Read the Education benchmark correctly

The 2.7% temptation

A chart in the RevenueCat report shows a 2.7% median day-35 download-to-paid conversion for Education apps with North American learners.[[1]](https://www.revenuecat.com/state-of-subscription-apps)

It is tempting to read that midpoint as a goal.

A median describes where sampled apps landed in 2025. It does not predict what your app will do, and it does not cause anything.

Who is actually in the chart

RevenueCat analyzed over 115,000 integrated apps with active subscription revenue and minimum install or revenue thresholds.[[1]](https://www.revenuecat.com/state-of-subscription-apps)

The Education bucket combines the store Education and Educational labels, so several product types across two stores sit together.

The target metric period is 2025, and this chart groups apps by learner location in North America.

Percentiles show spread

Percentiles rank the eligible apps on one measure, from weakest to strongest.

In North America, the D35 Education chart places P25 at 1.0%, the median at 2.7%, and P75 at 6.1%.[[1]](https://www.revenuecat.com/state-of-subscription-apps)

The distance between those quartiles is the real message. Top-quartile performance is not what an average new app should expect.

Geography and denominator traps

Two quiet details break comparisons.

Geography in the Education chart is the learner location, while other charts in the report use estimated developer headquarters.[[1]](https://www.revenuecat.com/state-of-subscription-apps)

Store downloads are also the denominator here. If you sell to a web-only visitor, that part of your funnel never enters this chart.

Placing a real app

Picture a language-learning app converting 3.1% at D35 on iOS with North American learners.

It sits above the 2.7% median and below the 6.1% upper quartile.[[1]](https://www.revenuecat.com/state-of-subscription-apps)

That placement says nothing about acquisition cost or learning quality. The chart never measured either one.

Your pre-benchmark checklist

Before quoting any benchmark, state five things: category, store or channel, geography, cohort window, and metric.[[1]](https://www.revenuecat.com/state-of-subscription-apps)

Then design experiments against the spread rather than the midpoint. Moving from P25 toward the median is a different job than chasing P75.

A 2025 sample gives useful context for a later product. It remains a sample statistic, never a plan.

Does the Education chart represent every learning app and every new launch?

• No, it represents a filtered eligible comparison set
• Yes, it represents all learning apps and new launches

Answer: No, it represents a filtered eligible comparison set

No. It covers a filtered set of RevenueCat-integrated apps with active subscription revenue and minimum install or revenue thresholds. It also combines specific Education labels across two stores.

For the North American Education D35 distribution, _____ has about one quarter of sampled apps at or below it, _____ is the midpoint, and _____ has about one quarter of sampled apps at or above it.

Answer: For the North American Education D35 distribution, P25 has about one quarter of sampled apps at or below it, median is the midpoint, and P75 has about one quarter of sampled apps at or above it.

P25, the median, and P75 locate outcomes within the distribution. In this chart, they are 1.0%, 2.7%, and 6.1%, respectively; they do not represent three forecasts for one app.

A language-learning app has a 3.1% D35 download-to-paid conversion for iOS learners in North America. What does the Education chart support concluding?

• It is above the median and below P75
• It is at the top-quartile benchmark
• It has lower acquisition cost than most apps
• It provides better learning quality than most apps

Answer: It is above the median and below P75

The app’s 3.1% result is above the 2.7% median and below the 6.1% P75. The chart does not measure acquisition cost or learning quality, and being below P75 does not mean the app is underperforming.