30
15 Comments

I rewrote my trial system to be credit card-less and now have 5x as many trials

Hey IndieHackers,

I'd like to share my early results with switching from a trial that requires a credit card to one that doesn't.

When I launched Polypane in May of last year, the easiest way to offer a trial was by using the trial method that Paddle, my payment provider, offered me. It required the least amount of code and got me up and running quickly.

Unfortunately it came with some major downsides:

  1. It wasn't possible to change the trial period after setting it, making experimentation really hard.
  2. People couldn't upgrade from individual to team during the trial.
  3. They couldn't decide to pay early.
  4. They had to fill in a credit card pr PayPal before being able to start a trial. Many developers (my target market) didn't feel comfortable doing that or simply didn't have access to the (company) credit card.

While the first three were mostly my problem, the last one caused many developers to bail during the sign-up process 😢

So, I knew it had to change.

The more people I spoke to, the more I realized that this credit card requirement was the biggest bottleneck for developers to experience how much Polypane could improve their work.

Work started 💪

I got to work and in about a month, in between writing massive guides to media queries and releasing new versions, I rewrote my entire trial flow to allow people to sign up without a credit card.

People can now sign up for Polypane with just a name and email. The trial length is controlled by an env variable on my server so experimenting will be really easy (14 days is kind of the standard, but who knows what happens if I move it to 7 or 30) and they can change to a multi user account any time during the trial to invite their team members. Lastly, they can convert to a paid subscription whenever they want. It's no longer tied to those 14 days so it can be after a few days as their employer walks by with the credit card, or a week after their trial ended.

Preliminary results: pretty good

It's been running "silently" for a week now. I haven't announced it anywhere so I could monitor if nothing broke before doing a big push. 🤞 So far, so good.

Even without promoting it trials are up by about 500%! 🤯

In particular the weekends have picked up. This makes sense, because devs previously needed the company credit card and would have waited until Monday, or just bailed.

With a two-week trial, I have to wait another week to see how many end up converting to a paid account, but just having so much more people try out my software and give feedback is already amazing.

Celebrating the new trial 🎉

So that's cause for a celebration! To celebrate the new trial method I decided to have an additional 25% discount for people that opt for a yearly license. Instead of $120 it's now just $89!

If you've seen Polypane before but didn't try it out because of the credit card requirement, now's the time to try it out!

  1. 2

    Any impact on the conversion rates @Kilian?

    Looks like quite a few of us are expecting to know that 😂

  2. 2

    Did your trial-to-upgrade conversion rate go up, down, or stay relatively the same?

  3. 2

    Thanks for sharing.
    I'm just getting ready to launch my product, so don't have any metrics yet. But I'm offering a tiered subscription plan that allows users to track 15, or 60 leads (plus other feature allowances). I'm also starting new accounts with a 'no card up-front' trial of 10 days with 5 leads.

  4. 2

    Thanks for posting, I'm on the same board regarding Paddle. I actually decided to for people to sign up, then pay, and then they get access. So that's the same as what you did before, including the trials. One argument for this is quality: putting in your creditcard shows you're serious about the service. So the real learnings will come in 14 days when those new trial customers need to decide whether they're going to pay. Please let us know here or in a new post how this goes!

    1. 1

      Yeah that's true, but I think it all depends on your target audience and their expectations.

      With my target audience being developers though, I think most of them really want to try something first before deciding to fill in a credit card, more so than other target audiences. We're after all used to just npm install something, or for Microsoft to give it away for free. So needing to fill in a credit card before trying doesn't fit with our usual flow.

  5. 2

    Sounds like a great idea @Kilian. Thanks for sharing this insight.

  6. 2

    Thanks for sharing these numbers. There are many variables in-play here, but have you seen a corresponding lift in conversions to paid plans too, or is it about the same as with the card-wall?

    1. 2

      I don't have numbers for that yet, as I'm just a week into this new trial system but my trial period is 14 days. I'll have to wait another week to see what happens there!

      As I replied to @Gabe, I expect the absolute number of conversions to paid plans to be roughly the same as before. Having more people in my trial can help me optimize that part of the funnel though, so with time improve the number of people converting to a paid plan!

  7. 2

    Congrats @Kilian. I’m curious to see how things change if you move to a 7 day trial or a 30 day trial. Would be interesting if you gave an update. I still have not decided which way is the best way to go with my product.

    Thanks

    1. 1

      When I launched I had a 7 day trial, but I got a lot of complaints about that being too short. I think it's pretty clear in the first hour or so whether Polypane works for you, and it's something most customers tell me happened for them, but I changed it to 14 and pretty much all complaints disappeared. Still, a 7 day trial is more interesting for me, so I'll definitely experiment.

      1. 1

        That's interesting. Have you noticed a difference in conversion to paid account from trial by adjusting the trial length?

        1. 1

          Adjusting the trial length helped getting more people to start a trail. I think with 7 days maybe a lot of people thing "well I'm busy this week, I'll try it some other time..." and then don't.

      2. 1

        Yes, I think agree with the 14 day trial. I guess I can always adjust if needed. I think its sufficient time for a user to test the product I plan to launch.

  8. 2

    Wow, congrats on these results! This is super insightful. The part about people not always having access to company credit cards makes a lot of sense. More people giving you their email also means more people to follow up with even if not all convert. Excited to see what the conversion rate is next week (please share).

    1. 2

      Thanks Gabe, will definitely do that!

      In researching for what to expect with this change the general trend seems to be that initially the number of people that end up being paying customers stays about the same despite having many more trials (because the barrier to try it is lower) but because of the extra feedback you get with the trials, you can then start improving the trial and onboarding flow and eventually end up with a better total number of people converting!

Trending on Indie Hackers
Finally! Webhooks as a Service 23 comments How many sales did you make on #GumroadDay? 18 comments Flowrite launch on Product Hunt ✏️ 13 comments 🧐 Lessons from hiring and firing my co-founder 11 comments "Get Started" Vs "Get Started for Free" - Which do you think converts better? **Updated with results** 10 comments What industry are you in and what made you want to join this specific industry? 9 comments