One of my biggest dilemmas in app growth is whether to remove the free trial.
The case for keeping it
Plenty of apps give a trial, sometimes even on a weekly plan, and they grow just fine. It also makes sense for the user, who gets to see the thing before paying for it.
If your app converts 50% from trial to purchase, and I have seen that, keep the trial, because that is a funnel that works. For a wider view than one app, RevenueCat’s State of Subscription Apps 2026 breaks trial to paid conversion down by trial length and category.
The case for cutting it
The other side looks at how many apps barely convert trial to paid, and the harder questions start.
When someone starts a trial for a complex app and never pays, are they just not taking it seriously? Or is it a sign the product market fit is not there yet?
And should a high intent channel like Apple Search Ads, where someone is actively searching for what you do, be treated the same as a discovery channel like Meta or TikTok, where nobody was looking for you? Those two users arrive with completely different levels of commitment, and giving both the same free entry is a decision, even when it is made by default.
I do not have a clean answer to all of it. But there is real value in running a campaign for purchases only and seeing what happens for yourself.
The Meta constraint that often decides it
Meta adds a dilemma of its own, which is how to optimize.
You can build a custom event and assign values so you bid for trials and direct purchases together. But there is no clean way to optimize for trial starts and purchases at the same time.
So on the app where I tested this, I optimized for purchases the whole time, even while the paywall still offered a trial. When I removed the trial, I kept optimizing for purchases, and the numbers looked a lot better, on a result that now meant a purchase only, which is worth sitting with.
How to decide
- What is your trial to paid conversion, and how many installs start a trial at all?
- High, on a healthy share of installs: the trial is doing its job, so keep it.
- Barely converting: go to 2.
- Where does your paid traffic come from? If it mixes high intent and discovery channels, consider splitting the treatment. Then go to 3.
- What do your Meta campaigns actually optimize for today? Also check whether your pricing favors ROAS bidding or purchase bidding on Meta.
- Purchases already: the trial dilutes the thing your paywall is being judged on without helping your bidding. You can test on the paywall instead of launching a new campaign. Remove the trial for paid Meta traffic only, keep the same campaign, creatives and event, and compare the weeks before and after. If your paywall tool can split new paid users between trial and no trial, do that; a before and after also catches whatever trend the account was already on.
- Something else: run a campaign for purchases only against your current setup before changing the paywall for everyone.
I ran a close version of that paywall test on one subscription app with paid Meta traffic, where the campaign already optimized for purchases. The full result, week by week, including what happened to conversion rate and cost per result, is in What happens when I remove the trial from my app?.