From pilot to contract: turning a trial into production revenue

Most deeptech pilots do not fail on technical results. They stall because nobody agreed in advance what success means, who owns the budget for the rollout, and which qualification steps the company must pass to become a supplier. A pilot that is not written down as a path to a contract is a paid experiment.

The gap between a technical yes and a commercial yes is filled with procurement work: supplier qualification, approved vendor lists, first article inspection, factory acceptance testing. None of it is glamorous, and all of it decides whether the pilot converts.

The guides below cover both sides: what to agree before the pilot starts, and what the buying organisation checks before it can place an order.

1. 01Your pilot worked. So why did nobody buy?The pilot proved your technology works. It did not prove the organisation can buy it. On pilot purgatory, buying groups, and why movement is not commitment.
2. 02The pilot agreement: five things to settle before anyone plugs anything inMost pilots fail commercially long before they fail technically. The reason is almost always a document nobody wrote: what success means, who decides, what happ
3. 03Design partners: free development department or your first real customer?Every hardware startup collects design partners. Few ask the one question that matters: is this company a customer in disguise, or a free consultancy with a log
4. 04An LOI is not an order: what actually counts as commercial proofYou have a signed letter of intent. An investor asks if it is revenue. A procurement team knows it is not. Here is what LOIs, MOUs and purchase orders really me
5. 05Your customer wants exclusivity. Excellent. Now make them pay for it.A broad exclusivity clause in a joint development agreement is not commitment. It is a request for yours. How to trade field-of-use exclusivity for real commerc
6. 06Your material is 20% better. The customer still won't qualify it.Industrial customers do not buy the best material. They buy a qualified one. On qualification timelines, requalification events, and picking the first customer
7. 07Your customer says they'll buy it. Will their bank believe them?Twenty interested customers can be worth less than one offtaker with its balance sheet behind future demand. On LOIs, conditional offtake, take-or-pay and what
8. 08Your chip passed the evaluation. So why is nobody designing it in?An evaluation board on an engineer's desk is not a deal in progress. It is an option someone bought cheap. The gap between eval and design win is where chip sta
9. 09TRL vs MRL: your prototype works, and that is not the question anyone is askingFounders pitch TRL 8. Buyers hear MRL 4. The gap between those two numbers is where deeptech deals, and deeptech companies, quietly die.
10. 10Manufacturing readiness levels, explained by someone who has watched a line stopYour prototype works. That is level 4 of 10. The other six levels are where hardware startups discover that making one thing and making a thousand things are di
11. 11Supplier qualification: the part of the sale that happens after they say yesThe buyer agreed in March. The first order came in September. Nothing went wrong. Here is what happens in those six months, and how to get through it faster.
12. 12The approved vendor list: why being better is not enoughYour product won the technical evaluation. The order still did not come. Here is how the approved vendor list actually works, and how small suppliers get on one
13. 13First article inspection: the paperwork that decides if you can actually make the partYou made one perfect part by hand. The FAI is where the customer asks you to prove the process made it, not the heroics. Here is what the inspection actually ch
14. 14The factory acceptance test: the exam you get to write yourselfThe FAT is the last gate before your machine ships. Startups treat it as a formality. Industrial buyers treat it as the moment of truth. Write the exam yourself