Most free trials start the day the app is installed. RevenueCat's 2026 report puts Day-0 trial starts at 78% (Productivity) to 89.9% (Business) (RevenueCat, State of Subscription Apps 2026). Adapty reports 90% (Adapty, State of In-App Subscriptions 2026). So in ASO for subscription apps, the product page is the last thing a user reads before the paywall, usually minutes earlier.
Yet most ASO playbooks optimize the install, and an install that never pays still costs onboarding, servers and often an Apple Ads tap. Since March 25, 2026, that trade-off has been visible, because App Store Connect now benchmarks day-35 download-to-paid conversion against your peer group (Apple Developer News, 2026).
The eight sections below run from one metric for the whole page to a test that reads subscribers. Screenshot order, ratings and video live in our install conversion tactics. Paywall design is out of scope.
Key Takeaways
- 78–90% of free trials start on install day (RevenueCat, 2026), so the product page pre-sells your paywall.
- Track payers per 1,000 impressions (conversion rate × D35 download-to-paid × 1,000), not installs.
- Disclose, don't price: tag premium in screenshots; let promoted IAPs and offers carry the price.
- Read every test twice: installs at significance, payers at day 35.
Why should ASO for subscription apps target payers, not installs?
Because payers = impressions × conversion rate × download-to-paid. One page change can push the last two in opposite directions. In 2026, App Store Connect benchmarks both conversion rate and Day 35 Download-to-Paid Conversion against your peer group (Apple, Peer group benchmarks). A variant that lifts installs 10% wins only if download-to-paid falls less than 9.1% (our arithmetic).
Apple defines download-to-paid as "the percentage of first-time downloads or redownloads that were followed by an In-App Purchase within 35 days of download" (App Store Connect Analytics Help, 2026). Our analytics guide explains how to read Apple's download-to-paid benchmark. This post covers the tactics that move it.
Payers per 1,000 impressions (our metric) = conversion rate × D35 download-to-paid × 1,000. It folds the whole page into one number. Both inputs are metrics Apple benchmarks against your peers. A change that raises installs but lowers this number has cost you subscribers, whatever the install chart says.
A worked example: the install winner that loses payers
An illustrative example: 10,000 impressions at a 5.0% conversion rate give 500 downloads and, at 2.0% download-to-paid, 10 payers. Variant B lifts conversion 10% to 5.5% but attracts browsers, and download-to-paid slips to 1.7%: 9.35 payers. Variant C cuts conversion 5% with a qualifying first frame and lifts download-to-paid to 2.3%: 10.9 payers. B wins the install test. C wins the business.
For any install lift a, the break-even drop in download-to-paid is a ÷ (1 + a). Lose less than that and the variant wins.
How much download-to-paid can you lose for more installs?
Treat it as a planning number. Apple's two denominators differ slightly. For a rough peer reference, multiply your peer group's median conversion rate by its median download-to-paid.
Non-paying installs aren't free. They consume ad spend and support. And a paywall the page never mentioned invites one-star reviews. No Apple source says the App Store ranks apps on subscription status, so the cost shows up in revenue and ratings.
What are good download-to-trial, trial-to-paid and download-to-paid rates in 2026?
Median D35 download-to-paid is 2.0% globally, 10.7% for hard paywalls and 2.1% for freemium (RevenueCat, 2026). For D30 download-to-trial, medians run from 4.4% in Gaming to 9.1% in Business, and trial-to-paid medians from 22.2% in Photo & Video to 43.5% in Travel. RevenueCat's State of Subscription Apps 2026 covers 115,000+ apps and $16B+ in revenue.
The paywall model moves download-to-paid more than anything on the product page: hard-paywall apps convert about five times as many downloads to paid by day 35 as freemium apps. So compare yourself with apps that share your category, paywall model and region. Never use an all-app average.
Adapty's 2026 averages across 16,000 apps are 10.9% install-to-trial and 25.6% trial-to-paid (Adapty, 2026). Don't blend the two vendors. Datasets, windows and definitions differ.
Same funnel, different benchmarks: Health & Fitness vs Gaming
Region: user location vs developer HQ
Region moves the numbers as much as category does. By user location, RevenueCat's median D35 download-to-paid is 2.8% for users in North America, where the top quartile is above 6.0%. The median is 2.4% for users in Asia-Pacific and 0.7% for users in India and Southeast Asia. If a storefront expansion drags your blended rate down, that may be storefront mix, not the page.
One trap: RevenueCat also cuts the same metric by developer HQ, where apps from North American teams read 2.6% and apps from teams in India and Southeast Asia 1.4% (RevenueCat, 2026). Both cuts are correct, but they answer different questions. For storefront decisions, quote the user-location cut.
Find the stage that leaks
Apple's benchmark tells you your position. Your subscription platform's stage split shows which stage of the conversion funnel leaks. Low conversion with normal download-to-trial is an install problem. Normal conversion with low download-to-trial means the page attracts people the paywall can't convert: fix screenshots, custom product pages and keywords. Low trial-to-paid is mostly paywall and onboarding. A low D35 with a month-plus trial is a window artifact.
Subscription app screenshots: disclose what's premium, don't price it
Disclose, don't price. Your description, screenshots and previews must "clearly indicate whether any featured items, levels, subscriptions, etc. require additional purchases" (Guideline 2.3.2). Names, subtitles, screenshots and previews "should not include prices, terms, or descriptions that are not specific to the metadata type" (Guideline 2.3.7). Both rules come from Apple's App Review Guidelines, checked September 28, 2026.
On iOS, price transparency is structural. Screenshots show the paid outcome and mark it as premium; promoted in-app purchases, the introductory offer and the paywall carry the price and trial terms. That satisfies both guidelines. It's our reading of Apple's rules, not Apple's wording.
Plenty of live apps put "free trial" in a caption anyway. A reviewer can still cite the rule. Four practices follow:
- Frame 1 qualifies. Lead with the outcome the subscription delivers for the audience you want to pay: "A 16-week plan to your first marathon" beats "Get fit."
- Tag premium with the paywall's own word. "Premium," "Pro" or your plan name, spelled as the paywall spells it, since most users meet the paywall minutes later.
- Match the model. Freemium apps show what's free and what's premium; hard-paywall apps never imply free use. Guideline 3.1.2(a) removes apps that "trick users into purchasing a subscription under false pretenses."
- Give frames 3–5 proof: ratings, outcomes, press. Previews follow the same disclosure rule.
Sizes and the text that gets screenshots rejected are in our screenshot guide; preview your first three frames as searchers see them before you ship. What you can say, and where:
| iOS surface | Say it's a subscription? | Price or trial terms? | Basis |
|---|---|---|---|
| App name, subtitle | Descriptive only | No | 2.3.7 |
| Keyword field | n/a | No pricing terms | 2.3.7 |
| Screenshots, app previews | Yes, required | No | 2.3.2 and 2.3.7 |
| Description, promotional text | Yes (required in the description) | Not named in 2.3.7; keep terms factual and identical to the paywall | 2.3.2; Terms of Use and Privacy Policy links required |
| Promoted IAP name, description | Yes, with the duration | Plan and duration; the store shows the price | Promoting In-App Purchases |
| Paywall (in app) | Full terms | Required; billed amount most prominent | Auto-renewable Subscriptions |
One more rule: Apple says "your app and App Store metadata must include links to your Terms of Use and Privacy Policy" (Apple, Auto-renewable Subscriptions, 2026).
Should your App Store page show the price?
Your App Store page can already show the price. Apple says promoted subscriptions and introductory offers appear on your product page and may also display in search results and on the Today, Games and Apps tabs (Apple, Auto-renewable Subscriptions, 2026). Higher prices don't obviously hurt, either. In RevenueCat's 2026 data, above-average-priced apps show a median download-to-trial of 8.9%, against 4.4% for below-average ones.
RevenueCat's medians by price tier put D35 download-to-paid at 2.8% for above-average-priced apps, 2.0% for average and 1.4% for below-average, with year-one realized LTV per payer of $62.19, $28.75 and $10.69 (RevenueCat, 2026). Tiers mix categories, brands and product quality, so that's correlation, not a pricing experiment. But it undercuts the fear that a higher price kills conversion.
Higher-priced subscription apps convert more downloads, not fewer
Which offer, plan and trial to show
Apple's own pages disagree on one detail. App Store Connect Help says an introductory offer "automatically appears on your app's product page" for eligible users (Apple, Set up introductory offers, 2026). The subscriptions guide says to "promote the offer's Apple In-App Purchase to display it on the App Store" (Apple, Auto-renewable Subscriptions). Promote it, and you satisfy both pages.
The real decision is which plan and offer appear first. In Health & Fitness, 68% of subscriptions sold are annual (RevenueCat, 2026), so if annual is your hero plan, promote it first and show it first on the paywall.
Trial length (Apple allows free trials from 3 days to 1 year) shows on the page too, though it's a paywall decision. RevenueCat finds 17–32-day trials convert at a 42.5% median, against 25.5% for trials of four days or less.
Keep page and paywall consistent. The plan, duration and offer the page shows should lead the paywall too. Apple requires the billed amount to be "the most prominent pricing element" at purchase (Apple, Auto-renewable Subscriptions), so don't lead promotional text with a monthly equivalent if the paywall leads with the annual total. At Apple's standard rate, you keep 70% in a subscriber's first year and 85% after. Small Business Program members keep 85% from day one (what you keep from each renewal).
How do promoted in-app purchases sell the plan before the install?
Promoted in-app purchases show the price before the install. In 2026, they put up to 20 plans on your product page and in search results, and Apple may also feature them on the Today, Games and Apps tabs. People can buy before they have your app: "they'll receive a prompt to download the app first" (Apple, Promoting In-App Purchases).
That makes a promoted subscription the page's most qualified entry point: whoever taps your annual plan has already seen its price. For subscription ASO, promoted in-app purchases are a pre-qualification surface first and a discovery surface second.
Promote the hero plan first, then the alternative plan and the introductory offer. Add a win-back offer for lapsed subscribers, too. Apple shows win-back offers on the Today, Games and Apps tabs and in Subscription settings (Apple Developer News, September 10, 2024).
Name each plan with its duration, as Apple asks. Skip generic names like "100 gems": a name such as "Marathon Coaching, 1 Year" fits inside 30 characters. Our character guide covers Apple's conflicting IAP name limits.
Handle the purchase at launch, as Guideline 2.3.2 requires. Apple's StoreKit docs name PurchaseIntent (StoreKit 2, iOS 16.4 and later) or the SKPaymentTransactionObserver delegate (iOS 11 and later). Use one, never both. A broken handoff turns your best-qualified install into a failed purchase.
Apple's SKProductStorePromotionController can hide or reorder promoted items per device, so returning subscribers see an upgrade instead of the plan they own. Just don't pitch promoted IAPs as a ranking lever. Whether IAP names lift your app's rankings is contested. The spec, which is check C9 in our audit checklist:
| Promoted IAP item | Apple's spec, 2026 |
|---|---|
| How many | Up to 20 at a time |
| Image | 1024 × 1024 PNG or JPEG; not a screenshot; no text overlay; lower-left corner clear |
| Display name | 30 characters (the product page guide says 35) |
| Description | 45 characters (guide: 55) |
| Subscription names | Include the duration; avoid generic names |
| Purchase flow | Can start before download; completes in the app |
Custom product pages: give each audience its own promise
A custom product page changes the screenshots, app previews and promotional text for one audience. In 2026, you can publish up to 70, and assigned keywords let a page "appear in search results for those selected keywords, rather than your default product page" (Apple, Custom Product Pages). Each page gets its own retention data.
For a subscription app, custom product pages are the only native way to see which promise produces payers. App Store Connect reports retention and average proceeds per paying user for each page, and its Product Page filter slices the rest of Analytics. Judge each page on download-to-paid and retention, not conversion alone.
In WWDC25 session 252, Apple's fictional demo app filters its cohorts by a running-themed custom product page. Only 1.3% of those users buy within 35 days, "more than twice as low as your average day 35 conversion" of 3%. Yet "once they subscribe, they tend to stay that way longer than average" (Apple, WWDC25 session 252). A page with a lower payer rate can still be your most valuable one.
Map pages to intents with different willingness to pay: a "marathon training plan" searcher and a "free workout" searcher shouldn't land on the same promise. Research, prioritization and build order belong to the five-phase CPP framework. The subscription layer adds one thing: judging pages by payers.
Carry the promise past the screenshots. Name the paywall's lead plan in the promotional text, and add a deep link (iOS 18 and later) so people land on the matching plan.
A page's data appears after five first-time downloads (App Store Connect Analytics Help, 2026). Apple's documentation says nothing about promoted in-app purchases per page, so assume every page shows the same promoted plans.
Which keywords bring subscribers instead of browsers?
Keywords whose searchers already want the paid outcome bring subscribers. Almost 65% of App Store downloads happen directly after a search (Apple Ads, worldwide 2022 data). In Adapty's July 2026 data, Apple Ads installs start trials at 2.5× the rate of other paid channels and convert to paid at 2.1× the rate (Adapty, 2026).
Keyword choice decides who reaches the page, so rank keywords by payers, not installs. Organic search gives no keyword-level payer data on iOS. Apple Ads does.
We sort keywords into four tiers by intent to pay. Specific outcome terms ("marathon training plan") earn the title, subtitle and a matching custom product page; category heads ("workout app") need a clear premium disclosure on the default page. Test competitor names on Apple Ads first. And skip "free" or "no subscription" modifiers: those searchers are telling you they won't pay.
Build the ranking from that paid data. Apple's attribution record includes campaign, ad group and keyword IDs (Apple, AdServices, 2026), and its token can go to your mobile measurement partner (MMP), which can join the keyword ID to trials and renewals. Sort keywords by payers, move the winners into your metadata, and repeat. The paid-to-organic keyword loop walks through it.
Keep high-value subscription terms on manual bids. Maximize Conversions optimizes installs, and Adapty warns that the bidder "cannot see trials, subscriptions, or retention" (Adapty, July 2026). For the bidding side, see why the auto-bidder can't see trials.
Check the channel mix too. App Store Connect cohorts now split by download source (Apple Developer News, 2026), so compare D35 download-to-paid for Search, Browse and web-referrer cohorts before you shift effort.
How do you test a product page change against trial-to-paid, not installs?
Read every test twice. Apple's Product Page Optimization (PPO) reports conversion rate, improvement, confidence and unique impressions per treatment, but in 2026 its help pages list no retention, proceeds or download-to-paid for treatments (Apple, Product Page Optimization help). So judge install conversion at significance, then the payer rate of the cohort the winner produces, 35 or more days later.
A Product Page Optimization winner is an install winner until its cohort has been read at day 35. As of September 28, 2026, Apple's Analytics filters include product page but not test treatment (Apple, Filters and dimensions), so the payer read compares the post-launch cohort with the baseline.
Apple counts purchases within 35 days of download, but trials can run up to a year. With a trial of a month or more, D35 and your peer position understate payers, so read trial-to-paid in your subscription platform instead. The metric also counts redownloads and any In-App Purchase, and Apple doesn't say whether a $0 trial start counts.
The protocol we use
- Baseline. Trailing 35-day cohorts: conversion rate, D35 download-to-paid and proceeds per download with peer percentiles, plus your subscription platform's stage rates. Compute payers per 1,000 impressions.
- Hypothesis in payers. "Tagging Premium on frame 2 cuts conversion at most 5% and lifts download-to-trial at least 10%." Check it against the break-even line from the first section.
- Run the install read. PPO for organic traffic (our sample size and false-positive rules), custom product pages for paid traffic.
- Apply the winner and freeze paywall, price and onboarding until the read is done.
- Read by download week. Download-to-trial once the first week closes, since most trials start on Day 0. D35 download-to-paid 35 days after that week's last download: day 42.
- Decide. Keep the change if payers per 1,000 impressions is at or above baseline, even if installs fell.
The two-read test timeline
Roll back in any of three cases. Week-one download-to-trial falls more than 10% (relative) while conversion holds. More 1–2★ reviews mention "subscription," "trial" or "charged" (the free App Review Analyzer surfaces them). Or payers per 1,000 impressions is below baseline at day 42.
Once the read is in, turn payer lift into ASO ROI. This loop is the core of our conversion rate optimization work.
Frequently asked questions
How is ASO different for subscription apps?
The goal is payers, not installs. With 78–90% of free trials starting on install day (RevenueCat and Adapty, 2026), the product page sets up the paywall. Subscription ASO adds premium disclosure, promoted subscriptions, intent-matched custom product pages and a day-35 payer read.
What is a good download-to-paid conversion rate for a subscription app?
It depends on paywall model, category and region. RevenueCat's 2026 medians put D35 download-to-paid at 2.0% globally, 10.7% for hard paywalls, 2.1% for freemium, and 2.9% in Health & Fitness against 1.0% in Gaming. For your own position, check App Store Connect's peer benchmark (25th, 50th and 75th percentiles).
Can you put the price or "free trial" in App Store screenshots?
Apple's Guideline 2.3.7 says screenshots and previews shouldn't include prices or terms. Guideline 2.3.2 requires them to show which features need a purchase. So tag premium features in screenshots. Let promoted in-app purchases, the introductory offer and the paywall carry price and trial terms.
Do promoted in-app purchases help App Store rankings?
The ranking effect is unproven. Promoted in-app purchases can appear as their own search results (consumables can't). People can also buy before they download. But whether IAP names lift your app's own rankings is contested. Treat them as a conversion and pre-qualification surface, not a keyword lever.
How long should you wait to judge a product page change on revenue?
Wait at least 35 days after the last download in the cohort you're reading, plus your trial length if it runs a month or more. Read download-to-trial after the first week, since 78–90% of trials start on Day 0 (RevenueCat and Adapty, 2026).
Does this apply to Google Play?
The principle carries over to Google Play. Its Metadata policy bans price and promotional claims like "10% off" in the title, icon and developer name, and its store listing experiments target unique-user install, open or pre-registration clicks, not revenue. So read payers in your subscription platform, and route intent with custom store listings.
The bottom line
In ASO for subscription apps, the product page is the paywall's opening act. Optimize it for the people who'll pay.
- Metric: payers per 1,000 impressions, not installs.
- Tolerance: +10% installs only wins if download-to-paid falls less than 9.1%.
- Screenshots: disclose, don't price.
- Promoted IAPs: the plan you want sold goes first, duration in the name, matched on the paywall.
- Custom product pages: one intent per page, judged on payers, retention and proceeds per paying user.
- Testing: installs at significance, payers at day 35.
Book a free 30-minute call. We'll compute your payers per 1,000 impressions against your peer group and show which lever is leaking. Most teams start with the fixed-price ASO audit.
