EN
Back to the archive

The archive · Business Models · Operational decision · 2005–2007

M-PESA makes Kenya's corner shops cash agents — the airtime network becomes a bank

Safaricom launched M-PESA in March 2007 by turning airtime resellers into cash-in/cash-out agents, so every corner shop became a bank teller for phone money.

Safaricom · Vodafone

The ideaReuse the airtime-reseller network as cash-in/cash-out agents so every corner shop becomes a bank teller for phone-based money.transformative

What it had to solve

Kenyan city workers earned cash they needed to get home, but the only routes — buses, post offices, money orders — were slow, risky and often out of reach for rural families. Banks served few; phones reached everyone.

How it works

In 2005, Vodafone's Nick Hughes proposed using the DFID-funded Financial Deepening Challenge Fund to test mobile payments for microfinance in East Africa. With Safaricom, they piloted loan repayment by phone with Faulu Kenya's customers in Thika and Mathare — but watchers noticed borrowers using the phone wallet as a safe way to carry cash from town to the countryside. The real problem was not loans; it was getting money home.

The decisive move was distribution: instead of building bank-like branches, the team used Safaricom's existing airtime-reseller network as cash-in/cash-out agents. SIM toolkit and SMS meant any basic phone worked, and agents were paid a fee per cash exchange, so a kiosk owner's income grew with every transaction. Safaricom paid dealers $1.50 per new subscriber, and an aggregator model meant a few thousand head offices could manage hundreds of thousands of agents.

M-PESA launched in March 2007 as 'Send Money Home'. It passed 2 million customers in under a year — after CEO Michael Joseph threw out the 300,000-user plan and demanded a million — and reached 1,041,522 active users by November 2007. By 2015 it was used by more than two-thirds of Kenyan adults through over 85,000 agents processing roughly 8 million transactions a day.

Why it lands

  • Corner shops already handled cash and were trusted locally, so they could cash money in and out without new infrastructure.
  • Agents were paid per exchange, aligning their income with usage — the network grew because every transaction paid its operator.
  • Airtime dealers already stocked Safaricom's products, so recruitment, training and trust came almost free compared with a branch rollout.
  • SIM toolkit and SMS worked on basic phones, so the channel reached people banks never could.

What it did

Reaching 2 million customers in under a year — and 1,041,522 active users by November 2007, after scrapping the 300,000 target — M-PESA grew to two-thirds of Kenyan adults, 85,000+ agents and 8 million transactions a day (~$20M) by 2015.

Write-upThe Making of M-PESA

What you can take

The distribution you already own is the moat: airtime resellers became M-PESA's branch network, so the service scaled to millions without building a single bank branch.

Since then

M-PESA stopped being just money transfer: it became the rails for bill payments, point-of-sale purchases and bank transfers, and enabled M-Shwari mobile banking, M-KOPA solar payments and government social-protection transfers. The agent network built in 2007 still carries the service — by 2015 more than two-thirds of Kenyan adults used it through over 85,000 agents processing about 8 million transactions a day, and the model was exported across Africa and beyond.

Sources

spotted an error? The archive wants to know.

Your turn

You just read one. Describe the brief you are staring at, and see who has been given the same problem.

Free account · 3 free questions · no card

Related cases