The Routing Line From Nairobi to Accra

There is a number that haunts everyone who has ever tried to sell a physical thing in Africa, and it has nothing to do with manufacturing. On this continent, getting a product from a warehouse to a doorstep can swallow as much as three-quarters of its retail price. That is not a rounding error. It is the difference between a business that scales and one that quietly closes after the third bad month. It is, in a sense, the tax that geography levies on ambition.

Leta, a software company founded in Nairobi in 2021 by Nick Joshi, has spent its short life arguing that this tax can be negotiated down with code. Not with new roads, not with cheaper fuel, not with the kind of grand infrastructure announcement that gets a ribbon cut and then waits a decade to matter. Just with software that decides, ten thousand times a day, which truck should go where, and in what order.

The unglamorous math of the last mile

What Leta sells is a routing engine. It plugs into the systems a company already runs, the inventory ledgers and point-of-sale tools and order managers, and then it does the thing a dispatcher with a whiteboard cannot do at scale: it matches every shipment to the right vehicle, sequences the stops, and re-plans when a road floods or a customer reschedules. The promise is that the same fleet, doing the same deliveries, simply costs less because it stops wasting kilometres.

The traction figures suggest the pitch is landing. The platform now processes over 10,000 trips a day and has handled 4.5 million deliveries, up from 500,000 in 2022. The fleet under its management has grown from 2,000 vehicles to 7,400 in two years. Revenue, the company says, has risen fivefold over the same stretch. Its client list reads like a tour of African consumer demand: KFC, Diageo, the brewer EABL, the Kenyan retailer Gilani.

These are not consumer names you would recognise from a glossy startup launch. They are the boring, enormous accounts that move millions of crates and care about exactly one thing: cost per delivery. That is the quiet genius of the business. Leta is not trying to charm the public. It is trying to win the spreadsheet.

Why the map keeps redrawing itself

In March 2025, Leta closed a $5 million seed round led by the European firm Speedinvest, with participation from Google's Africa Investment Fund and the climate-focused fund Equator. It was the kind of round that, in a leaner funding year, signalled conviction rather than froth. And it underwrote a deliberate expansion. The company now operates across Kenya, Nigeria, Uganda, Zambia and Zimbabwe, and in mid-2025 it added Ghana, planting a flag in West Africa's most watched logistics market.

The Accra move is worth dwelling on, because it reflects a wider pattern. African startups that began in one corner of the continent are increasingly redrawing their expansion maps to run through Ghana, treating it as a gateway rather than an afterthought. For a routing company, each new country is not just a market. It is more roads, more traffic data, more failure modes for the algorithm to learn from. The map gets denser, and the software gets sharper, and the moat gets a little harder to cross.

The most valuable thing a logistics startup owns is not its trucks. It is the accumulated memory of every trip that went wrong.— An observation on why routing data compounds

The pivot hiding in plain sight

Here is where the story turns from a tidy software business into something more interesting. Leta is now piloting embedded finance: fuel cards, vehicle financing, supply-chain credit aimed at the fleet operators who already live inside its software. This is the move that logistics platforms across the continent keep gravitating toward, and the logic is almost gravitational.

Once you know exactly how many trips a truck makes, how much it earns, and how reliably it shows up, you know something a bank in Lagos or Lusaka could never easily learn: you know whether that truck is creditworthy. The routing data is not just an efficiency tool. It is an underwriting file. The company that sees the fleet most clearly is the company best positioned to lend to it.

That is the bet, and it is a familiar one. The risk is equally familiar. Lending is a different animal from software; it has a balance sheet, a default rate, a regulator, and a way of punishing companies that mistake confidence for capital. Plenty of African platforms have reached for embedded finance and found that the margins that looked so inviting came attached to risks they were not built to absorb.

But the direction of travel is unmistakable. Leta says it wants to double revenues and push further into Africa and the Middle East. Whether it becomes a logistics company that does a little lending, or a lender that happens to route trucks, is the question its next chapter will answer. What is already clear is the thesis that got it here: in a continent where the road is the most expensive part of the sale, the most valuable real estate is the algorithm that decides which way to turn.