Keyword research for the App Store is not the same as keyword research for Google. The rules are different, the data is noisier, and the most common mistake is copying tactics from SEO without understanding how Apple’s algorithm actually works.

This guide walks through a repeatable process for finding keywords that are worth targeting: specific enough to rank on, popular enough to drive installs, and relevant enough to convert.

What makes a keyword worth targeting

Three factors determine whether a keyword is worth putting in your metadata:

  • Search volume: how often people search for it each month
  • Competition score: how hard it is to rank in the top 10
  • Relevance: whether someone searching for that term would install your app

Most developers focus entirely on the first factor. That’s how they end up targeting “photo editor” with a freshly launched app and wonder why it doesn’t rank. Volume without the other two is just noise.

The sweet spot is a keyword with moderate-to-high volume, low-to-medium competition, and strong relevance to your core use case.

Step 1: Start with what your app actually does

Open a blank document and write down every verb and noun that describes your app’s functionality. Be literal first, creative second.

If you have a keyword research app, your list might start with: keywords, ASO, App Store rankings, keyword tracker, keyword research, app optimization.

Then expand outward. What problems does the app solve? What would someone type if they didn’t know your app existed? What adjacent tools do users compare yours to?

This brainstorm typically produces 30 to 50 seed terms. Not all of them will be searchable, but every one of them can lead somewhere useful.

Step 2: Look at what’s actually being searched

Take your seed terms into a keyword research tool. In Marteso, you can enter any term and see its search volume score, competition score, and current ranking apps.

Pay attention to three outputs:

Volume score: Marteso shows this on a 0-100 scale. A score above 30 typically indicates real search activity. Below 10 is often negligible unless it’s a very specific niche term.

Competition score: Also 0-100. Below 30 is generally accessible for a newer app. Above 60 is difficult without an established install base and review count.

Ranking apps: Look at what’s currently ranking in the top 10 for the keyword. If those apps have thousands of ratings and have been in the App Store for years, that’s a signal of how hard the term is to crack regardless of the competition score.

Step 3: Identify the keyword ladder

A keyword ladder is a set of terms you can rank for now that eventually lead to harder, broader terms.

Here’s how it works in practice:

You can’t rank for “photo editor” on day one. But you can rank for “photo editor for social media” or “photo editing with filters” or “photo editor minimal.” Those narrower terms have real volume, lower competition, and users who are specifically looking for what you offer.

When you start ranking for those terms and getting downloads, Apple’s algorithm builds a signal profile around your app. That profile influences how you rank for adjacent, slightly broader terms. Over months, as your install velocity and conversion rate improve, you become competitive on terms you couldn’t touch at launch.

The ladder is about earning positions, not just picking them.

Step 4: Check your competitors’ metadata

Go to the App Store pages for your top 3 competitors. Look at their titles and subtitles. Those are visible keyword choices made by teams who have already done research.

You’re not looking to copy them. You’re looking to find:

  • Terms they’re targeting that you haven’t considered
  • Gaps where they’re weak (low competition around a term they’re dominating by default, not by optimization)
  • Opportunities to differentiate by owning a keyword angle they’ve ignored

In Marteso’s competitor view, you can see which keywords a competitor ranks for and where they rank. This is often faster than building a full list from scratch because you’re starting from validated signals.

Step 5: Fill your 100 characters wisely

The App Store keyword field is 100 characters. Every character counts.

Basic rules that are often ignored:

  • Do not repeat keywords that already appear in your title or subtitle. Apple indexes those separately; duplicating them wastes space.
  • Do not use spaces between words if you can avoid it by using commas. “tracker,keywords,rankings” takes fewer characters than “tracker, keywords, rankings” and functions identically.
  • Do not include your app’s name, category name, or any word Apple already associates with you.
  • Singular and plural of the same word count as variations; you only need one.

A well-structured keyword field typically holds 10 to 15 distinct root terms. That’s your targeting surface. Use it.

Step 6: Track what you ship and iterate

Keyword research does not end when you submit your metadata. It starts there.

After your next App Store release, track how your target keywords perform. Marteso shows rank movement over time. You’re looking for:

  • Keywords climbing in the first two weeks, which confirms Apple is indexing and testing them
  • Keywords stuck outside the top 30, which usually means competition is too strong for your current authority
  • New keywords appearing organically that you didn’t explicitly target, which can reveal related terms worth adding

Update your keyword field at least once per major release cycle. Apps that treat metadata as a configuration file, set once and forgotten, consistently underperform compared to apps that treat it as an active experiment.

What to do next

If this is your first time doing keyword research, start with steps 1 and 2. Generate your seed list, run it through a tool, and find 10 to 15 terms in the medium-volume, low-competition range.

If you’re revisiting an existing app, go straight to your competitor analysis (step 4) and rank movement data. The fastest gains usually come from replacing stale terms that have dropped outside the top 50 with fresher alternatives.

The process is not complicated. The mistake is not doing it at all.