The Safaricom Story: Government monopoly to most innovative telco in the world
On the sixth of March, 2007, in a small office in Nairobi, a Kenyan telecoms company launched a product that nobody else in the world thought would work.
It was called M-Pesa. It let you send money — actual money — by SMS. The banking regulator did not know what to call it. The telecoms regulator did not know who should regulate it. Vodafone, which co-owned the company, had tried versions of the same idea in other markets and watched them fail.
In the first month, M-Pesa had nineteen thousand, six hundred and seventy-one active users.
Within a decade, it was moving more money than Western Union does worldwide. Today, it processes more than forty-one trillion Kenyan shillings a year — about three hundred and twenty billion US dollars.
This is the story of how a state telephone department with twelve thousand customers became the company that taught the world how to bank a continent.
Five Men and a Working Capital Cheque
To understand Safaricom, you have to begin where it began. Not as a company. As a department.
In 1993, the Kenya Posts and Telecommunications Corporation — the state monopoly — switched on an experimental cellular network. The technology was a dying analogue standard called ETACS. The phones were the size of bricks. The customers were not customers in any meaningful sense; they were a few thousand handsets in the pockets of Nairobi’s political and business class. If you wanted a phone line in Kenya in 1993, you joined a waiting list. The waiting list moved in years.
In 1997, the department was incorporated as a company. They called it Safaricom. The name was deliberate — Safari, the one Kenyan word the world already knew, and Com for communications. The branding was green and white, designed to feel like national infrastructure rather than a foreign import. Safaricom Limited was a subsidiary of Telkom Kenya, which was itself being carved out of the soon-to-be-dismantled state monopoly. It inherited about twelve thousand analogue subscribers and a network that was already obsolete.
For two years, that was the entire story. A small unit. An old standard. Twelve thousand customers in a country of thirty million.
In May of 2000, the second moment that built this company happened in London.
Vodafone Group PLC paid roughly forty-two million US dollars for a forty percent stake in Safaricom, and took management control. Without that cheque, this episode does not exist. The company stays a department. The story ends.
The next month, a fifty-four-year-old South African telecoms engineer named Michael Joseph boarded a flight to Nairobi with a team of five Vodafone employees. They walked into the building on Waiyaki Way, inherited seventeen thousand subscribers and about twenty million dollars in working capital, and got to work.
Joseph was not from Kenya. He was a University of Cape Town engineer with a US passport and deployment scars from emerging markets across three continents. His reputation, even then, was for being blunt to the point of difficulty. The line he is most associated with — “I do not suffer fools gladly” — is something he has said about himself, repeatedly, on the record.
On the twentieth of October, 2000, Safaricom officially relaunched on a new 2G GSM network. They invested three point seven billion Kenyan shillings in initial infrastructure. They competed against the Telkom Kenya monopoly. They competed against KenCell, the second-licence operator, which had foreign money behind it and a head start.
The bet Michael Joseph had been hired to make was simple, and it was not, at the time, obvious. The conventional wisdom of African telecom in the year 2000 was that mobile phones were a product for the African middle class. A luxury good. Joseph was about to argue the opposite, and stake the company on it.
The Twenty-Shilling Bet
In 2000, a SIM card and a basic handset in Kenya cost about fifty-four thousand shillings. At the exchange rate of the time, that was around seven hundred US dollars — multiples of what most Kenyans earned in a year. The cheapest way to make a phone call was to walk to a public phone booth and feed it coins.
Every telecom executive in the region looked at those numbers and reached the same conclusion. Mobile was for the rich. The mass market was unreachable. The right strategy was to harvest the top of the pyramid and wait for prices to fall.
Michael Joseph reached the opposite conclusion.
His bet was that ordinary Kenyans — written off by the industry as too poor to afford mobile — would pay for connection if you priced it the way they already paid for everything else in their lives. In small amounts. Cash. Today.
Safaricom built a prepaid airtime system around scratch cards in tiny denominations. Five hundred shillings. Two hundred and fifty. One hundred. Fifty. Eventually twenty shillings — about a quarter of one US dollar. They introduced per-second billing so that a forty-second call did not get charged as a full minute. They built distribution into every duka, every kiosk, every petrol station, until buying airtime in Kenya was as routine as buying bread.
What happened over the next four years is the closest thing in modern African business to a stampede.
The customer base went from seventeen thousand at the relaunch to several million by 2004. The Safaricom green-and-white became one of the most visible brands in the country — on the side of every kiosk, painted on the walls of village shops, flying behind every matatu. The company did not just sell mobile service. It built the retail layer that would later carry M-Pesa, M-Shwari, Lipa Na, Fuliza — every product that would come after, distributed through the same hundreds of thousands of small Kenyan businesses Safaricom had spent four years signing up.
Inside Safaricom, the culture matched the man at the top. Joseph ran a performance regime that people who worked under him still describe in the same two words — high pressure. Targets were aggressive. Excuses were rare. The reputation Safaricom developed for being abrasive in the market — pushing distributors hard, squeezing rivals, treating regulators as obstacles to manage — came from this period. It is also, honestly, part of why the company executed at the speed it did.
In June of 2004, the Telkom Kenya monopoly formally ended. The competitive licensing framework that had been promised for years finally arrived. KenCell — by then renamed Celtel, later sold to Zain, eventually rebranded as Airtel Kenya — was supposed to be the rival that disciplined Safaricom. It never did. Through every ownership change and every relaunch, Airtel never closed the gap. Safaricom out-invested it on network coverage and out-distributed it on retail.
By 2006, Safaricom was profitable, dominant, and growing. It was, by any reasonable measure, a successful African telecom.
It was about to become something for which there was no category at all.
The Hack That Became the Bank
The story of M-Pesa does not start in Kenya. It starts in London, in 2003, in the corporate social responsibility department of Vodafone.
A CSR executive named Nick Hughes had been thinking about something for a while. Microfinance institutions in developing countries were lending small amounts of money to very large numbers of very poor borrowers, and the single most expensive part of that business was not the lending. It was the repayment. Loan officers riding motorcycles into villages to collect installments in cash. Inefficient. Dangerous. Slow.
Hughes applied for a matching grant from the UK government’s Department for International Development. He won it — a million British pounds, matched by Vodafone, to design a mobile phone system that would let microfinance borrowers repay their loans by SMS.
Vodafone seconded a product manager named Susie Lonie to Nairobi to lead the pilot. The partner was a Kenyan microfinance institution called Faulu Kenya. The location was Thika — a small industrial town an hour north of Nairobi. The product was a tightly scoped, single-purpose tool for loan repayment.
Within weeks of the pilot going live, the pilot users were doing something the product had not been built for.
They were sending the money to relatives. They were paying for goods. They were using it to settle small business debts. The microfinance use case was secondary. The use case the customers had invented for themselves was that this was a way to move money — any money, for any reason, to anyone on the network — without going to a bank, without writing a cheque, without queuing.
The product team did the thing that all good product teams do when their users tell them what the product actually is. They listened. They redesigned. The pilot was no longer about loans. It was about money.
The decision to launch was not automatic. Inside Safaricom, there was real internal scepticism. The argument against was straightforward — a mobile-money product would cannibalize voice and SMS revenue, which were the company’s profit engines. People were going to talk less and text less if they were using their phone for money instead. The numbers did not, at first, justify the risk.
Michael Joseph overruled it.
On the sixth of March, 2007, M-Pesa launched commercially. One product. One feature. Send Money. The marketing line was as direct as a slogan can be — Send Money Home — aimed at the millions of Kenyans who worked in Nairobi or Mombasa or the export-flower towns of the Rift Valley and supported families they had not lived with in years.
Nineteen thousand, six hundred and seventy-one Kenyans signed up in the first month.
By November of the same year, a million had.
The bet had been that Kenyans, written off by the financial industry as too poor to be banked, would pay for the ability to move money in small amounts if you priced it the way they already paid for airtime. In small amounts. Cash. Today.
It was the same bet Michael Joseph had made on prepaid airtime in the year 2000, scaled to a different product. He was right again, by an order of magnitude.
From Telco to Operating System
In March and April of 2008, the Government of Kenya sold twenty-five percent of its sixty-percent stake in Safaricom to the public. Ten billion shares at five Kenyan shillings each. The IPO raised about fifty billion shillings, around seven hundred and ninety-two million US dollars. It was the largest privatisation in East African history.
Roughly eight hundred and seventy thousand new retail investors entered the Nairobi Stock Exchange to buy Safaricom shares. For most of them, it was the first listed equity they had ever owned. When the company crossed a market capitalisation of one trillion shillings later that year, it was the first Kenyan company ever to do so.
M-Pesa, meanwhile, was doing something no product in financial history had done before.
By 2010, the network of M-Pesa agents — the duka owners, the petrol attendants, the small shopkeepers licensed to accept cash deposits and dispense withdrawals — outnumbered every ATM, every bank branch, and every post office in Kenya. Combined. The most accessible financial counter in the country was no longer a bank. It was a corner shop.
What followed was a layered build. Not a product. A platform.
In November 2012, Safaricom launched M-Shwari in partnership with the Commercial Bank of Africa — a fully mobile savings and micro-lending account, with credit decisions made by algorithm in seconds, on transaction history that Safaricom already had. In June 2013, it launched Lipa Na M-Pesa, the merchant-payment rails that turned every duka into a payment terminal. In 2019, it launched Fuliza — the world’s first contextual mobile-money overdraft, extending credit at the moment of payment to anyone whose M-Pesa balance fell short.
By the mid-2010s, M-Pesa was moving the equivalent of roughly half of Kenya’s annual GDP through its rails.
The cultural change inside the company in this period had a name. Bob Collymore.
Collymore was a Guyanese-born, British-raised telecoms executive who took over from Michael Joseph in November 2010. He inherited about seventeen million subscribers. By his last full year, the base was over thirty million. The share price, over his tenure, roughly quadrupled.
Where Joseph had been blunt and high-pressure, Collymore was quiet, values-driven, and unusually public about corporate governance, transparency, and the role of business in African civil life. He became, in the years before his death, one of the most respected African chief executives of his generation.
He was diagnosed with acute myeloid leukaemia in 2017. He sought treatment in the United Kingdom, returned home to Kenya in 2018 to continue working while ill, and died at home on the first of July, 2019. He was sixty-one. The company he handed over was no longer a telco. It was, in everything but legal form, the country’s largest financial institution.
The State Midwife Steps Back
On the first of April, 2020, Peter Ndegwa became the first Kenyan chief executive in Safaricom’s history.
Ndegwa was not a telecoms person. He was a Diageo lifer who had spent two decades selling beer across East and West Africa and Western Europe. The product he is most associated with — Senator Keg, a low-cost sorghum lager designed to compete with informal alcohol — became one of Diageo’s most successful global innovations. He understood mass-market Africa. He had not, until that morning, run a network.
The two largest bets of his tenure are the two largest bets in the company’s history.
The first is Ethiopia. In May 2021, a Safaricom-led consortium won the unified telecoms licence for Ethiopia for eight hundred and fifty million US dollars — the largest foreign direct investment deal in the country’s history. Safaricom Ethiopia launched commercial services in October 2022, M-Pesa Ethiopia in August 2023. By May 2025, the Ethiopian business had crossed ten million subscribers. The target for 2030 is seventy million. The investment so far is over two billion US dollars. Profitability is projected for the financial year ending 2027.
The second is harder to summarise.
In June 2024, mass protests against a finance bill swept Kenyan cities. Several perceived protest leaders were abducted. Civil society organisations accused Safaricom of sharing subscriber data with security agencies. Safaricom — and Peter Ndegwa personally — publicly denied the allegations. In 2026, the Law Society of Kenya filed a constitutional petition at the High Court alleging that investigators had routinely accessed subscriber data without judicial authorisation. The litigation is ongoing. The trust gap with younger Kenyans — who are also M-Pesa’s most engaged future cohort — is a strategic risk the company has acknowledged publicly.
In December 2025, the Government of Kenya sold an additional fifteen percent of its Safaricom shareholding to Vodacom for roughly two hundred and forty-four billion shillings. Vodacom consolidated to fifty-five percent effective control. The state, which had midwifed Safaricom into existence twenty-five years earlier, stepped back to a twenty percent minority — retaining two board seats, the right to consult on offshore expansion, and a contractual requirement that the chief executive be Kenyan.
The company that closed the financial year ending March 2025 had over three billion US dollars in annual revenue, fifty-seven million customers across two countries, and an M-Pesa platform processing forty-one point six eight trillion shillings a year.
The product Vodafone could never replicate anywhere else in the world is now the operating system of a country.
Every empire has an origin.
Safaricom’s was a state telephone department with twelve thousand customers, a dying analogue network, and a Vodafone cheque. Five people in a Nairobi office. A bet that the Kenyans every other operator had written off would pay for connection if it was priced right.
They were right.
And then they paid for everything else through it.
This is Asili Africa.
Key Takeaways
- Five Men and a Working Capital Cheque. To understand Safaricom, you have to begin where it began.
- The Hack That Became the Bank. The story of M-Pesa does not start in Kenya.
- From Telco to Operating System. In March and April of 2008, the Government of Kenya sold twenty-five percent of its sixty-percent stake in Safaricom to the public.
- The State Midwife Steps Back. On the first of April, 2020, Peter Ndegwa became the first Kenyan chief executive in Safaricom’s history.
Also available on YouTube — search “Asili Africa” or subscribe to our channel.