worked.
The pattern was clear: there was massive pent-up demand for a way to send money from one place to another. The question was: who would solve it — banks or telcos?
The 4 decisions that defined M-Pesa
There are many ways to tell the M-Pesa story. The way I find most useful — and it's how we structure it in the SMX simulator — is as a sequence of four strategic decisions, each with plausible alternatives Safaricom could have taken. In each one, the "obvious" choice differed from what they actually did.
Telco or bank as the protagonist?
When the UK's Department for International Development (DFID) funded the initial research in 2003, the original idea was to use the mobile phone as a channel for microcredit. The telco would just be the "pipe" — the operator of the service would be a local microfinance bank. That's the "safe" model: banks understand money, telcos understand networks.
Safaricom took the counterintuitive decision to be the financial operator itself. This meant dealing with compliance, security, anti-money-laundering, KYC — everything that banks have historically done better than telcos. In return, Safaricom kept full control of the customer experience, pricing, and — crucially — the relationship with the end user.
This decision matters because in nearly every country where mobile money was tried with banks as the protagonists (the Philippines, India, several African markets), the product failed to scale. Banks treated the mobile channel as "just another product"; telcos treated it as the product. That focus difference translates into investment, marketing, and iteration.
Microcredit or pure transfer?
The original M-Pesa business plan in 2005 was sophisticated: users would take out microloans, receive funds on their phone, and repay installments the same way. It was a full financial product, with credit scoring, interest rates, and delinquency management.
In the pilots, something unexpected happened: users started using the system to send money to relatives in other cities. The transfer functionality had been conceived as a means to move loan funds — but it turned out to be what people used most, even without ever taking a loan.
Safaricom did something few companies manage to do in that moment: they listened to the product. They killed the microcredit layer, killed the risk scoring, killed the complicated parts. They relaunched in 2007 with a brutally simple proposition: "Send Money Home." Three words. One use case. Zero interest. Zero complexity.
The effect was devastating. Within 18 months, M-Pesa had 5 million users. Microcredit did come later, in 2012, with the M-Shwari product — but only after the habit of using M-Pesa for other things was already entrenched.
Cash-in/cash-out: own network or partners?
A digital money service only works if users can, on one end, put physical cash into the system, and on the other end, take physical cash out. That's the "last mile" of mobile money, and it's where most projects die.
Safaricom could have built its own M-Pesa branches — expensive, slow, but controlled. It could have used ATMs and banks — fast, cheap, but only serving those near existing branches. Or it could create a network of third-party agents, small merchants who would earn a commission on every deposit and withdrawal.
They chose the third option — and pushed it to the extreme. In 2007 they launched with 750 agents. By 2010, they had 20,000. By 2020, more than 250,000 agents — many more than the combined bank branches of all East African banks.
The design matters. M-Pesa isn't dominant because it has the best app, the best interface, or the lowest prices. It's dominant because, in any Kenyan village, there is a merchant wearing a green Safaricom t-shirt who can hand you your cash in three minutes. It's a logistics rocket dressed as a fintech.
Regulation: wait for approval or launch and negotiate?
This is where the story gets interesting, and where most M-Pesa imitators in other countries failed. In 2006, there was no specific regulation for mobile money in Kenya. Technically, what Safaricom was proposing — collecting deposits, managing balances, operating a payment system — was banking activity. And Safaricom wasn't a bank.
Safaricom could have waited for regulation to arrive (which, on average, takes a decade in emerging economies). It could have lobbied for legislation (slow and expensive). It could have given up and left the market to the banks.
Instead, it opened a direct dialogue with Kenya's Central Bank, proposing a specific model: customer balances would be 100% protected in escrow accounts at commercial bank partners, Safaricom would not engage in financial intermediation (no lending, no deposit-taking), and the service would be audited. The Central Bank agreed to allow the launch under a "money remittance" classification with close monitoring.
That regulatory flexibility — often called a "regulatory sandbox" avant la lettre — was decisive. When Nigeria, India and several other countries tried to replicate the model with more cautious regulators, the product never reached Kenya's scale.
What M-Pesa has become
The current numbers are hard to process without losing sense of scale.
Over 50% of Kenya's GDP is transacted through M-Pesa every year. Salaries are paid via M-Pesa. Electricity, water and cable bills are paid via M-Pesa. Bank loans are disbursed via M-Pesa. Taxis in Nairobi are paid via M-Pesa. Even donations in rural evangelical churches happen via M-Pesa.
The economic impact is even more interesting. An MIT study published in 2016 (Suri & Jack, "The long-run poverty and gender impacts of mobile money") estimated that M-Pesa lifted roughly 194,000 Kenyan households out of extreme poverty — an effect especially strong in female-headed homes, where women gained autonomy over family finances. It's one of the few documented cases where a private technology product had measurable, population-scale impact on poverty reduction.
Today M-Pesa is an ecosystem: it offers savings (M-Shwari), credit (Fuliza), insurance, investments, e-commerce integration (Lipa na M-Pesa) and even APIs for developers to build applications on top of the system. It is, essentially, the parallel financial infrastructure of a country.
Want to live the decision, not just read about it?
At SMX, you step into the CEO's chair at Safaricom in 2005 and make the four decisions in this case in 40 minutes. Strategic feedback, comparison with the actual decision and certificate. It's free.
Try the M-Pesa case in the simulator →3 lessons that travel beyond Kenya
M-Pesa is tempting to read as a story about a specific emerging market. It isn't. The principles underlying the four decisions show up, with different costumes, in every major financial inclusion success of the last two decades — from Brazil's PIX to India's UPI to Alipay in China.
1. Brutal simplicity beats sophistication
M-Pesa's original plan was complex — and it pivoted to a three-word proposition. Nubank in Brazil started with a single product (a purple no-fee credit card) and refused, for years, to become a "full bank." PIX launched with one feature only: instant, free, 24/7 transfers. Financial products scale when you remove options, not when you add them.
2. Physical infrastructure matters as much as software
M-Pesa won because it had 250,000 physical agents on every corner. PIX won because Brazil's Central Bank built a real-time public settlement infrastructure. Fintechs that only have an app and a banking-as-a-service layer are playing on hard mode — the differentiation comes from whoever controls deeper layers of the financial stack.
3. Regulatory flexibility is structural
Kenya's Central Bank said "yes, let's try" in 2007. Brazil's Central Bank built PIX, opened Open Finance, created the regulatory sandbox. It is no coincidence that the two countries most innovative in financial inclusion have regulators who see themselves as partners of innovation, not just referees of it. Entrepreneurs in jurisdictions where the regulator is defensive rarely build something resembling M-Pesa or PIX.
Conclusion: what M-Pesa teaches about strategy
If you read the classical competitive strategy books — Porter, the Five Forces, SWOT analysis — none of them cleanly predict a case like M-Pesa. That's not a failure of theory. It's a reminder that strategy is, in practice, a sequence of decisions made under uncertainty, where the "right" answer only becomes obvious in retrospect.
The real value of studying cases like M-Pesa isn't memorizing what Safaricom did. It's training the muscle of making real strategic decisions — with incomplete information, reluctant counterparts, brutal trade-offs, and windows of opportunity that close. You don't become a good CEO by reading about CEOs. You become a good CEO by making hard decisions and receiving honest feedback on them.
That's exactly what we've built at SMX.