When seed equity thins and debt becomes the larger source of startup capital, founders need to build products that can reach a paid, repeatable workflow before financing becomes urgent. The product decision moves earlier: choose a narrow customer, a priced problem, and a delivery plan that can survive a repayment date.
In April 1970, Apollo 13 lost an oxygen tank on its way to the Moon. Jim Lovell, Jack Swigert and Fred Haise could no longer follow the mission they had launched to complete. NASA’s teams had to preserve power, water, air and a route home with the equipment already in the spacecraft. The lunar module became a lifeboat; engineers devised an adapter so the command module’s square lithium hydroxide canisters could help remove carbon dioxide in the lunar module.
NASA’s Apollo 13 Flight Journal documents a mission shaped by constraint after the original plan had disappeared. The outcome remained uncertain for days. Every action had to protect the next one.
That is closer to the product reality facing an African founder than the usual seed-round narrative. A debt facility can keep a company alive, but it also turns time into a product requirement. Your roadmap has to produce cash before the next obligation arrives.
Build around the cash-producing workflow
A founder with a broad AI platform idea often starts by building the most impressive version of the product. That choice gets expensive when capital is patient. It gets dangerous when revenue must arrive on a schedule.
The earlier question is simpler: which person has a recurring task costly enough to pay for now?
For a logistics business in Accra, that may be reconciling delivery records before a customer disputes an invoice. For a small lender in Lagos, it may be preparing a reviewable decision record before a credit officer acts. For an operations team serving clients from Berlin or New York, it may be turning scattered requests into an owned queue with a clear handoff.
Each is smaller than “AI for operations.” Each can become a paid workflow.
The product should show its value in the customer’s existing work before it asks them to change how the whole company works. That means a narrower first integration, a manual fallback, and a person responsible for the output. A polished demo with no owner creates activity without a route to revenue. [Ama’s pilot had the same problem](./blog/ama-s-polished-demo-no-pilot-until-the-workflow-had-a-clear-owner-96247638/).
Treat repayment timing as a design constraint
Debt changes the cost of product uncertainty. It does not make a weak product stronger. It shortens the time available to discover that the customer’s problem was less urgent than it looked in a discovery call.
Before adding a feature, ask three operational questions:
- Can this feature help close or retain a customer within the period our cash plan requires?
- Can we deliver it without adding a permanent cost that depends on a future round?
- If the model fails or the data is incomplete, can the customer still complete the task?
These questions can produce unglamorous answers. A founder may postpone a custom dashboard, keep a human review step, or decline a large request that would turn one customer into a separate product line. Those choices protect the product from becoming a collection of unpaid promises.
The important distinction is between revenue that looks good in a deck and revenue that funds the next month of building. A one-off implementation can be useful, especially when it reveals the workflow. It becomes a problem when the team quietly depends on another custom project to cover every fixed cost.
Price the risk you are carrying
Debt is often discussed as a finance decision. Customers experience it through product decisions: how much setup you offer, how quickly you promise results, how much custom work you absorb, and whether you charge for work that only looks like onboarding.
If an AI product needs customer data cleaned, policies agreed, and exceptions reviewed before it can operate safely, that work has value. Price it, scope it, or reduce it. Do not hide it inside a low monthly fee because the team wants the contract to feel easy.
This matters across African, European and US markets, where a buyer may expect a different procurement rhythm but still needs a clear reason to trust the output. A paid pilot should have an owner, a defined workflow, and a decision about what happens after it ends. Otherwise, the founder has borrowed against ambiguity.
The first repayment date can expose this sharply. [This is the moment a loan can turn into an emergency](./blog/the-first-repayment-date-that-can-turn-a-loan-into-an-emergency-8de092e3/), especially when product work has been funded by hope rather than a customer commitment.
Keep enough oxygen for the return journey
Apollo 13 did not reach the Moon. The mission succeeded because the crew and ground teams changed the objective from landing to getting home.
Founders do not need to abandon ambition when funding conditions change. They need to name the immediate objective accurately. It may be reaching three paying customers with the same workflow. It may be proving that a customer renews after the first month. It may be stopping a feature branch that consumes engineering time without improving retention.
Write down the product decision that protects the next six months of learning. Then make the next build serve that decision. A thinner seed market rewards the company that can keep learning while cash is constrained.
Comments
No comments yet.