Writing for Play versus iOS
Expert: with sources and numbers. Plain: the same, without them, for beginners.
Why the same copy cannot serve both stores. Not because of character counts, but because the indexed text does opposite jobs.
The character limits are the least important difference. Copy fails when it moves between stores because the indexed text is doing a different job.
The difference that matters
On iOS, the indexed text is mostly hidden. A Apple’s text-relevance inputs are title, subtitle, keywords and primary category, and the keyword field is invisible to users. The description is not named as a search input. (App Store search, verified 2026-09-18)
On Play, every indexed field is visible. A Google identifies apps from metadata including title, description and category. (App Discovery and Ranking, verified 2026-09-18) There is no hidden field.
So:
- iOS separates the two jobs. Terms go in the keyword field where nobody reads them. Visible text is free to persuade.
- Play combines them. Every term you want indexed sits in text a human reads. The description must rank and read at once.
That is the whole problem. Everything below follows from it.
What this changes in practice
The description. On iOS, write it for the reader. It is a conversion asset. On Play, it is the main indexed body text and a conversion asset, so it needs natural coverage of how people describe the problem, in prose someone will actually read.
Term repetition. On iOS, repeating a word across fields is waste, the store combines terms across fields (30-04). On Play, B frequency appears to carry some weight, but A Google explicitly warns against “repetition of words” and publishes a keyword-list example of what not to write (Best practices). The answer on Play is natural use, not counting.
Field mapping. There isn’t one.
| iOS | Play | Why it does not transfer |
|---|---|---|
| Subtitle, 30 chars | Short description, 80 chars | Different length, position and job. CONFLICT-003 |
| Keyword field, 100 characters | nothing | No hidden field exists on Play |
| Description, 4,000, not a search input | Full description, 4,000, indexed | Same size, opposite function |
Two mistakes, symmetrical. Writing iOS copy the Play way produces a keyword-dense description that does nothing for iOS search and reads badly. Writing Play copy the iOS way produces a description that reads well and leaves Play’s main indexed field carrying no relevant terms.
The claim that comes from crossing the wires
C “The first 245 characters of the description are crucial.” Asserted for iOS reviewed here, where it is wrong twice over: Apple does not name the description as a search input at all, so no part of it is established as indexed.
Its likely origin is Play, where the visible portion before “Read more” genuinely does matter. But as the text that earns the expand, not as a magic character count. Full history in CONFLICT-002.
How to work with it
Write each store’s copy from the same brief, not from each other’s output. Agree positioning, proof points and priority terms once. Then write twice.
Do keyword research once, apply it twice differently. The terms people search are largely the same. Where they go is not: iOS hides them, Play shows them.
Never translate iOS copy into Play copy or the reverse. Not across stores, and not across languages, 30-05.
Schedule them separately. iOS indexed text batches to the release; Play does not (20-07).
What goes wrong
One copy deck for both stores. Produces text that is wrong for at least one of them.
Assuming length equivalence means field equivalence. Both descriptions are 4,000 characters and they do opposite things.
Carrying a Play habit onto iOS. Keyword-dense descriptions are effortful and, on iOS, inert.
Carrying an iOS habit onto Play. A beautifully written description with no relevant terms leaves Play’s main indexed field empty of meaning.
In short
Every fact on this page in one line, with its evidence tier, source and verification date, 4 A · 1 B · 1 C. Generated from the claim register, so it cannot drift from the page above.
How much room each centre gives you is the least important difference between them. Copy breaks when it moves from one centre to the other because the readable writing itself is doing a genuinely different job in each.
The difference that actually matters
On one side, what the desk reads is mostly hidden. The desk's own stated inputs are the sign, the line beneath it, the hidden filing words, and the floor. And that hidden pocket of filing words is never seen by a shopper at all. The leaflet isn't named as something the desk reads.
On the other side, everything the desk reads, a shopper reads too. The desk works out what a shop is from its own writing: sign, pitch line, leaflet, floor. There's no hidden pocket at all.
So:
- One side keeps the two jobs separate. Filing words go in a pocket nobody reads. The visible writing is free to do nothing but persuade.
- The other side combines both jobs into one. Every word you want the desk to notice sits in writing a real person also reads. The leaflet has to be found and read at once.
That's the whole problem. Everything else follows from it.
What this actually changes
The leaflet. On the hidden-pocket side, write it purely for the reader. It's a persuasion piece. On the other side, it's the desk's main reading material and a persuasion piece at once, so it needs natural coverage of how people actually describe their own problem, in writing someone will genuinely want to read.
Repeating a word. On the hidden-pocket side, repeating a word across pieces of writing is wasted room (the desk already combines separate words on its own. On the other side, frequency does appear to carry some weight) but the centre itself explicitly warns against "repetition of words" and even publishes its own example of exactly what not to write. The right answer there is natural use, not counting.
Matching the pieces up. There isn't a clean match.
The line under a sign on one side is a short phrase in a fixed small spot. The pitch line on the other side is a full sentence in a different spot doing a different job: different length, different position, different purpose, and treating them as the same thing is a common, costly mistake.
The hidden pocket of filing words simply doesn't exist on the other side at all. There's nothing to map it onto.
The two leaflets are the same generous length, but one is read purely by a person while the other is the desk's own main reading material. Same size, opposite function.
Two mirror-image mistakes. Writing hidden-pocket-side copy the other way produces a leaflet dense with repeated words that does nothing for that side's own desk and reads badly to a person. Writing the other side's copy the hidden-pocket way produces a leaflet that reads beautifully and leaves that side's own main reading material carrying no relevant words at all.
The belief that comes from crossing the wires
You'll hear that a fixed early stretch of the leaflet is "crucial" for being found. Asserted about the hidden-pocket side, it's wrong twice over there, the desk on that side never names the leaflet as something it reads at all, so no part of it is established as feeding anything.
Its real origin is almost certainly the other side, where the visible portion before a shopper has to expand really does matter. But as the writing that earns the click, not as a magic length. Worth remembering exactly which centre a piece of advice actually came from before applying it to the other one.
How to work with this
Write each centre's copy from the same brief, not from each other's finished output. Agree your positioning, your proof points and your priority words once. Then write it twice, properly, for each place.
Do your homework on words once, apply it twice, differently. The words people actually search for are largely the same. Where those words belong is not.
Never translate one centre's copy into the other, and never across languages either. Both moves lose the thing that made the original work.
Schedule the two separately. One side's indexed writing batches up behind a rebuild; the other doesn't wait for anything like that at all.
What tends to go wrong
One single copy deck for both centres. Produces writing that's wrong for at least one of them.
Assuming the same length means the same job. Both leaflets might be exactly the same length and still do opposite things.
Carrying a habit from one side onto the other. A leaflet dense with repeated words is effortful, and on the hidden-pocket side, completely inert.
Carrying the opposite habit the other way. A beautifully written leaflet with no relevant words in it leaves the other centre's own main reading material effectively empty.