I turned down the bank pilot because it would have consumed the people, attention and product decisions we needed to reach the smaller businesses the company was built for. Prestige can extend a conversation with investors, but it cannot replace evidence that ordinary customers will keep using and paying for the product.
At 3:40 p.m. on a Friday, Yaw was sitting in a shared office in Accra with a cold cup of sobolo beside his laptop and a draft pilot agreement open on the screen. This is an illustrative composite, but the choice is familiar. The bank’s name would have changed how people introduced him in rooms where his startup was still unknown. The pilot came with senior meetings, a long procurement path, security reviews and a request for workflows that belonged to the bank’s internal operations.
His own product served small merchants who spent their evenings chasing payment confirmations and reconciling messages from different channels. They had been asking for a simpler way to see what had cleared, what had failed and what needed a human call. The team had enough runway to choose carefully, not enough to build two products at once.
By Monday, the bank expected an answer. If Yaw accepted, the merchant roadmap would pause. If he declined, he might lose the most recognisable opportunity the company had received.
The pilot offered credibility, then asked for the roadmap
The first version of the proposal looked manageable. A contained pilot. A limited set of users. A chance to prove the company could work with a major institution.
The detail changed the decision.
The bank needed approval flows, reporting fields and integration work that made sense inside a large regulated organisation. Each request was reasonable on its own. Together, they would pull the product toward a buyer with a different job to be done.
Yaw had already seen the early signs. In calls with merchants, people did not ask for internal reporting structures. They asked why a payment marked as sent still had not appeared. They wanted to know which customer to call before closing the shop. Their problem was immediate and repetitive. The bank’s problem was important, but it was different.
A prestigious pilot can create a dangerous form of momentum. Everyone becomes busy. The calendar fills. The founder has an impressive answer when someone asks what the company is doing. Meanwhile, the people who would use the product every week wait for a roadmap that has quietly stopped belonging to them.
That is the harder part to see from inside a small team. The cost rarely arrives as one large invoice. It arrives as six weeks of engineering work, another meeting added to Tuesday, a delayed customer interview, and a product manager explaining why the promised fix has moved again.
Runway belongs to the customers you chose to serve
Yaw wrote two columns on a sheet of paper. One listed what the pilot might bring: a respected logo, a possible route to revenue, introductions, a stronger story for future fundraising.
The other listed what the team would give up: the next merchant release, direct learning from current users, and the ability to correct the product while the team was still small enough to change direction quickly.
The second column was longer.
This did not make the bank a bad prospect. It made the timing wrong. A startup can take work that funds the company while still protecting the product, but that requires a clear boundary: what the contract pays for, what it does not change, and who decides when customer work resumes. [Ama's customer contract]( /blog/ama-s-customer-contract-six-weeks-to-prove-paid-use-without-losing-the-roadmap-d2557dc1/ ) explores a related tension between proving paid use and keeping the roadmap intact.
The useful test was simple: would the work help the company learn faster from the customers it planned to serve next year?
For Yaw, the answer was no. The pilot would teach the team how to satisfy one institution. It would not tell them whether a market trader, a small logistics firm or a growing retailer would return to the product next week.
That distinction matters more as fintech companies move into regulated services. Operating maturity and regulatory readiness matter. So does knowing which maturity work supports the business you are actually building, rather than the most impressive meeting on the calendar.
Saying no created a smaller, more useful commitment
Yaw did not send a dramatic rejection. He explained that the team could not commit to a bespoke pilot without changing its product priorities, and that he would rather be clear than promise work they could not deliver well.
Then he made the alternative concrete. The team would spend the next release cycle on the recurring reconciliation problems already appearing in merchant conversations. They would ask existing users to show them the moment the process broke, rather than describe it after the fact. They would keep a narrow list of work that could be reversed if the evidence changed.
The following Wednesday, Yaw sat with a merchant owner after closing time. A receipt printer was still clicking behind the counter. She opened her phone, found two payments that had been marked incorrectly, and pointed to the exact point where she stopped trusting the system.
That conversation gave the team a decision they could act on the next morning.
The bank pilot might have produced a better slide. This gave them a better product question.
Comments
No comments yet.