A hard paywall buys speed, not loyalty
Published
Rhys Waters compares hard paywalls with freemium from the ad account's side: RevenueCat's 2026 medians show hard-paywall apps convert about five times as many installs to paid within 35 days and earn about eight times the revenue per install by day 60 — while year-one retention is nearly identical. The advantage is speed and signal for paid acquisition, not stickier subscribers; freemium still wins when free users drive growth no ad report can see.
What to take into your own app
Speed and signal are the prize, not loyalty
RevenueCat's medians put download-to-paid within 35 days at 10.7% for hard-paywall apps against 2.1% for freemium, and revenue per install at day 60 at $3.09 against $0.38 — while year-one retention on yearly plans is 27% against 28%. RankBlob takeaway: choose the gate for how fast you learn and get paid, not for keeping subscribers longer.
A hard paywall buys Meta enough events to learn
Meta's learning phase needs about 50 results in a week; the article's worked example has a hard-paywall app reaching 50 trial starts at about £3,600 a month against about £9,000 for a freemium app at the same install cost. RankBlob takeaway: if Meta is your main channel and the budget is modest, check whether your model generates enough deep events before you scale spend.
Freemium's tail falls outside the attribution windows
22.9% of freemium conversions arrive in week six or later, past Meta's 7-day click window and Apple's day-35 SKAdNetwork window, so paid reporting understates freemium ROAS. RankBlob takeaway: judge ads against each other inside the platform, and judge budgets on cohort revenue from your own subscription data.
Know what these notes establish
These notes use the article's reading of RevenueCat's and Adapty's 2026 reports; the medians describe apps that chose each model, not what switching will give yours. RankBlob takeaway: test a switch per install — download-to-paid at day 35 and revenue per install at day 60 — and watch organic installs and ratings, which paid reporting cannot see.
A small experiment to try
RankBlob exercise: take one campaign and work out what a paying subscriber is worth to you, then the cost per install each model needs to break even at the article's median conversion rates. If the numbers only work behind a hard paywall, test the gate on one cohort before committing — and judge it on retention at first renewal, not just conversion.
The original source
Good ideas get better together.
Enjoying the read? Meet indie app builders, compare notes, and share what you’re working on.