Our free trial asks for a credit card, and most people leave at that step

What we did

On October 27 we changed Enjoy to one paid monthly plan. The first week is free, but you have to enter a credit card to start it. If you don’t cancel during the week, we charge the card.

Most people who get to the credit card form leave without entering a card. That surprised me, because the week is free and canceling is easy. I didn’t know if this was normal. I also didn’t know if I should stop asking for a card, or drop the free week and charge right away.

So I looked at our numbers in Stripe and read what other companies have written about this. I’m sharing what I found because anyone building an app with a subscription has to make the same choice.

What our numbers show

About 63% of the people who reach the card form leave without starting the trial. Very few of them come back later, about 6%.

For the people who did start a trial, and whose week has ended:

• 35% paid.
• 49% canceled during the week.
• 16% didn’t cancel, but their card was declined.

A lot of cards were declined. When the week ended we charged everyone who hadn’t canceled, and almost a third of those charges failed, mostly because the card didn’t have enough money on it. So having someone’s card doesn’t mean you get paid.

We also lowered the price by about half on November 7. Of the trials that ended before then, 25% paid. Of the trials that ended on or after that day, 47% paid. It’s early, so it might not mean much. But some people in the first group gave “too expensive” as their reason for canceling, and nobody in the second group did.

A lot of people cancel on the first day

Of the people who canceled, 56% did it within a day of starting, and 39% did it within the first hour.

I don’t think they tried the app and decided against it that quickly. I think they entered a card because they had to, and then canceled right away so they wouldn’t be charged by accident. For those people, asking for a card doesn’t get us a payment. It’s just an extra step.

RevenueCat reports something similar in its 2026 report on subscription apps: more than half of the cancellations of three-day trials happen on the first day.

What PhoneBurner found

The most useful thing I read was a post from PhoneBurner. For a little over three months they ran two versions of their signup at the same time. One asked for a card and one didn’t. Their results:

• Without the card, 239% more people signed up.
• 73% of the trials with a card became paying customers, compared with 22% of the trials without one.
• Overall, the version with the card produced 35% more paying customers and 16% more revenue.
• The customers from the version with the card canceled at about twice the rate.

Even though the card version had better numbers up front, they chose the version without a card. They wrote that “lifetime value is a far more important metric than conversion.” They said the biggest lesson was that not asking for a card put them on the same side as their customers.

The point I found most interesting was about the people who don’t pay. Without a card, a person who tries the product and doesn’t upgrade still has an account. The company can stay in touch with them, and they might pay later. With a card, a person who leaves at the card form never tries the product at all.

That matches what I see with Enjoy. Only about 6% of the people who leave at our card form come back, and the rest never really get to use it.

Other companies got different results

Nusii removed the card requirement for four weeks in 2017. They got five times as many trials, but the share that paid dropped from 30% to 3%, and their growth stalled. They went back to requiring a card.

Simple Analytics also removed the card requirement. They got more signups and a smaller share of them paid, so the number of new paying customers didn’t change.

First Page Sage collected numbers from 86 software companies. In their data, about 70% fewer people start a trial when it requires a card, and a much larger share of the people who start end up paying: about 49%, compared with 18% for trials without a card.

So what’s happening with Enjoy looks normal for a trial that requires a card. Whether removing the card helps seems to depend on the product.

What I’m going to do

I haven’t decided yet.

With the card, roughly 13% to 17% of the people who reach our card form end up paying. If we removed the card and got results like PhoneBurner’s, that number would go up. If we got results like Nusii’s, it would go down a lot. I don’t know which one we would be.

I don’t think charging right away, with no free week, is the answer. People would still have to enter a card, and they would be paying before they had used the app, so I’d expect even more of them to leave.

For now we’re doing one thing. Since November 9, when someone leaves at the card form, we send them one email offering the free week without a card. The first of those trials end around November 15, and I’ll update this post with how many of them pay.

If you’re deciding this for your own app

A few things I’d suggest, based on all of this:

• If you require a card, expect most people to leave at the card form. That’s normal.
• Compare the two options by how many people end up paying and how long they stay, not by how many start a trial.
• Check how soon people cancel. If a lot of them cancel on the first day, they’re probably just avoiding the charge.
• Expect some cards to be declined.
• Think about whether you can stay in touch with people who try the app and don’t pay. You give that up when you require a card.

Where the numbers come from

Our numbers are from Enjoy’s Stripe account as of November 10, 2026. They cover about two weeks, so they could change. The numbers for other companies are from their own posts, which are linked above.