Exact match vs broad match in Apple Search Ads: when to use each

If you're running Apple Search Ads on your own app, the choice between exact and broad match is one of the few real levers you have. Apple doesn't let you tune creative the way Google or Meta do — your knobs are country, keywords, bids, and your product page. Match type is how you tell Apple how loosely to interpret each keyword you add, and getting it wrong is one of the easiest ways to burn budget on searches that will never convert.

This post is a practical walk-through of what each match type actually does, when to reach for which, and how to structure your campaigns so the two work together instead of fighting each other.

What each match type actually means

On Search Results campaigns, every keyword you add has a match type:

Exact match: your ad is eligible to show for that specific search term and very close variants — plurals, common misspellings, minor reorderings. If you bid on budget tracker exact, you won't show for budget tracking app for couples.
Broad match: Apple decides what counts as a related search. You'll show for synonyms, longer phrases that include your term, related concepts, and sometimes things that feel pretty far from the seed keyword.

There's also Search Match, which isn't really a match type on a keyword — it's a toggle at the ad group level that lets Apple match your ad to relevant searches based on your app's metadata, with no keyword from you at all. Think of it as broad match with the keyword field left blank.

A useful mental model: exact match is a scalpel, broad match is a net, Search Match is a trawler. They catch different things and cost different amounts per useful thing caught.

When to use exact match

Exact is where most of your money should sit once a campaign is mature. Reach for it when:

You already know a search term converts. You've seen installs and ideally revenue from it, and you want to control the bid precisely.
You're bidding on your own brand. Brand terms should almost always be exact (and in their own campaign), so a competitor or a loose synonym doesn't drag your CPT up.
You're bidding on a specific competitor name. You want that exact name, not whatever Apple thinks is "related".
You're protecting CPA on a high-intent keyword. Exact gives you a clean signal: same search, same bid, observable conversion rate.

The big advantage of exact is that the data is interpretable. If expense tracker exact has a 4% conversion rate at a $1.20 CPT, that number means something — it's tied to a real search string you can reason about. Bid changes have predictable effects.

The disadvantage is that exact match never finds you new keywords. Whatever you don't already know about, you'll miss.

When to use broad match

Broad is a discovery tool first and a scaling tool second. Use it when:

You're hunting for keywords you haven't thought of. Real users phrase things in ways you won't predict — typos, two-word combos, niche jargon. Broad surfaces these as search terms in your reports.
You want to cover long-tail variations of a core concept without manually adding fifty keywords.
An exact keyword has high intent but low volume. Running the same term on broad in a separate ad group can catch related phrasing with similar intent.
You're testing a new category or country and want to see what Apple thinks is relevant before you commit to a keyword list.

The catch: broad match will absolutely show your ad on searches that have nothing to do with your app. A meditation app bidding broad on sleep might end up matched to sleep number bed searches. You won't know until you look at the search terms report and see the taps that never installed.

That's why broad needs more babysitting. It's not "set and forget" — it's "set, watch, and prune".

A structure that actually works

One mistake I see indie devs make is dumping exact and broad versions of the same keyword into one ad group. Don't. They compete with each other, the bids get muddled, and the data becomes hard to read.

A cleaner pattern for a single country campaign:

Ad group: Brand — exact. Your app name and obvious variants. Low bids usually win here.
Ad group: Category exact. Your proven, high-intent keywords on exact match.
Ad group: Category broad (discovery). The same seed keywords on broad, with a lower bid and a tighter daily cap. This is where you find new terms.
Ad group: Competitors — exact. Each competitor name as exact, often in its own ad group if you want different bids per competitor.
Ad group: Search Match. No keywords. Let Apple match from your metadata. Lower bid. Pure discovery.

Now you have a flow: Search Match and broad discover terms, you graduate the good ones into the exact ad group at a controlled bid, and you negative-match them out of the discovery ad groups so you're not double-paying.

Using negatives to keep broad honest

Negative keywords are how you stop broad match (and Search Match) from being a money pit. Two habits worth building:

Promote winners, exclude them from discovery. When a search term from your broad ad group starts converting, add it as an exact keyword in your exact ad group, and add it as a negative in the broad and Search Match ad groups. Otherwise both will compete for the same user.
Cut obvious junk. Scan the search terms report weekly. Anything clearly off-topic that got taps — add it as a negative at the campaign level.

Negatives can be exact or broad too. Broad negatives are powerful but blunt; use them when a whole theme is wrong for your app (e.g. a paid productivity app negative-matching free).

What to actually watch

A few signals tell you whether your match-type setup is working:

TTR by ad group. Very low TTR on broad usually means Apple is matching you to searches where your app doesn't look relevant. Either the keywords are wrong or your product page needs work.
Conversion rate by ad group. Exact should generally convert better than broad on the same seed. If broad is way lower, prune the search terms.
CPA / ROAS by keyword (on exact). This is where revenue data earns its keep. Apple won't tell you per-keyword revenue, but if you have purchase attribution wired up (via AdServices and something like RevenueCat), you can connect installs back to keywords and see which exact terms actually pay back.
New search terms per week from broad and Search Match. If discovery has dried up, lower the bid or pause it — you're paying for taps without learning anything new.

A simple weekly loop

If you do nothing else, run this loop:

Pull the search terms report for your broad and Search Match ad groups.
Promote anything that installed (and ideally generated revenue) to exact in its own ad group.
Negative anything that got real tap volume but no installs.
Check your exact ad groups for CPA drift and adjust bids.
Repeat.

This is mechanical, and once you've done it a few times it takes maybe thirty minutes. It's also the kind of thing tools like AdsBuddy are built to surface — reading your ASA data alongside your revenue and proposing the specific promotions, negatives, and bid changes worth making that day, so you just approve the ones that make sense. But the logic is the same whether you do it by hand or not.

Takeaway

Exact and broad aren't alternatives — they're two halves of the same system. Broad and Search Match find candidates. Exact runs the proven ones at controlled bids. Negatives keep them from stepping on each other. Get that loop running and Apple Search Ads stops feeling like a black box and starts behaving like a knob you can actually turn.