“Memory games” has a popularity score of 97 in the US App Store. Pi Digits does not rank for it.
That is not a failure. It is the data telling you where the real opportunity is.
Most small iOS apps are optimizing for the wrong target. They add high-popularity terms to their metadata, watch rankings stay flat for weeks, and conclude that ASO does not work. The actual problem is that they are competing for terms where Apple’s algorithm already has strong opinions. Lumosity and Elevate have years of ratings, installs, and engagement signals locking in the top results for “memory games for adults” and “brain training.” A small app cannot buy or shortcut that authority.
But small apps do rank for something. And it is almost never the term with the biggest popularity number.
What Pi Digits ranks for, and what it does not
Here are the head terms Pi Digits tracks in the US App Store, pulled from Marteso today:
- “memory test”, popularity 97, difficulty 78, current rank: none
- “memory games for adults”, popularity 97, difficulty 95, current rank: none
- “math games”, popularity 93, difficulty 89, current rank: none
- “brain exercise”, popularity 92, difficulty 98, current rank: none
None of them rank. These terms have real search volume and entrenched competition. Apple has already decided who belongs in the top results for each one.
Now here is what Pi Digits actually ranks for:
- “pi test” (US), popularity 74, difficulty 100, rank: #7
- “memorize pi” (US), popularity 72, difficulty 33, rank: #16
- “number memory challenge” (US), popularity 79, difficulty 48, rank: #27
- “learn pi” (UK), popularity 74, difficulty 26, rank: #19
- “numero pi” (Spain), popularity 86, difficulty 71, rank: #3
- “memoria numérica” (Mexico), popularity 81, difficulty 49, rank: #23
These rankings exist today. They generate impressions. They belong to Pi Digits, and no established brain-training app is competing for them.
Why difficulty is the number that actually matters
Popularity tells you how often users search for a term. Difficulty tells you how strong the existing competition is. For a small app, the signal that matters is the gap between the two.
Look at “memorize pi”: popularity 72, difficulty 33. That gap, enough search volume for real impressions, low enough competition that a small app can win, is exactly where specific keywords live.
Compare that to “brain exercise”: popularity 92, difficulty 98. The demand is real but every established brain-training app in the store is competing for it. A small app without ratings authority and browse history will not appear in the top 30. Not this week, and probably not this year.
The common mistake is sorting keyword lists by popularity descending. That view shows you a list of terms you cannot win. When you filter instead for terms where difficulty is meaningfully below popularity, you start seeing the terms where a small app can actually place.
What rank #7 on “pi test” means in practice
“Pi test” has a popularity score of 74 and a difficulty score of 100. Pi Digits ranks #7 for it.
That looks like a contradiction. Why does a small app rank #7 for a term with maximum difficulty?
Because difficulty reflects competitive strength from a broad pool of apps that could plausibly be relevant. Pi Digits is the most specifically relevant app for this particular search intent. When a user types “pi test,” they want to test how many digits of pi they know. Pi Digits does exactly that and nothing else. Apple’s relevance model cannot argue with that match.
That specificity creates a moat. Lumosity does not rank for “pi test” because Lumosity is not a pi app. The very narrowness that makes the term lower-traffic also makes it winnable and defensible. A rank #7 on a 74-popularity term is a real impression driver, and it establishes that Apple has connected this app to this specific job. Those are the first two conditions for building a keyword ladder upward over time.
How to find your version of “memorize pi”
Every app has a version of this: a set of terms specific to its core job, where it is unambiguously the best result, and where competition is thin because the query is narrow. The process for finding them is repeatable.
- Start from your app’s exact core job. What does it do, in a single verb-object phrase? For Pi Digits: “memorize pi digits.”
- Generate variants. Different phrasing, different intent angles. “memorize pi,” “pi digits trainer,” “learn pi,” “pi test,” “number memory challenge.”
- Check popularity and difficulty for each variant. You are looking for terms where popularity is above 50 and difficulty is at least 20 points below popularity.
- Check whether you already rank. If you appear in the top 30 and the app genuinely matches the intent, you have a proof keyword. Protect it in your metadata.
- Build from there. Once you rank for proof terms, bridge terms into broader categories become accessible. You earn your way into the ladder rather than forcing it.
The proof terms you find today are not a ceiling. They are the foundation.
One keyword audit you can do today
Pull your current ranked keywords. Sort by rank ascending, your best rankings first. For each term in the top 30, ask: is this term explicitly in my current title, subtitle, or keyword field?
If it is not, you have an alignment gap. The app is ranking for something Apple inferred without explicit metadata support. Adding that term to your keyword field reinforces the ranking rather than building from scratch.
Then look at terms where you rank between 30 and 80. These are bridge opportunities. You have partial signal. A direct mention in the subtitle or keyword field might be enough to move into the top 20.
The ranked terms are your proof set. Do not add new head terms before you have fully reinforced the terms where you already have evidence.