Pulled from Google Trends on 20 August 2026, over a rolling twelve months worldwide: the single most common related query for "app store optimization" is not a tool, a service or a price. It is "what is app store optimization" — the top related query at a relative value of 100, and still rising at +50%. "app store optimization guide" is up +60%.
People are not looking for a vendor. They are looking for the definition.
So here it is, with the awkward part included: App Store optimization is the practice of writing and structuring an App Store listing so that Apple's search matches it to the right queries, and so that people who see it install. The first half has four documented inputs. The second half has none, because it is a judgement about your product.
The one sentence Apple publishes
Almost everything defensible about App Store search comes from a single sentence on Apple's App Store search page:
"Search results are based on a number of factors, including text relevance (matches for your app's title, subtitle, keywords, and primary category), as well as user behavior (downloads, ratings and reviews, and more)."
Read that carefully, because it is more specific than its reputation and it is doing two jobs.
Text relevance names four things: your app's title, subtitle, keywords and primary category. Apple states elsewhere on the same page that "Your primary category and optional secondary category are indexed by our search algorithm."
User behavior names downloads, ratings and reviews, "and more". Apple also says, separately, that "Ratings and reviews appear on your product page and in search results, and can influence how your app ranks in App Store search."
That is the documented surface. Four text fields, plus behaviour Apple describes but does not quantify.
What is not on that list
The description is not on it. Neither are release notes, in-app events, in-app purchase names or your screenshots. Promotional text is a special case — Apple names it only to rule it out, saying on the same page that it "doesn't affect your app's search ranking".
This post does not claim Apple ignores them — Apple's sentence says "including", which leaves room. It says only that Apple names four text fields and none of those is among them. If you have read that the description is your main keyword surface on iOS, that claim does not come from Apple.
The distinction matters because it changes what the description is for. It is not a matcher's input. It is the copy that turns a product-page visit into an install, which is a bigger lever anyway — and most of it is never read.
The fields, and what each costs you
Every cap below is from App Store Connect's own reference pages.
| Field | Limit | Named by Apple as a search input? | Visible to users |
|---|---|---|---|
| App name | 30 characters | Yes | Yes |
| Subtitle | 30 characters | Yes | Yes |
| Keyword field | 100 bytes or 100 characters — Apple says both | Yes | No |
| Primary category | one choice | Yes | Yes |
| Description | 4,000 characters | Not named | Yes |
| Promotional text | 170 characters | Documented as having no effect | Yes |
| What's New | 4,000 characters | Not named | Yes |
Two things in that table catch people out.
The keyword field's size is given two ways by Apple: "up to 100 bytes of content" on App Store Connect's reference page, and "limited to 100 characters total" on the App Store search page. In English the two agree. In Japanese, Greek or Hindi one character costs two to four bytes and they do not. Plan against the smaller reading.
Promotional text is the one field Apple rules out rather than omits: "promotional text doesn't affect your app's search ranking so it should not be used to display keywords." That is a documented negative, which is a stronger fact than an absence.
And Apple tells you not to waste it: "Don't repeat any words any words included in your app name, subtitle, or category." That instruction, typo and all, is on Apple's page. The no-repeat rule people trade as a hard-won secret is documentation.
What ASO cannot do
Being clear about the ceiling is more useful than another tactic.
It cannot tell you what people search for. There is no public App Store search-volume figure — not from Apple, not from anyone. Tools that quote you a volume have modelled it, and the honest ones say so. Apple's autocomplete orders its suggestions, which is a relative signal about popularity and not a volume.
It cannot read a competitor's keyword field. Verified against Apple's public API across 133 result objects: subtitle, keywords and promotionalText are not merely empty, the keys do not exist. Anyone offering you a rival's keyword field is showing you a guess.
It cannot make a bad product rank. Downloads, ratings and reviews are in Apple's own list of factors, and all three are consequences of the app rather than the listing.
It cannot promise a rank. Apple names factors, never weights or a function. Any specific promise about position is invented.
Where the effort actually goes
If you strip out everything undocumented, ASO reduces to a small, unglamorous list:
- Fill the four documented fields properly. Distinct words across name, subtitle and keyword field; no wasted bytes; a category that matches what people would call your app.
- Fix the opening of the description, because that is what a visitor reads before deciding.
- Make the creative do its job — the icon decides whether anyone reads anything, and the first screenshots decide the rest.
- Localize where you already have users, not everywhere.
- Earn ratings honestly. They are in Apple's list. Buying or trading them is a terminated developer account, and Apple says so plainly in the App Review Guidelines.
None of that needs a subscription. The ASO guide walks through it end to end, the listing kit puts every field on one sheet with the caps checked, and keyword research builds candidates from Apple's own autocomplete rather than from a volume figure nobody can source.
What this cannot tell you
- Not the weights. Apple names factors and never their relative importance. Anyone ranking the fields for you is reporting practice, not documentation.
- Not the volume of any term. No public source has it.
- Not whether an edit worked. Rank moved is an observation. Rank moved because of your edit is an attribution, and a competitor's release or an editorial placement moves the same number.
- Not what happens on Google Play. Different store, different documentation, different fields.