Apple shipped the biggest App Store Connect analytics update since launch on March 25, 2026: over 100 new metrics, cohort analysis, proceeds-per-download, and a revamped monetization view. For indie iOS developers this is a free Mixpanel-ish dashboard that Apple just handed us, and most devs have not opened it yet. This post is the opinionated walkthrough of which 10 metrics out of 100+ actually predict business outcomes, how they map to RevenueCat and Superwall, and what to set up this week.
Short version: the new cohort analysis, proceeds-per-download ratio, and the download-to-paid conversion metric are the three new numbers every indie dev should check weekly. Everything else is context. The App Store Connect update is free and better than most paid analytics tools for measuring what actually matters: whether the App Store is sending you profitable traffic.
What changed on March 25, 2026
Apple's update added or reworked roughly these categories:
- Cohort analysis across retention, active usage, and subscription spend.
- Monetization view with proceeds-per-download and lifetime value by acquisition source.
- Sources dashboard with attribution for App Store search, referrals, and web-to-app installs.
- Subscription lifecycle metrics including trial conversion, retention by tier, churn reasons.
- Search terms dashboard with the actual App Store queries that brought installs (where Apple has the data).
- Device-level segmentation (iPhone 15 Pro and newer vs older, iPad, Vision Pro).
- Pre-order and pre-purchase metrics for apps that support them.
Total count is past 100 now, and most of them are new. Apple's own announcement undersold the significance; the ecosystem picked up on it quickly with 9to5Mac, The Apple Post, and Neowin all covering it the same day.
The 10 metrics that actually matter for indie devs
Of the 100+, these are the numbers I check weekly. Every other metric is interesting but not actionable week-over-week.
| Metric | What it measures | Threshold / target |
|---|---|---|
| Proceeds per download | Average revenue per install net of Apple's commission | $0.50 or better for freemium; $2.00+ for paid |
| Trial to paid conversion (7-day) | Percentage of users who start a trial and convert to paid | 35-55 percent for optimized apps |
| Annual subscription retention (Month 12) | Percentage of annual subscribers who renew | 40-60 percent for indie utility apps |
| Monthly subscription retention (Month 3) | Percentage of monthly subscribers still active at 90 days | 25-40 percent is strong |
| Day-1 retention | Users who return the next day | 35 percent or higher |
| Day-7 retention | Users still active after a week | 15-20 percent is solid for utilities |
| Day-30 retention | Users still active after a month | 8-15 percent for subscription apps |
| Impressions to product page conversion | App Store search/browse impressions that click through | 3-5 percent for name-match queries |
| Product page to install conversion | Viewers who install after reaching your page | 25-40 percent for strong pages |
| Refund rate | Percentage of purchases that get refunded | Keep under 3 percent; App Review watches this |
Why proceeds-per-download is the single most important number
If you only have time to check one metric, make it proceeds-per-download. It collapses the entire monetization funnel into one number: how much revenue did your app make per new install last month? If it is climbing, your paywall, pricing, and onboarding are working. If it is flat or falling while install volume rises, you have a quality problem that will hurt you at scale.
Indie-level benchmarks I have seen across my own apps and consulting work:
- Under $0.20 per download: something is broken. Paywall invisible, onboarding killing conversion, or free tier too generous.
- $0.20 to $0.50: working, but thin. Paid ads at typical $8-12 CAC are not viable here.
- $0.50 to $1.50: healthy freemium. Paid ads break even in this band with good creative.
- $1.50 to $4.00: strong. Sustainable paid acquisition.
- $4.00+: elite. Niche apps, high-intent verticals, or tight paywalls right after value delivery.
Cohort analysis: the feature devs waited years for
Before March 2026, you had to pay Mixpanel or Amplitude to get cohort analysis on iOS. Apple now ships it for free. Three questions cohort analysis answers that nothing else does cleanly:
- Are newer cohorts converting better than older ones? This tells you if product changes are improving the funnel. If July 2026 cohorts convert 15 percent better than April 2026 cohorts, your recent paywall A/B test is winning.
- Are retention curves improving over releases? If Day-30 retention is 12 percent for the v4.0 cohort and 9 percent for the v3.8 cohort, you ship the right things.
- Does acquisition source affect cohort quality? Users from Apple Search Ads vs organic search vs web referrals almost always have different retention. Now you can see this without a custom attribution setup.
Mapping App Store Connect to RevenueCat and Superwall
App Store Connect shows you what Apple sees. RevenueCat and Superwall show you what your app code sees. They will disagree by small percentages because of event latency, refunds, and Apple's own trial-to-paid rules. The pattern I use to keep both in sync:
| Metric | App Store Connect | RevenueCat | Superwall |
|---|---|---|---|
| Total proceeds | Truth (Apple reports) | Near-truth (webhook latency) | Close |
| Trial conversion | Apple's version | RevenueCat's version | Paywall-level |
| Cohort retention | Best, now built-in | Limited | Limited |
| Paywall-level conversion | Not available | Entitlements only | Truth (per-variant A/B) |
| Churn reasons | New in March 2026 | Detailed | Partial |
| Refund rate | Apple's version | Reflected via webhooks | Reflected |
| Search term attribution | Apple-only | N/A | N/A |
Rule of thumb: trust App Store Connect for proceeds, cohorts, and attribution. Trust RevenueCat for entitlement state and webhooks. Trust Superwall for paywall-variant-level conversion numbers.
Five things to set up this week
If you are opening the new App Store Connect for the first time after this update:
- Bookmark the new Monetization dashboard. Check it weekly. Make proceeds-per-download your single health metric.
- Create a cohort saved view for each major release. So you can compare release-over-release retention without manual filtering.
- Turn on search terms attribution. Apple added this as an opt-in feature. Will surface queries bringing your installs.
- Export to CSV monthly. Apple retains data for 14 months rolling. Snapshot into your own sheet if you want longer history.
- Compare to your RevenueCat and Superwall dashboards. Small discrepancies are normal; large ones (over 10 percent) indicate instrumentation bugs.
What this replaces (and what it does not)
The new analytics are good enough to replace several paid tools for most indie devs:
- Replaces Mixpanel-for-iOS-conversion for funnel and cohort analysis on install-to-paid pipelines.
- Replaces AppsFlyer or Adjust for simple App Store attribution. Enterprise attribution (paid UA campaigns, cross-device) still needs a real MMP.
- Replaces some of TelemetryDeck for basic engagement. TelemetryDeck still wins for custom events and privacy-first defaults.
- Does NOT replace Crashlytics or Sentry for crash reporting.
- Does NOT replace PostHog or Amplitude for custom event tracking and session replay.
- Does NOT replace RevenueCat or Adapty for subscription state and entitlement management.
Gotchas and limitations
- Data lag. App Store Connect analytics refresh daily at best. For real-time launch monitoring, use RevenueCat webhooks.
- User-level data is not available. Apple aggregates everything at the cohort or population level. You cannot look up a single user's behavior.
- Country segmentation is coarse. Top 20 countries broken out; the rest bucketed.
- Historical backfill only goes to March 25, 2026. Older data is the previous format.
- API access is limited. Most of the new metrics are not yet in the App Store Connect API. Dashboard only for now. Apple says API parity by Q3 2026.
What The Swift Kit ships
The Swift Kit ships with RevenueCat and PostHog pre-wired to match the App Store Connect event model: every subscription event mirrors into your PostHog dashboard with the same cohort structure Apple uses, so you can cross-check the numbers without manual mapping. Superwall integration is a single config flip for paywall-level A/B testing on top.
$99 one-time, unlimited commercial projects. See every integration on the features page or jump to pricing.
Final recommendation
Open App Store Connect today. Create a cohort saved view per release. Make proceeds-per-download your single weekly health metric. If you have been paying for a paid analytics tool purely for iOS conversion funnel data, seriously consider downgrading. Apple just did a quiet favor to every indie iOS developer, and most have not even opened the new dashboard yet.