To Michael Joseph’s frustration, his customers keep flashing. At all times of the day. For no more than a second but long enough to catch each other’s attention. In Kenya, where his company Safaricom is the dominant mobile phone operator, it has nothing to do with indecent exposure.

To flash is to call a mobile and hang up before the call is answered, a cost-free way of letting the owner know you want to be called back. People do it because they are low on pre-paid credit, or because they think the other person has a better reason to pay for the conversation.

It is a habit borne of poverty in an African country where gross national income per capita is $530 a year and 46 per cent of its 36m people live on less than a dollar a day. But flashing, which congests the network, bothers Mr Joseph so much that, for no charge, customers can now send a standardised text message that reads: “Please call me. Thank you.”

“It gets people off our network to allow other people to make calls that will mean revenue for us,” says the chief executive, who was parachuted into Safaricom by Vodafone, which owns 35 per cent of the company via an entity called Vodafone Kenya.

Flashing is one irritant in the Kenyan mobile market. But cannily creating products specifically for low-income customers has helped Mr Joseph post the biggest profit this year of any company in east Africa.

Much has been made of mobile phones transforming lives in Africa: they connect distant relatives, save arduous bus journeys on dreadful roads and help casual labourers find out where to go for work.

But what Safaricom proves is that delivering social change can be a lucrative business – even for a group that has a reputation for high pricing, network congestion and ponderous customer service.

The government is preparing to float 25 per cent of the company on the Nairobi stock market – before the end of the year if possible, but more likely in 2008 – and Safaricom set the stage for the initial public offering in June by posting a record-breaking pre-tax profit of KSh17bn ($270m) on revenue of KSh47bn for the year to March.

Safaricom’s profits were up 41 per cent from the previous year whereas Vodafone’s group profits – earnings before interest, tax, depreciation and amortisation of £12bn – edged up just 0.2 per cent , a comparison that underscores why the mobile giant has seized on emerging markets to offset slowing growth in its core European business.

Vodafone bought its Safaricom stake in 2000 and dispatched Mr Joseph to Kenya from Hungary. He arrived in Nairobi to find a poorly performing business with about 20,000 customers.

The group’s only competitor, now called Celtel and owned by Kuwait’s MTC, had about 85,000 customers and a brand new network by the end of 2000. But with a $20m investment from Vodafone and more borrowed money, Mr Joseph relaunched Safaricom and a year later had won more customers than his rival.

Today, with roughly one in four Kenyans owning a mobile, nearly 8m use Safaricom and about 3m are on Celtel. Both companies have benefited from state neglect of the country’s fixed-line infrastructure, which does not reach many rural areas and is unreliable in the cities, creating a captive market for mobile services.

What differentiates Safaricom, according to Mr Joseph, is its targeting of low-income workers in the informal sector, people who sell fruit, fix cars, or make tools: “Not the guy with the credit card and the cheque book.”

The company focused on pay-as-you-go tariffs, which are now used by 90 per cent of its subscribers, and opted to bill them per second. Recognising that many Kenyans prefer to buy things in small increments – including sugar by the teaspoonful in the slums – it sells top-up cards in denominations as small as KSh50, which secures about six minutes of talk time.

“People use the air time as they can afford to,” says Mr Joseph. “You will notice here that people pull into a gas station in the morning, they put in 100 shillings of petrol, and at night when they go home they put in another 100 shillings. Kenyans don’t like to spend money when they’re not sure what they’re going to need it for.”

The incomes of some are so low that even the cheapest handsets available in Kenya – such as a standard Nokia for KSh2,700 – are beyond their reach. Many people have taken to buying only a Sim card, which costs KSh50, and clubbing together with family members and neighbours to share a handset, passed between Sim-owners when they want to make calls or send messages.

The company has also responded to demand for agricultural information from farmers in remote areas who tend to lose out in transactions with commodity traders if they do not have independent access to market data.

Picking his own phone out of his pocket, Mr Joseph demonstrates how they can send a text message that, seconds later, triggers an automated response from a system tied to the Kenya Agricultural Commodities Exchange. “So a 126kg bag of cabbage,” he says, reading from his screen, “is 1,000 shillings in Nairobi and 2,200 in Mombasa. So you’d better go and sell your cabbage in Mombasa.”

But competition does not just come from its sector rival: “For consumers, it’s a choice between buying a beer, buying a cigarette, walking a bit further rather than taking a bus, or buying airtime. So, our biggest competitor in fact is not Celtel. Our biggest competitor is the breweries. We are fighting for their share of purse.”

He says Safaricom’s average revenue per user – a key industry measure – is “much lower” than that of US and European operators: “One tenth, maybe.” Its fat profit margin has more to do with wages and base station rents that are much cheaper than those of peers in the developed world. “We are still growing. They are maxed out and their pricing is probably much more competitive on a relative basis, though of course ours is much lower,” he says.

Safaricom’s record earnings, however, have sparked a backlash from cash-strapped subscribers. Many are outraged that the company some call “Safaricon” can earn such large profits yet, they say, do little to lower prices, reduce network congestion, cut queues in customer service or fix its erratic billing system.

“We do screw up, yes. Every company does,” Mr Joseph says, adding that Safaricom needs big profits to invest in solving its problems. “We are the only company or the only industry that, every time we say or do something, is asked to push its prices down. Where is the pressure on the breweries, who make a not inconsiderable profit, to push their beer prices down?”

Small game hunter: turning text messages into shillings

David Mwangi emerges from a dusty concrete cubby-hole on the edge of Mathare, a Nairobi slum where the smell of rotten fruit hangs over lop-sided shacks.

Gripping his mobile, he has just visited a Safaricom agent and used the phone to send KSh500 ($7.50) across Kenya with the company’s innovative money transfer service.

Launched in March and known as M-Pesa, the service allows the “unbanked poor” like Mr Mwangi to deposit, transfer and withdraw cash with their mobile phones. It also shows how mobile companies in Africa are having an influence beyond the sphere of telecoms.

M-Pesa has spread rapidly in Kenya and now has around 225,000 customers. But in a country with a vibrant informal economy, where only 19 per cent of adults have a bank account, its potential for growth is huge.

Users can put up to KSh50,000 into an account by depositing cash with one of 625 local agents, who are based in Safaricom outlets, petrol stations and supermarkets. They transfer money to other mobiles by sending a text message with a pin code. The recipient collects cash from another agent by showing the code, identification and answering a security question.

Mr Mwangi uses the service to send money back to his wife in his home village. Before M-Pesa, delivering money meant a six-hour bus journey on poor rural roads.

Safaricom profits by charging customers about 5 per cent of the value of their transfers. The company’s biggest challenge is to keep up with soaring demand by finding new agents and ensuring they have enough cash to turn text messages back into shillings.