Increasing VIP Conversion by 122% with a Simple Change to Our Free Trial Model
- Hching Lin
- Dec 15, 2021
- 3 min read
Updated: Jun 22
Key results: ✦ Cut non-VIP reading costs by roughly 80% ✦ Increased our VIP conversion rate by more than 120%
Some context before we start
For nearly a decade, Kono offered new users a 14-day free trial. No credit card required. Just sign up, read freely, and then—hopefully—subscribe.
But over time, this once-solid strategy started showing cracks. Two big ones, in fact:
We split profits with publishers for every article read, and non-paying users were reading a lot.
People figured out they could just keep creating new accounts to read for free. Forever.
We needed a change, and we needed it fast.
This case study is all about how I used data to rethink our trial model, test a new idea, and ultimately boost our conversion rates by more than 120%.
Data observation: Trouble in free trial paradise
Once we saw how much non-paying users were costing us, we dug into the data. Turns out, non-VIPs were 5x more expensive than VIP readers.
With only 7–8% of users converting to VIP, we were bleeding money. Our return on investment just wasn’t there.
Redefining the mission
To turn things around, we had two options: we could either cut down on how much non-paying users could read, or get more users to convert to VIP, or both.
So we set two clear goals:
1. Decrease the total number of articles read by non-VIPs 2. Increase conversion from new user to VIP (not just trial user to VIP)
I had three possible ways to fix the funnel
Option 1 - A shorter free trial: I could cut the free trial from 14 days to 7 days or less.
However, I saw from the data that 80% of non-VIP reading happened on day 1 to day 2. Unless we made the trial really short, we wouldn’t see much of a cost drop.
Option 2 - Limit free articles: I could switch from a time-based trial to a content-based one—give new users X free articles instead of unlimited reading.
However, our data showed that the more users read, the more likely they were to subscribe. Cutting them off early could possibly do more harm than good.
👑 Option 3 - Subscribe first: Instead of giving a trial before someone subscribes, we give it after.
Why this made sense:
Users mostly convert on day 1 or after day 14.
My guess was users who subscribed on day 14–17 probably decided on day 1 but waited for the trial to end.
Others may have meant to subscribe but got distracted and never came back.
By flipping the model, we could:
Cut off freeloaders with no intent to subscribe
Nudge serious users to take action sooner
The risk? If my assumptions were wrong, this could tank our conversion rate.
So, we ran an A/B test.
Launching the A/B test experiment
We split new users into two groups:
Path A (Control) - The original flow: 14-day trial starts right after sign-up.
Path B (Experiment) - New flow: 14-day trial starts after you subscribe.
Other than that, everything else was the same—same welcome letter, same onboarding emails. The only change was when the trial kicked in.
Success metrics were defined as:
Non-VIP article reads should decrease 📉
VIP conversion should increase 📈
We quickly saw major success
Sample size
Data was gathered from 2400 new users.
Reading behavior
Non-VIP article reads dropped by a whopping 80%.
Conversion breakdown
Metric | Path A (Old) | Path B (New) | Growth |
|---|---|---|---|
Converted in 30 days | 8.94% | 19.9% | +122% |
Still VIP after 45 days | 7.55% | 14.98% | +98% |
Even though 24.7% of users in Path B canceled during or shortly after the trial, we still ended up with nearly 2x more retained VIPs.
6 months later... did it stick?
We launched the new flow based on the test results, along with an extensive redesign of our subscription page.
▲ Check out the full case study → Redesigning a New Web Subscription Flow for Kono
Six months later, even without a control group, the conversion gains held strong.
Timeframe | Old model | New model | Growth |
|---|---|---|---|
Within 21 days | 7.05% | 20.53% | +191% |
Within 30 days | 8.25% | 20.86% | +152% |
Within 45 days | 9.05% | 21.12% | +133% |
The takeaway
Sometimes the answer isn’t to remove something—it’s to rethink when and how you offer it.
By simply changing when the free trial kicks in, we cut non-VIP reading costs by roughly 80% while simultaneously increasing our VIP conversion rate by more than 120%. I'd call that a win!
