How to make an on-premise software trial actually expire

An on-premise evaluation ends when the customer takes an action: uninstalling, decommissioning the VM, telling their team to stop.

Look at the incentives on their side. Ending the pilot costs them effort and removes something their team has started relying on. Not ending it costs them nothing. There is no deadline in their calendar, no invoice, and nobody in their organisation whose job it is to switch off software that works.

The asymmetry is total. You need them to act, and they have no reason to.

So the sixty-day pilot passes day sixty. The champion who sponsored it changes role. Budget slips a quarter, then a year. Procurement never issues an order because nobody escalated it.

Meanwhile the deployment sits on their infrastructure doing real work. In the worst version it quietly becomes production, because a team started depending on it while the deal stalled, and now nobody wants to be the person who removes it.

You are supplying a customer who has never paid you, and your pipeline still shows the deal as open.

The standard remedy is a sequence of increasingly awkward emails asking whether a decision has been made and noting that the evaluation period has passed.

Those emails damage the deal. They move the conversation from whether the product is worth buying to whether you are being difficult about a technicality, and they put your champion in the position of defending you internally over an administrative matter.

You are also asking for something you cannot verify. Whatever they reply, you still do not know what is running.

When the protected parts of your software check for permission, the pilot expires by itself. Nobody at the customer has to act, and nobody at your company has to ask.

What happens next is the useful part. A stalled evaluation becomes a live conversation, opened by them, at the moment the product stops. Those are the best conditions you will ever get for a buying discussion, because the value has just been demonstrated by its absence.

Extending becomes a decision you make in a dashboard, which is a considerably better position than a chase email.

The second-order effect is that you can be more generous with pilots rather than less.

Most vendors restrict evaluations precisely because they are hard to end: limited data volumes, limited users, limited features. A pilot that reliably stops on schedule can be a full-fidelity pilot, and those convert better.