Skip to content
The Product Guys
All patterns
Money5 min read

The Free Trial

Time-boxed full access, given on the bet that habit will outlast the clock.

The problem it solves

People will not pay for software whose value they cannot feel yet. A trial moves the payment decision to after the experience instead of before it.

Where a trial actually converts

1SignupIntent is highest here2SetupFirst drop-off3ActivationReal data or a teammate lands4HabitSecond and third session5Trial endsThe decision is mostly already made
Conversion is decided in the setup and activation steps, not on the day the trial expires. A reminder email cannot rescue an account that never reached a useful state.

A trial trades certain cost for probable revenue. You give away the product for a fixed window and hope that by the end, the person has data, settings, or teammates inside it. The design question is not how long the trial runs. It is what state the account has to reach before the clock stops mattering.

Card up front or card later

Card required

  • Fewer signups, higher share of them convert
  • Conversion is partly inertia, so early churn runs high
  • Filters out students and tyre-kickers, and also cautious buyers
  • Requires an honest reminder before the first charge

No card

  • More signups, lower share of them convert
  • Every conversion is a decision someone made on purpose
  • Needs a real upgrade moment inside the product
  • Support load is higher, and so is the top of the funnel

Length should follow the natural cycle of the work, not a round number. A tool used daily can prove itself in seven days. A tool used at month end needs to survive a month end. Fourteen days is a default, not an answer.

  1. 1Name the activation state: the specific thing an account must contain for the product to be useful, such as one project with three collaborators.
  2. 2Set the trial length to cover at least one full cycle of the work that produces that state.
  3. 3Instrument the gap between signup and activation, and treat that gap as the trial's real metric.
  4. 4Start the clock at activation rather than signup if the product needs setup time, or offer a one-time extension when setup stalls.
  5. 5Send a reminder before the trial ends that states the date, the amount, and how to stop it.

Worked example

A hypothetical extension

A team notices that accounts which import data convert far more often than those that do not, and that importing takes about nine days on average. They could extend the trial to twenty-one days for accounts that have started an import and left it unfinished. The bet is that the blocker is elapsed time rather than intent. It is testable, and it can fail: the extension may simply delay the same no.

The end of a trial is the honest moment. A product that quietly starts charging and calls it a conversion is using forced continuity, which regulators in the United States, the United Kingdom, and the European Union have all moved against. A reminder email costs a sliver of revenue and buys the right to keep the customer's goodwill.

When it fits

  • Value depends on the user's own data or team being inside the product, which a demo cannot fake.
  • The product is cheap enough to serve that giving it away for a few weeks does not hurt.
  • There is a clear activation: The moment a new user first gets the value the product promised. state you can define and measure.
  • The buying cycle is short enough that a trial window can span it.

When it backfires

  • Setup takes longer than the trial, so the clock runs out before the product has done anything.
  • The trial exposes every feature at once, so the person leaves confused rather than convinced.
  • The card is taken up front and the first charge arrives with no warning, which converts a customer into a chargeback.
  • The product is bought by a committee, where a fourteen-day clock is shorter than a single procurement email thread.

Products using it

  • Figma

    Keeps a free tier alongside trials of paid plans, so the account survives the trial ending rather than going dark.

  • Ahrefs

    Has historically restricted or charged for trial access, treating the trial itself as a filter against low-intent signups.

  • Netflix

    Ran card-up-front free months for years and then withdrew them in many markets, showing that trials are a growth phase choice rather than a permanent fixture.

The psychology under it

The takeaway

A trial buys time for activation, so measure activation and let the length follow from it.

Finished the teardown? Bank it and the day counts toward your run.