A free pilot is supposed to be a fast way to prove a product works before asking someone to pay for it. In practice, plenty of pilots run for months, get genuinely used, generate a friendly relationship with a champion inside the company, and then quietly stop, with nobody ever quite saying no. That isn't a product failure. It's a structural one, and it's set in place on day one of the pilot, usually by the founder, who was so relieved to get a yes to "try it for free" that they never asked the harder question underneath it: what has to be true for this to become a paying account, and by when.
An open-ended pilot has no reason to end
Free access with no expiry date is comfortable for everyone, which is exactly why it never resolves. The champion who agreed to try it has no reason to push for a decision, since the free version keeps working and nobody above them is asking about the bill. You have no reason to force the issue either, because an active pilot feels like progress, and pushing for money risks the relationship you've spent months building. So it drifts. Eighteen months later you'll find a graveyard of pilots that were technically still running, in the sense that a login still worked, and had stopped meaning anything in the sense that mattered around month four.
The fix has to be in place before the pilot starts, not raised awkwardly once it's underway. Set an end date at the outset: this runs for six weeks, and on this specific day we get on a call and decide, together, whether it becomes a paid contract. A pilot with a real end date behaves completely differently from one without, because both sides know a decision is coming and can't quietly avoid it.
Agree on the price and the finish line before anyone starts using it
The other half of the same mistake is negotiating price after the pilot proves itself, which sounds fair and turns out to be the worst possible time to do it. By then the champion has invested real effort making it work internally, and raising a number for the first time at exactly that moment feels, to them, like a bait and switch, even though nothing was hidden. Negotiate the price, and the specific outcome that would justify it, before the free period starts. "If this saves your team roughly the four hours a week you're describing, the price is this, and we'll both look at the numbers together in six weeks" is a completely different conversation from introducing a number for the first time once the free thing is already working and everyone's gotten used to not paying for it.
Agree on the finish line the same way. Not "let's see how it goes," which nobody can fail and nobody can point to, but a specific measurable thing: response time under a set number, a defined volume processed, a number your champion picked themselves, because they know their own team's actual bar for success better than you do. A pilot with no defined success looks successful to everyone and converts to nobody, because there's nothing concrete for either side to point at when the six weeks are up.
Your champion needs help you probably haven't offered
A pilot that goes well at the level of one person's daily work doesn't automatically create a budget line, and this is the part founders miss most often. The person running the pilot day to day is frequently not the person who signs off on a new vendor, and a pilot succeeding doesn't hand that person the internal case for spending money, it hands them a favorite tool they now have to go justify to someone else. If you only prove the product to your champion, you've done half the job. The other half is giving them the material to make the case upward: a short, specific summary of what changed, in language their manager or their finance person will actually read, not a features list aimed at the person who was already convinced.
Ask directly, partway through the pilot, who else needs to say yes, and build toward that person on purpose rather than assuming your champion will handle it alone once they're sufficiently happy.
What happens at the actual decision call
Show up to the end-date call with the numbers you agreed to track, not a fresh pitch. If the target was hit, this is where the pre-agreed price gets confirmed and turned into a contract, and it should feel like closing a loop you opened together six weeks earlier, not like a negotiation starting from zero. If the target was missed, find out honestly why before assuming the product simply wasn't good enough. Sometimes it wasn't used the way you both intended, sometimes the six weeks weren't long enough for the actual workflow, and sometimes the answer really is no, in which case a clean no on a set date is worth more than another three months of a pilot nobody was going to convert.
Where the tracking actually lives
Once a pilot turns into an actual conversation about money, it should live somewhere with real structure rather than in a thread you'll have to scroll back through. Rocketship keeps the whole account, from the first outbound message that opened the door through every call, every reply and the quote you eventually send, in one pipeline with a stage for exactly this: pilot in progress, decision pending. When the day comes, sending the quote and taking the first payment through Stripe happens in the same place the relationship has lived the whole time, instead of switching into a separate invoicing tool at the one moment that actually matters.
The rule underneath all of it
A free pilot that never had an ending was never really a test. It was a comfortable way to avoid finding out. Put the date, the price and the finish line in place before anyone starts using the product, and the six weeks that follow will tell you something real, whichever way it goes.