What is ASO? App Store Optimization from zero
App Store Optimization is the work of making an app findable in store search. Indexation, relevance, measurement and priority - the whole job in one piece.
· 9 min read
App Store Optimization (ASO) is the work of making an app appear on the right terms in store search, and of convincing the person who finds it to install. It is not played with an advertising budget but with the store's own rules: word placement, relevance and user behaviour.
Bad news for anybody looking for a magic word: there isn't one. Good news: the job is largely inventory management. You have a fixed number of characters, a fixed number of image slots, and each of them has a measurable consequence. Knowing the inventory is knowing where to invest.
ASO splits in two: indexation and relevance
Apple first associates your app with a search term, then places it in an order for that term. We call the first indexation and the second relevance, and they are fixed by completely different work.
- Indexation is decided by metadata: title, subtitle, the 100-character keyword field, in-app purchase names and the developer name.
- Relevance is decided by behaviour: people arriving from that search installing the app, opening it, and staying.
The split has a very practical consequence. You cannot rank for a word you were never indexed for; there the problem is metadata and it is solved in an afternoon. On a word you are indexed for but sit thirtieth on, the problem is usually where that word stands in your metadata - again a single edit.
The hard case is the third one: the word is in your title and you still cannot reach the first page. There the text is finished, and it is the screenshots, the first-run experience and conversion that matter.
Fixed inventory: 30 + 30 + 100
The App Store's text inventory is clear. Title 30 characters, subtitle 30, keyword field 100. Nobody ever sees the keyword field; it only enters the search index. That makes it the purest inventory item in ASO: every character either indexes a word or goes to waste.
Apple reads the three fields together and combines words across them. With “habit” in your title and “tracker” in your keyword field you are already indexed for “habit tracker”. Knowing that alone wins back a fifth of the hundred characters.
- Leaving a space after a comma: Apple splits on the comma, the space indexes nothing.
- Repeating a word already in the title inside the keyword field: the same word does not earn weight twice.
- Writing both the singular and the plural of a stem: the store already matches them.
What you measure
There are three numbers in ASO, and none of them means anything unless all three are read together. Rank is where you stand for a word. Popularity is how much that word is searched. Difficulty is how much competition climbing it takes.
On measurement one rule matters more than everything else: an unmeasured value is not zero. “Popularity 0” and “popularity could not be measured” are different things, and showing the second as the first sends somebody to invest in the wrong word. rankcusp never confuses the two; what it could not measure, it leaves blank.
Where to start: the words you are already near
Most beginners start hunting for new keywords. The pool grows, the work gets crowded, nothing changes. Meanwhile the cheapest gain sits in the words you already rank for.
On a word where you rank 11-30 Apple has already associated you with the term; it just did not find you relevant enough for the first page. The difference is usually where the word sits in your metadata. Climbing from 150th is a separate war: it takes content, volume and time.
That is why the same effort returns several times more inside the threshold band. The distance is already closed; what is left is placement.
The order of the first week
If you had to fit the job into a week, this is the order:
- Pick the rivals properly: same category, same country, genuinely appearing in the same searches.
- Build the keyword pool from three channels: the store's search suggestions, rival titles, rival reviews.
- Measure every candidate's popularity and permanently drop the ones below the floor.
- Measure your real rank across what is left and separate out the 11-30 band.
- Sort that band by popularity and make one move for the word at the top.
- Ship the change and measure the result - which words rose, which fell.
Repeated monthly, this loop stops being a project and becomes a habit. rankcusp exists to automate exactly that loop: it builds the pool, measures every day, separates the band, and proposes one concrete move per word.