App Store Optimization gets treated as a keyword exercise. It is really two separate jobs that happen to live on the same page: getting found, and getting tapped. Keywords do the first. Icon and screenshots do the second, and the second is usually where the money is.
A useful way to think about the size of the prize: if your store page converts at 25% and you improve it to 30%, every acquisition channel you own just got 20% cheaper. No new spend, no new features.
Step 1: Know your baseline
Before changing anything, write down where you are. You cannot claim an improvement you cannot measure.
- Impressions, from App Store Connect and Google Play Console
- Product page views
- Conversion rate, both impression to install and page view to install
- Where installs come from: search, browse, referral, or paid
That last split decides your priority. If almost everything comes from search, keywords matter most. If it comes from browse or paid, creative matters most. Teams routinely spend a month on keywords when 80% of their traffic never used the search box.
Step 2: The icon
The icon is the single highest-leverage asset you own. It appears in search results, in browse, in ads, and on the home screen. It is also the thing most studios never test.
What works, consistently:
- One focal element, readable at 48 pixels. Open your icon on a phone at actual size. If you cannot tell what it is, neither can anyone else.
- High contrast against both light and dark backgrounds. Test on a real device in both modes.
- No text, with the rare exception of a number or a single short word that is genuinely the brand.
- Different from your competitors. Open the top ten results for your main keyword and lay the icons out side by side. If yours disappears into the grid, the design has failed regardless of how nice it looks in isolation.
Step 3: Screenshots
Most players decide in the first two screenshots, which are the only ones visible without scrolling. Treat those two as the entire pitch and the rest as supporting evidence.
- Lead with the hook, not a menu screen or a logo splash.
- Caption every screenshot. Short, benefit-led, readable at thumbnail size. A raw gameplay grab without a caption wastes the slot.
- Portrait for portrait games. Landscape screenshots on a portrait game shrink to nothing in search results.
- Localise the captions for your top markets. Localised text consistently outperforms English everywhere.
- Use an app preview video, but make the first three seconds gameplay. Autoplay means a logo animation burns the only attention you get.
Step 4: Title, subtitle and keywords
The two stores work differently and the same copy will not serve both.
App Store. The title (30 characters) and subtitle (30 characters) are indexed, plus a separate hidden keyword field (100 characters). Do not repeat words between fields, since repetition wins you nothing. Do not use commas and spaces in the keyword field; commas alone fit more terms. The store builds combinations across fields, so single words generally beat phrases.
Google Play. The title (30 characters), short description (80 characters) and full description (4000 characters) are all indexed. The short description carries real weight and is also the first thing a human reads. Write it for a person first, then check the keywords survived. Keyword stuffing the long description is both obvious and ineffective.
For research, look at what your closest competitors rank for rather than what has the highest search volume. A term you can realistically reach the top ten for is worth more than one you will never place in.
Step 5: Test, do not guess
Google Play gives you store listing experiments, which are free and genuinely good. Use them. Apple has Product Page Optimization, which allows up to three treatments against the original.
Rules that keep tests honest:
- One variable at a time. Change the icon, or the first screenshot, not both.
- Run to significance, not to the first encouraging day. Early leads reverse constantly.
- Beware seasonality. A test spanning a holiday weekend is comparing two different audiences.
- Keep a log of every test, the hypothesis, and the result. In a year this log is the most valuable ASO asset you own.
Step 6: Ratings and reviews
Rating affects both ranking and conversion, and it compounds. Two things matter more than everything else:
- When you ask. Prompt after a genuine win, a level cleared or a reward earned, never on launch or mid-failure. The same prompt in a different moment produces a completely different rating.
- Reply to negative reviews. It is visible to everyone reading, it often gets ratings updated, and on Google Play it materially affects perception.
Step 7: Keep going
ASO is not a launch task. Competitors change their creative, seasons shift what resonates, and the algorithms move. A sensible cadence is a keyword review each quarter, a creative test each month, and a seasonal icon or screenshot refresh where it fits the game.
The short version
- Record your baseline and find out where installs actually come from
- Fix the icon first and test it against competitors at real size
- Make the first two screenshots carry the whole pitch, with captions
- Write title, subtitle and keywords separately for each store
- Test one variable at a time, run to significance, log everything
- Ask for ratings at the right moment and reply to the bad ones
- Review quarterly, forever
