Based on Good Strategy Bad Strategy: The Difference and Why It Matters by Richard Rumelt. 

The Kernel

Every good strategy has three parts working as one unit. The Diagnosis names the single most important challenge the company faces. The Guiding Policy decides how to meet that challenge by ruling out what the company will not do. The Coherent Actions are the specific moves in hiring, pricing, marketing, and operations that all reinforce the same logic. Remove any one of the three and what remains is not strategy.

In early 2003, David Kimani sat in the back room of the Ronald Ngala Street branch counting receipts against supplier invoices. The numbers, as they had been for twenty-four months, did not agree. Nakumatt was four hundred metres from his front door, Uchumi was five hundred, and Tusker Mattresses had a branch near the OTC terminus that pulled in morning commuters before most Nairobi shoppers had finished breakfast. David had arrived in 2001 with the intention of beating the competition. He priced below them where his cost base allowed and waited. The customers came, but not in the numbers the rent required, and not fast enough to compensate for the supplier credit his rivals had secured and he had not.

What made Nairobi expensive was not rent alone. The suppliers who had extended the Mukuhas credit in Rongai and Naivasha, where the brothers' payment history was personal and well-known, had no relationship with David in the capital. The established retailers had already secured the best terms, and a new entrant from Naivasha negotiated at the end of the queue. Without direct supplier access, David bought through middlemen. Each middleman added cost. Each added cost narrowed the price gap that was the only argument David had to make to a Nairobi shopper who already knew Nakumatt's address.

In February 2004, David drove to Machakos, 65 kilometres southeast of Nairobi. What he found was a market that had never had even a single retailer. The shops along the main street priced without competition, restocked without consistency, and sold to customers who had no reference point for what fair pricing looked like.

The diagnosis was that the gap Naivas could fill was not in cities where formal retail already existed. It was in every Kenyan town that formal retail had bypassed because the basket sizes were smaller and the commercial streets were less impressive. The guiding policy that followed ruled out, once and for all, the ambition to compete with Nakumatt and Uchumi on their own terms in their own streets. Naivas would enter markets where the competing set was dukas and market traders, not supermarket chains. It would price against local trader rates, not against Ukwala's shelf prices. It would also close any branch that could not sustain itself, because a loss-making store in the wrong market consumed the cash a profitable store in the right one was generating.

The coherent actions were each an expression of that policy. Every new store was funded from operating profit, because borrowing to expand into an untested market compounded the risk of a wrong location. Every site went through a feasibility study before commitment, a structured assessment of purchasing power, competitive conditions, and realistic sales projections, because the Ronald Ngala Street experience had already shown what happened when a store opened in the wrong market with the wrong cost base. By 2009, the company had ten stores, eight outside the capital.

Chain-link Systems

Some businesses are built so that every part depends on every other part, and the weakest link determines the strength of the whole. Once every link is brought to a high standard, the business becomes very hard to copy, because a competitor must match every link simultaneously, not just the most visible ones.

In 2019, Adenia Partners, a Mauritius-based private equity firm, acquired majority stakes in Quickmart and Tumaini Self Service in consecutive transactions, merged both chains under the Quickmart name, and started expanding at the pace that private equity capital permits. By 2024, Quickmart had 84 stores across 16 counties and employed more than 7,000 people. In the same year, Naivas had 114 stores across 38 counties and employed 12,000 people. The gap between the two was a measure of what capital alone cannot purchase.

To understand that gap, count the links that Naivas had assembled before Quickmart had its first investor. The first was geographic depth. Naivas had entered Eldoret, Embu, Kisii, Kitui, and Garissa before a credible competitor appeared in any of them. In each of those towns, Naivas had arrived as the only formal retailer and had spent years becoming the default store in the customer's weekly routine. A competitor entering the same town in 2023 faced a market where Naivas had already formed the habits. 

The second link was the supplier network. Naivas had spent the 1990s buying stock in cash, because suppliers in Nairobi in 2001 extended credit to retailers they knew and declined it to retailers they did not. The founding brothers spent those years building what supplier credit actually required: a payment record. By 2024, the company sourced from more than 4,500 supplier partners, with over 80 percent of its products made in Kenya, and more than 200 suppliers feeding the fresh produce department alone. Those relationships had been built over two decades of consistent payment. A competitor arriving in 2019 with capital to spend and no payment history to show negotiated from the position Naivas had occupied in 2001. 

The third link was the NaivasCard. By 2024, it had 3.5 million registered members and recorded over 285,000 daily customer interactions. The data those interactions produced told Naivas which products sold in which branches on which days, which categories peaked before specific holidays, and which promotions had moved volume in previous years. A competitor could launch a loyalty program in an afternoon but could not compress three million customer purchase histories into a database in the same afternoon. The data Quickmart was generating in 2024 was what Naivas's data looked like in 2014. Purchase intelligence compounds in direct proportion to how long a company has been collecting it.

The fourth link was... Continue reading this playbook below 

This story first appeared on episode two of Kenyan Founders, The History and Business Strategy of Naivas Supermarket.