---
title: External payment links vs in-app purchase (US, 2025+)
slug: external-payment-links-vs-iap
summary: RevenueCat's Dipsea test after the 2025 US ruling — web-only checkout earned $1.96/user vs $2.09 for IAP at a 30% fee (≈ a wash), fewer trials but higher trial conversion and far lower auto-renew-off (3.5% vs 19%); at the 15% fee IAP wins; Skylight saw +100% conversion moving a QR flow from web to Apple.
domain: mobile-apps
tags: [monetization, iap, web-checkout, app-store-fees, experiment, ios]
related: [why-web2app-funnels, web2app-paywall-best-practices, revenuecat-charts-and-experiments, free-trial-toggle-pricing]
sources: [yt:R3gbKaowcq4, yt:AilakPuHdf4]
updated: 2026-10-24
---

# External payment links vs in-app purchase

## TL;DR
- Since a 2025 US court ruling, iOS apps in the **US** can link out to web checkout without Apple's "scare sheet".
- RevenueCat's 4-variant test on its own app Dipsea: proceeds per user **$2.09 IAP-only vs $1.96 web-only** (after 30% Apple fee and ~6% web costs) — about a **6% gap, roughly a wash**.
- At the **15% small-business fee**, sending users to web is a clear loss.
- Web: far **fewer trial starts**, but **higher trial-to-paid**; early renewal signal favours web (**3.5% turned off auto-renew vs ~19%** on App Store).
- Given a choice with **30% off on web**, most users still bought in-app: 68 web vs 132 IAP (75 annual + 57 monthly).
- Recommended: a **hybrid** — some users to web, some IAP — if you pay 30%; stay IAP-only if you pay 15%.

## Details
**Setup (May 2025).** David Barnard explains RevenueCat bought Dipsea to run public experiments. Variants: A = Dipsea's existing React Native paywall; B = same paywall rebuilt in RevenueCat Paywalls (to rule out tooling bias); C = choice — blue IAP button for annual/monthly, red "try for free and save 30%" button to web; D = web-only, identical to B except the trial button opens web checkout.[^1][^2][^3]

**Results.** Proceeds per user after fees: web-only (D) $1.96, IAP-only (A/B) $2.09; the IAP+web choice variant (C) slightly underperformed.[^4] Trial starts dropped sharply in D — consistent with the industry: even with Apple Pay, people start fewer trials on the web — but web trials converted at a much higher rate, partly closing the gap; B had the most paying users overall.[^5] Realised LTV so far: ~$10,000 for B vs ~$6,800 for D — almost exactly the 30% fee difference.[^6] In C, despite the 30% discount, only 68 users went to web vs 75 (annual) + 57 (monthly) in-app — perhaps preference for IAP or path of least resistance; adding a monthly option to the web path is a suggested next test.[^7]

**Renewal signal.** Of annual subs eligible to renew in May 2026, only ~3.5% of web-billed users had turned off auto-renew vs almost 19% on App Store — so web LTV will likely end up higher; final numbers pending.[^8][^9]

**Recommendation.** If still in the Small Business Program (15%), keep everyone on IAP; if paying 30%, use a hybrid.[^10]

**Friction beats fees (Skylight).** Skylight's on-device QR code used to open a web checkout; switching it to Apple IAP lifted that flow's conversion **over 100%**. Mark Ungerer says IAP "pays for itself" through reduced friction in their one good test; the fee only applies to a fraction of subscribers (most subscribe via their own website) and drops to 15% in year two.[^11][^12] Michael Segal's framework: it's **growth vs profit** — easy subscribing grows, avoiding the 30% "tax" juices profit; choose by ambition and profit profile. David adds web isn't free money (more like 20% saved than 30%), and many customers prefer paying through Apple (cancellation is easy).[^13][^14]

## Disagreements / caveats
- One app (Dipsea), US only, early cohort; wording on the web button wasn't optimised (David invites better variants).
- Platform rules are date-sensitive (US injunction in 2025); check current Apple guidelines and the status outside the US.
- Web checkouts in web2app funnels (paid ads → web quiz → checkout) behave differently from in-app link-outs — see [[why-web2app-funnels]].

## See also
- [[web2app-paywall-best-practices]]
- [[people/david-barnard]], [[people/mark-ungerer]]

[^1]: yt:R3gbKaowcq4 @ 00:00 — David Barnard
[^2]: yt:R3gbKaowcq4 @ 00:49 — David Barnard
[^3]: yt:R3gbKaowcq4 @ 02:19 — David Barnard
[^4]: yt:R3gbKaowcq4 @ 03:29 — David Barnard
[^5]: yt:R3gbKaowcq4 @ 05:02 — David Barnard
[^6]: yt:R3gbKaowcq4 @ 08:07 — David Barnard
[^7]: yt:R3gbKaowcq4 @ 06:36 — David Barnard
[^8]: yt:R3gbKaowcq4 @ 09:41 — David Barnard
[^9]: yt:R3gbKaowcq4 @ 11:15 — David Barnard
[^10]: yt:R3gbKaowcq4 @ 12:26 — David Barnard
[^11]: yt:AilakPuHdf4 @ 56:07 — Mark Ungerer
[^12]: yt:AilakPuHdf4 @ 46:27 — Mark Ungerer
[^13]: yt:AilakPuHdf4 @ 48:54 — Michael Segal
[^14]: yt:AilakPuHdf4 @ 48:09 — David Barnard