App Store Ranking Factors, Ranked by Influence

Neither Apple nor Google publishes its ranking algorithm. What follows is drawn from their own developer documentation, the behaviour both stores demonstrate consistently, and what changes when you change something. Ordered by how much difference each one makes.

Last updated 20 August 2026 · by the ASOWeb team

Tier one: the factors that decide most outcomes

1. Keyword presence in the app name. The single most heavily weighted text field on both stores. Thirty characters on each. If your most valuable term is not here, nothing else you do compensates fully.
2. Download velocity. Recent installs relative to your category, not your all-time total. This is why a launch spike lifts rankings and why the lift decays without sustained installs.
3. Conversion rate from impression to install. Both stores can see how many people who saw you chose you. A better icon and first screenshot raise this, and the raised rate feeds back into ranking. Creatives are ranking work, not just conversion work.
4. Rating average and count. Both feed ranking directly and both feed conversion, which feeds ranking again. This compounding is why a modest app with 2,000 well-tended ratings often outranks a better one with 200.

If you only work on four things, work on these four.

Tier two: real, but slower to move

5. Subtitle (iOS) and short description (Play). Heavily weighted second fields. Eighty characters on Play against thirty on iOS, so the same idea needs different phrasing per store.
6. The 100-character keyword field (iOS only). Invisible to users and indexed for search. Play has no equivalent.
7. Full description (Play only). Fully indexed on Play, entirely unindexed on iOS. This is the biggest single difference between the two stores.
8. Retention and engagement (Play especially). Google can see whether people keep your app. A high uninstall rate suppresses ranking in a way Apple's store cannot replicate.
9. Android vitals (Play only). Crash rate and ANR rate are explicit ranking inputs. On Android, fixing stability is genuinely an ASO task.

Tier three: worth doing, small effect

10. In-app purchase names (iOS). Indexed, lightly. Thirty characters each, and most apps leave them as "Pro Subscription".
11. Developer name. Indexed on both. Studios whose name carries a category term have a small standing advantage.
12. Category choice. Being a large fish in a smaller relevant category beats drowning in a bigger one.
13. Localisation coverage. Each localisation is a separate index. Adding en-GB beside en-US is close to free and roughly doubles indexed terms for that language.
14. Review text (Play). The words users write contribute to what Play thinks you are relevant for.

Things that do not affect ranking

• Your iOS description. Not indexed by Apple at all. Write it for the human deciding whether to tap Get.
• Keyword density. Not a concept in either store the way it once was in web search. On iOS, repetition is a wasted character; on Play, two to four natural mentions is sufficient.
• Backlinks to your store page. They can drive traffic, and that traffic can convert, which helps indirectly. The link itself is not a ranking signal the way it is on the web.
• Update frequency by itself. Shipping empty updates does not raise ranking. Shipping updates that improve ratings and retention does.
• Your app's file size, framework or language. No effect, beyond what size does to install completion rates on poor connections.

What you can actually control this week

Ranking factors split into three groups: text you can edit today, behaviour you can influence over months, and things you cannot touch. Spend your time accordingly.

The practical reading: your metadata is the only tier-one-adjacent lever that moves in days rather than months, which is why it is where almost every ASO effort should start.

Frequently asked questions