---
title: Reverse trials — converting users who closed the paywall
slug: reverse-trial-freemium-conversion
summary: Give users who dismiss the paywall a free week of Pro, push them to enable Pro features, then warn that Pro is ending and offer a discount — Momingo's free-user conversion went 0.4% → 4.5%.
domain: mobile-apps
tags: [monetization, reverse-trial, freemium, onboarding, retention]
related: [paywall-exit-offers, pay-what-you-want-discount-slider, paywall-in-onboarding, hard-vs-soft-paywalls, what-to-gate-behind-the-paywall, removing-the-free-trial]
sources: [yt:2soA7LyyQ64, yt:bg0_DavI-DE, yt:FWF_AbsR0qc, yt:L-W6tm-Yk5E, yt:-5zgc14t2gc]
updated: 2026-10-24
---

# Reverse trials

## TL;DR
- For freemium apps that won't use a hard paywall.
- Paywall → user starts trial ("Welcome to Pro") **or** hits X ("No problem — enjoy a week of Pro on us").
- Both paths get a **secondary onboarding** that makes users switch on Pro features (give, then take away).
- Before expiry: "Your Pro membership ends in 5 days" → tap → discount offer.
- Momingo: free-user → paid 0.4% → 4.5% (~10x); typical install-to-paid is ~2%.
- Other forms: **Ladder** — 7-day free trial with **no credit card**, product does the selling; **BoldVoice** — closing the paywall unlocks day one (assessment + first plan), trial required from day two.
- A reverse trial won't save a bad product — it exposes it (David Barnard).

## Details
Source: John, founder of **Momingo** (bus/train times app), on Steve's podcast/livestream; John coined the approach.[^1][^2]

**Flow.**
1. *Value-based onboarding* — benefits not features ("get there on time stress-free", "never miss your stop"), social proof on every screen, permissions, a pre-paywall proof screen, the "we'll remind you 2 days before" screen.[^1][^3]
2. *Paywall* — yearly with 7-day trial, "pay as you go" weekly, lifetime.[^4]
3. *Fork:* trial/purchase → "Welcome to Pro"; X → "Enjoy a Pro week on us".[^5]
4. *Secondary onboarding* (same for both): screens that get users to turn on Pro features, e.g. custom notifications marked Pro — the features they'll lose.[^4][^6]
5. *Expiry nudge:* push/in-app "your Pro membership ends in 5 days" → tap → small discount.[^2]

**Result:** users who dismissed the paywall ("freeloaders") converted at 0.4% before and 4.5% after. Steve compares with ~2–3% average install-to-paid (Apple's numbers per Steve) or ~2% of free users.[^2][^6][^7]

**Fit:** works if your app has no high per-use costs (AI apps with compute costs should be careful); can be stacked with a [[pay-what-you-want-discount-slider]] at the end of the reverse trial.[^6][^8]

**Variants from Sub Club.** Ladder (fitness) gives 7 days free without asking for a card and aims to get users through one or two workouts; Greg Stewart would split effort ~90% activation vs 10% deals. David Barnard calls it a reverse trial and warns it exposes products people don't actually value.[^9] BoldVoice: users who X the paywall get the speech assessment, practice set and coach video for day one; the next daily plan requires a 7-day trial — serving skeptics who won't enter a card even for a free trial, alongside users who start the trial in onboarding.[^10][^11] Metered variant: card-scanner apps show the full product with a visible "scans left" counter.[^12]

## Disagreements / caveats
- One app; the 10x figure is free-cohort conversion, not total revenue.
- Requires entitlement logic for a granted (non-store) Pro week.

## See also
- [[paywall-exit-offers]]
- [[hard-vs-soft-paywalls]]

[^1]: yt:2soA7LyyQ64 @ 07:17 — Steve P. Young
[^2]: yt:2soA7LyyQ64 @ 08:08 — Steve P. Young
[^3]: yt:FWF_AbsR0qc @ 08:31 — Steve P. Young
[^4]: yt:FWF_AbsR0qc @ 09:44 — Steve P. Young
[^5]: yt:bg0_DavI-DE @ 03:48 — Steve P. Young
[^6]: yt:FWF_AbsR0qc @ 10:55 — Steve P. Young
[^7]: yt:bg0_DavI-DE @ 04:36 — Steve P. Young
[^8]: yt:FWF_AbsR0qc @ 12:27 — Steve P. Young
[^9]: yt:L-W6tm-Yk5E @ 08:06 — Greg Stewart, David Barnard
[^10]: yt:-5zgc14t2gc @ 24:55 — Anada Lakra
[^11]: yt:-5zgc14t2gc @ 27:26 — Anada Lakra
[^12]: yt:-5zgc14t2gc @ 25:41 — David Barnard