To choose ASO keywords, keep only the terms that pass five tests: high popularity, real ranking potential, low competition, high relevance to what your app does, and high efficacy (short words that combine into many phrases). Then run each survivor through yes/no gates until you run out of keywords or metadata space.
Key takeaways
- A right keyword scores on all five criteria: popularity, ranking potential, low competition, relevance and efficacy.
- Relevance is about search intent. Users will not download an irrelevant app, even if it ranks #1.
- For a new keyword, a store search with under 10 results is a strong signal to select it.
- Keep an old keyword only if it already ranks high; otherwise cut it.
- Stop when you run out of good keywords or out of characters, not when the list "feels complete".
What are the five tests of a right keyword?
The five tests are popularity, ranking potential, competition, relevance and efficacy. I call them the Five Keyword Criteria, and a keyword has to pass all five, not three out of five.
- Popularity. Check search volume or audience size. A keyword is a search term used by real humans; its volume is the size of a market.
- Ranking potential. Look for evidence your app can hit the top spots, such as having ranked for it before.
- Competition. Favor terms with fewer apps fighting for the space.
- Relevance. Match the search intent (the need behind the search) to what the app actually does.
- Efficacy. Prefer short, common terms that combine into many longer, long-tail phrases. Long-tail keywords are longer, more specific phrases that fewer people type but that show clearer intent.
The efficacy test is the one most teams skip. On iOS you have 160 indexed characters in total. A short word that appears in five long-tail phrases is worth more than one long phrase that matches a single query.
What does a vanity keyword look like?
A vanity keyword scores high on popularity and collapses on relevance and competition. My usual example: a car rental app targeting "taxi". The volume is there. But someone typing "taxi" wants a ride now, not a rental, and the space is crowded with apps built for exactly that intent.
| Test | Wise keyword | "taxi" for a car rental app |
|---|---|---|
| Popularity | High | High |
| Ranking potential | Evidence it can rank | Weak |
| Competition | Few strong rivals | Crowded |
| Relevance | Matches what the app does | Wrong intent |
| Efficacy | Combines into many phrases | Combines, but into the wrong phrases |
Relevance also protects your conversion rate (CVR, the share of people who saw your app and installed it). CVR is itself a ranking factor, so irrelevant keywords can lift traffic and damage the signal that holds your rank.
How do you decide, keyword by keyword?
Run every keyword through ordered yes/no gates. I call this the Keyword Kill-or-Select Gates. It turns a long scored list into a decision, one term at a time.
- Kill anything without high volume or high relevance. No exceptions for "nice to have" terms.
- If it was in your metadata before, keep it only if it ranks high. Old keywords do not get a free pass.
- If it is new, select it when a store search shows under 10 results. That is a gap you can fill.
- Otherwise, keep only great short combinations with a high conversion rate. Kill the rest.
- Stop when you are out of keywords or out of space.
The gates are deliberately strict because characters are scarce. On iOS the title, subtitle and keyword field share 160 characters, and a word repeated across them adds nothing. On Google Play the title and the 80-character short description carry the most weight. Placement rules differ per store, so choose the keyword first and decide its field second.
Where do teams go wrong when choosing keywords?
Most mistakes come from checking one criterion and ignoring the rest. These are the signs I look for:
- Celebrating #1 rankings without checking whether users would pay. A top rank for a zero-volume or unrelated term is worth nothing.
- Keywords stuck in the top 100, never breaking the top 10. That usually means a guarded, high-value term chased without the muscle to win it.
- Keyword stuffing that makes texts unreadable. Both stores penalize stuffing, and users bounce from it.
- Low-volume terms sitting in the title, the most valuable field you have.
- Old keywords kept without checking their rank.
- Long phrases used where short combinations would win.
When a high-value keyword is out of reach, do not abandon it and do not wait forever. Target easier variants first, rotate what fails, and escalate as your app grows stronger. That method is explained in how often to update ASO keywords.
From the field
The cost of not choosing keywords on purpose builds up quietly. One world-leading app had paused keyword optimization for three years. When we checked, only 0.4% of its mission-critical keywords ranked #1, and 40% sat below the top 50. The brand was strong. The keyword set had simply stopped being chosen.
Does adding a keyword mean you will rank for it?
No. Adding a keyword gets your app indexed for it; ranking depends on everything else. Keyword optimization finds the best keywords your app can rank for under its current health and marketing conditions.
If the barriers are elsewhere, such as download velocity, ratings or technical health, improve those ranking factors instead of rewriting metadata again. App store ranking factors covers what to fix first.
Brand terms deserve their own decision too. Searches for your own name follow demand you already created, so I treat them separately from generic terms; see branded vs non-branded keywords.
A quick checklist before a keyword goes live
- It came from a keyword data source, not a brainstorm.
- It has meaningful search popularity.
- The top results for it look like your app, so the intent matches.
- Few strong apps compete for it, or a store search shows under 10 results.
- It is short, or it shares words with other target phrases.
- It fits into natural copy without stuffing.
- It does not contain another app's brand name, and it is not a "free" or "best" filler word that the algorithm ignores anyway.
Frequently asked questions
What makes a good ASO keyword?
A good ASO keyword passes five tests at once. It has real search popularity, your app has evidence it can rank in the top spots, few apps compete for it, its search intent matches what your app actually does, and it is short and common enough to combine into many longer phrases. A keyword that fails relevance or competition is a vanity keyword, even at #1.
Should I target high-volume keywords?
Only if they also pass relevance and you can realistically rank for them. High-volume terms are often guarded by big brands with bigger ad budgets and faster downloads. Your keyword can sit in the top 100 for months and never break the top 10. Start with easier variants you can win, then escalate to the high-value terms as your app grows stronger.
How do I check keyword competition quickly?
Search the term in the store. For a new keyword, a search that shows under 10 results is a strong select signal. For a keyword you already used, keep it only if it ranks high. Everything else has to earn its place as a short combination with a high conversion rate, or it gets cut.
Is it worth ranking #1 for a low-volume keyword?
Rarely. Ranking #1 for a term nobody searches, or one unrelated to your app, brings no installs and no revenue. Check that the keyword has popularity and that people searching it would actually download and pay. Celebrating a #1 rank without that check is one of the most common keyword mistakes I see.
These five tests sit inside the wider keyword research process. If you want them applied to your app's keyword pool, that is part of our ASO strategy work.
About the author
Binh Dang founded Appkitekt after leading ASO at Flo, Kovalee and Jellyfish, where he was Senior Director of ASO, and starting out as an ASO consultant at Phiture. He wrote The ASO Black Book.
See how Appkitekt runs ASO