The Endowment Effect and What a Trial Should Feel Like
Ownership raises what people will accept, but only when they genuinely own something.
Tessellate sells a floor plan tool to estate agents. The 14-day trial gives full access to everything. Trial-to-paid sits at 4 percent. The team's theory is that the trial is too short, so they extend it to 30 days. Conversion stays at 4 percent, and the median trial user opens the product twice. Length was never the variable.
What a trial is trying to manufacture is ownership. A user who has built something in your product is deciding whether to give it up. A user who has looked at your product is deciding whether to buy it. Those are different decisions and they convert at different rates.
The gap ownership opens
The mug studies, and the argument about them
Kahneman, Knetsch and Thaler published a set of experiments in 1990. Half the participants in a room were given a college mug, then a market was run between owners and non-owners. Owners asked for roughly twice what buyers would pay, and far fewer trades happened than standard theory predicts. The gap was named the endowment effect, and it is usually explained by loss aversion: Losing something feels worse than gaining the same thing feels good.: giving up the mug is coded as a loss, acquiring it as a gain.
Be aware of the critique. Plott and Zeiler published work in 2005 arguing that much of the measured gap comes from features of the experimental procedure, including how the task is explained and what participants infer about the experimenter's expectations. With different procedures the gap shrank or disappeared. The effect has not been dismissed, and plenty of field evidence is consistent with it, but the size is contested and the mechanism is not settled. Cite it as a useful frame, not as a number.
Three levels of trial ownership
- Access
- The user can see the product. Nothing is theirs. This is what Tessellate's 14 days provided, and extending it just gave people longer to not start.
- Artefact
- The user has made something: a floor plan, a report, a configured pipeline. Ending the trial now costs them work. This is where the effect has something to grip.
- Dependency
- Other people rely on the artefact. A client has the link, a colleague has a login, a Zapier job runs off it. Cancelling is now a conversation with someone else, which is a much higher bar.
The design job in a trial is moving users from access to artefact as fast as honestly possible. For Tessellate that means the trial should not open on an empty canvas. It should ask for one real address, pull the floor plan from the listing, and hand the agent something they could send to a client within four minutes.
Trial copy selling access
- Start your 14-day free trial. Full access, no credit card.
- Explore all Pro features
- Your trial ends in 3 days. Upgrade now to keep access.
Trial copy building an artefact
- Make your first floor plan. Paste a listing URL and we'll draft it.
- Your 3 plans are saved. Add a client to share them.
- Your trial ends Friday. Your 3 plans stay in your account, view-only, until you choose a plan.
The third row matters more than it looks. Threatening to delete work is a manufactured loss and it produces angry conversions and refunds. Saying the work is safe but frozen keeps the real ownership intact while removing the coercion. The user still has something, they just cannot use it the way they want to, which is an honest description of not having paid.
Worked example
What Tessellate changed
They cut the trial back to 7 days and put a single question on the first screen: what is the address of a property you are listing this week? The product produces a draft plan from the listing photos in about 90 seconds. activation: The moment a new user first gets the value the product promised., defined as one plan shared with a client, goes from 9 percent to 31 percent. Trial-to-paid follows it to 12 percent. The trial got shorter and converted better because the users owned something by day one.
A closing caution for anyone about to quote this in a growth review. The endowment effect is an explanation for why artefact-first trials work, and it is contested enough that you should not lean on it alone. The simpler explanation runs in parallel: a user who has finished one real task has evidence the product works for them. You do not need to pick between these. You do need to avoid stating either as a proven mechanism when the thing you actually measured was activation.
Quick check
Why did shortening Tessellate's trial from 14 days to 7 raise conversion?
The takeaway
A trial converts on what the user has built in it, so spend the trial producing one real artefact rather than granting broad access.
Try this tomorrow
Define one artefact a trial user must produce to count as activated, then measure what share reach it on day one. Redesign the first screen around that single outcome.
Answer the check above, then bank the day.
Where this comes from
- Experimental Tests of the Endowment Effect and the Coase Theorem (1990), Daniel Kahneman, Jack Knetsch and Richard Thaler
- The Willingness to Pay / Willingness to Accept Gap (2005), Charles Plott and Kathryn Zeiler
- Thinking, Fast and Slow, Daniel Kahneman
User Psychology is one of six tracks. These lessons summarise and build on the work above, they do not reproduce it. Buy the books, they are better.