← All posts
11 min read

Apple Ads Keyword Research: First Campaign Terms

Pick Apple Ads keywords for a first search-results campaign using relevance, Apple suggestions, search popularity, competitor context, and a clean test.

Orange dithered key emoji for Apple Ads keyword research

For a first Apple Ads search-results campaign, start with searches that describe the job your app does. Build a short candidate list from that job, your brand, your category, and relevant competitors. Then check each term against Apple's suggestions, search popularity, the live App Store result page, and the promise on your product page.

The result should be a small list you can explain, not every phrase a tool can generate.

Apple Ads is the current product name, although many people still call it Apple Search Ads. This guide is about choosing keywords for App Store ads, from seed research through the first manual test.

The short workflow

StepQuestionOutput
Define the jobWhat does the user want to accomplish now?One clear search intent
Build seedsWhich words describe the brand, job, category, use case, or competitor?15 to 30 candidates
ExpandWhat does Apple suggest for those seeds?Related App Store searches
ScreenIs the term relevant, searched, winnable, and supported by the page?Keep, hold, or reject
LaunchWhich terms deserve the first controlled spend?Five to ten exact terms
LearnWhich actual searches install and create value?Promote, refine, or exclude

The numbers are practical starting ranges, not Apple rules. A focused list is useful because it makes the first result easier to diagnose.

Write an intent card before opening the keyword tool

Keyword research gets noisy when the app's promise is vague. Write these five lines first:

  • User: who is searching?
  • Job: what are they trying to do?
  • Moment: why are they looking now?
  • Proof: what in the app or product page shows you can help?
  • Value: what outcome would make this user worth acquiring?

For a beginner running app, the card might read: "A new runner wants a week-by-week 5K plan today. The app generates the plan during onboarding and the first screenshots show the schedule. A useful acquisition reaches the plan and starts a trial."

That card gives you a testable intent. "Fitness audience" does not.

Build seeds from five useful buckets

Start with words you already understand. Do not ask a tool to invent the strategy.

Brand

Include the app name and common brand variants. Brand searches are usually the clearest intent, but keep them in their own campaign or ad group so they do not make generic acquisition look stronger than it is.

Core job

Write the plain phrase a customer would use for the main outcome. Examples include "5k training plan," "budget planner," or "duplicate photo remover."

Category and generic terms

Add the accepted category language, but flag terms that describe a market rather than a job. "Finance" may be relevant to a budgeting app without being a sensible first paid keyword.

Use cases and constraints

Add narrower versions shaped by audience, format, or moment: "beginner 5k plan," "shared family budget," or "delete duplicate photos."

Relevant competitors

Apple's own keyword guidance suggests considering popular brands that provide similar services. Treat those terms as a separate competitor test. They have a different intent, auction, and product-page challenge from category terms.

At this stage, duplicates and rough wording are fine. The goal is to expose the vocabulary around one job. Evaluation comes next.

Use Apple's tools as expansion signals

Apple does not currently document a separate Google Ads-style product called Keyword Planner. Its closest built-in research tools are the keyword suggestion flow and campaign Recommendations.

When you add keywords to a Manage Bids search-results ad group, Apple provides recommendations based on the app and genre. Entering your own seed can produce more related suggestions. Apple's interface also displays keyword popularity based on App Store searches.

Once campaigns have run, the Recommendations page adds another layer. Apple says it updates recommendations daily and can show estimated installs, spend, average CPA, and search popularity from 1 to 5 for suggested keywords.

Use those tools for three jobs:

  1. Find the phrasing customers may use.
  2. Compare relative demand inside the App Store.
  3. Discover adjacent terms after real campaign activity exists.

Do not treat a suggestion as approval. A popular phrase can still be wrong for the app, too broad for the page, or uneconomic after the install.

Score every candidate with evidence you can inspect

Use this worksheet before spending:

CheckKeep whenHold or reject when
RelevanceThe app directly completes the searched jobThe relationship needs a long explanation
SpecificityYou can describe one likely customer and outcomeThe term names an entire industry
Search popularityApple shows enough relative demand for the testDemand is absent or too small for the goal
Result-page realityComparable apps appear and your offer has a credible angleResults reveal a different intent or an unwinnable mismatch
Product-page fitThe icon, title, subtitle, and first screenshots support the promiseThe page sells something else
Downstream valueThe intent can plausibly reach activation and revenueIt is likely to create curiosity installs only
StorefrontLanguage, competitors, pricing, and product fit make sense locallyThe term means something different in that market

Relevance is the veto. High popularity does not rescue a term the app cannot satisfy.

The live result page matters because the phrase in your head may not match the market's interpretation. Search the intended country's App Store, inspect the apps that appear, and read their titles and first screenshots. If the page is full of meal planners when you meant project planning, change the seed before paying to learn the same lesson.

AppSprint keyword research table with popularity, difficulty, country, rankings, and competitor apps
The research view should keep the term, storefront, popularity, difficulty, current rank, and competing apps together.

The free App Store keyword checker can help with the same preflight: compare popularity, organic difficulty, and the current competitors before the term receives ad spend.

A worked example: a beginner 5K app

Suppose the product creates personalized beginner 5K schedules. Its first screenshot shows the weekly plan, and onboarding asks for a race date and current ability.

The raw list might look like this:

CandidateWhat the result-page check tells youDecision
5k training planDirect match to the product and pageFirst exact test
beginner 5k planNarrower version of the same jobFirst exact test
5k running planSame outcome with different wordingFirst exact test
couch to 5k planKeep only if the program truly supports beginners from zeroTest or reject on product fit
running planRelevant but wider than the first promiseHold for a separate test
running coachCould imply live coaching rather than a generated planHold until the result page is checked
fitnessBroad category with no clear jobReject
marathon training planValuable intent, but unsupported by this version of the productReject until the product and page support it
Competitor brandDifferent comparison intentSeparate competitor test

Notice what the list does not contain: invented volume estimates, fifty synonyms, or terms kept only because they are popular.

The first three terms test one claim: people searching for a 5K plan will recognize this product as the answer. If that claim fails, the team can inspect the keyword, result-page creative, product page, onboarding, and downstream conversion without five other intents muddying the result.

Pick five to ten terms for a small first test

Five to ten tightly related terms is AppSprint's practical default for a small starting budget. It is not a platform maximum and it is not the right size for every account.

The list is ready when:

  • Every term belongs to the same intent or a clearly labeled brand or competitor bucket.
  • You know why a customer would search it.
  • The intended storefront has enough relative demand to learn something.
  • The product page can prove the promise quickly.
  • You can afford enough taps to evaluate the chosen outcome.

If fifteen terms all meet those conditions, split them by intent instead of placing them in one blended group.

Use exact match for a controlled first answer

Apple supports broad match and exact match keywords. New keywords default to broad match. Apple also notes that a saved keyword's match type cannot be changed; you pause it and add it again with the other type.

For a small first test, choose exact match before saving. Exact match still includes close variants that Apple considers equivalent, but it gives a tighter starting boundary than broad match.

This is a testing recommendation, not a claim that broad match is bad. Broad match is useful for discovering nearby searches. Search Match can discover relevant searches without bidded keywords. Both belong in a separate discovery campaign or ad group where their search terms and budget remain readable.

The Apple Ads match-types guide covers close variants, broad discovery, Search Match, and negative keywords. The campaign-structure guide shows where brand, category, competitor, and discovery groups belong.

Make the product page pass the keyword test

A keyword is a promise before it is a bid target.

Search the term and look at your likely ad next to the organic results. Then ask:

  1. Does the app name or subtitle make the fit understandable?
  2. Does the first screenshot prove the searched outcome?
  3. Does the first product experience deliver the same job?
  4. Can the paywall explain why this outcome is worth paying for?

If a term is strategically valuable but the default page is too generic, consider a custom product page for the Apple Ads intent. Do not create custom creative to disguise a product mismatch. The experience after install still has to keep the promise.

Know the difference between a keyword and a search term

The keyword is the term you bid on. The search term is what the customer actually typed.

They are not always identical. Exact match can include close variants, broad match covers wider related searches, and Search Match does not require a bidded keyword at all.

After launch, review the Search Terms report. For each meaningful query:

  • Promote a relevant discovery winner into a controlled exact group.
  • Add it as a negative in discovery when you want to reduce overlap.
  • Exclude an irrelevant or consistently weak query.
  • Keep waiting when a subscription cohort has not matured enough to judge.

This is how the first hand-built list becomes evidence instead of a permanent opinion.

Judge the term beyond the install

Popularity tells you whether searches exist. It does not tell you whether those users become customers.

Read the first campaign through a small funnel:

SignalWhat it helps diagnose
ImpressionsRelevance, demand, and auction access
Tap-through rateSearch-result promise and creative fit
Install conversionProduct-page persuasion
Activation or trialWhether the app delivers the searched job
Paid conversion and revenueWhether the acquired intent creates business value

For a subscription app, do not label a fresh active trial as a failed keyword. Define the maturity window before launch. The keyword ROAS playbook covers the longer measurement and scaling loop.

Common first-list mistakes

Copying the organic ASO list into ads

Organic metadata and paid keywords can inform each other, but they are not the same decision. An organic term may be useful for discoverability without deserving a paid bid. Use the App Store keyword research guide for name, subtitle, keyword-field, and rank-tracking work.

Starting with the biggest category word

Broad category terms often hide several jobs. Start with the phrase that describes the outcome your app can prove.

Accepting every Apple suggestion

Suggestions reveal language and demand. They do not know your activation quality, subscription economics, or whether a specific product page is ready.

Mixing brand, category, and competitor intent

These buckets behave differently. Separate them before their averages blur the decision.

Ignoring the storefront

Popularity, competitors, language, and conversion can change by country or region. Research and test in the market where the campaign will actually run.

Expanding before the first claim is readable

Broad match and Search Match can find useful queries, but opening discovery immediately can make a small test harder to interpret. Prove or disprove the known intent, then fund discovery separately.

Your first-keyword checklist

Before publishing the campaign:

  • Write the user, job, moment, proof, and value.
  • Build seeds across brand, core job, category, use case, and competitor buckets.
  • Expand them with Apple's keyword suggestions.
  • Check search popularity in the intended storefront.
  • Inspect the live App Store results and competing pages.
  • Reject every term the product or page cannot honestly satisfy.
  • Choose five to ten terms for one small controlled intent.
  • Select exact match before saving if control is the goal.
  • Set a budget and outcome you can evaluate.
  • Review actual search terms and downstream value after launch.

Research Apple Ads keywords in AppSprint with storefront popularity, difficulty, rankings, competitors, spend, trials, revenue, and ROAS in the same workflow.

Sources

Research, analyze, optimize

Find the right keywords, study the competitors already earning attention, and turn that into a stronger App Store page.

Available for macOS 14.6 and above

AppSprint developer 1AppSprint developer 2AppSprint developer 3AppSprint developer 4AppSprint developer 5AppSprint developer 6AppSprint developer 7AppSprint developer 8

Loved by 544 users

Related articles

See all posts →