What is the difference between phrase and broad match?

Phrase match requires the meaning of your keyword to be present in the search, possibly with extra words around it. Broad match is looser and can show your ad for searches the system judges related, even if they do not include the keyword's terms, using signals such as your landing page and other keywords.

Does exact match still mean exact?

No. Exact match now covers searches with the same meaning or intent, including close variants such as misspellings, plurals and reordered words. It is still the tightest option, but you should check search terms for exact match keywords as well.

What happened to broad match modifier?

Google retired broad match modifier and folded its behaviour into phrase match some years ago. Existing modified broad keywords were treated as phrase match. If you see older guides recommending it, the advice is out of date.

Should I use broad match for a small budget?

Usually with caution. Broad match explores widely and needs conversion data to steer it. On a small budget with few conversions, exploration can consume most of the spend. Start with phrase and exact, then test broad in a ring-fenced campaign once you have reliable data.

Do match types apply to negative keywords?

Yes, but negative keywords behave differently. They do not match close variants or synonyms in the same way, so you may need to add plurals and misspellings explicitly. Check current Google documentation for the rules on negative broad, phrase and exact match.

Keyword Match Types Explained: Broad, Phrase and Exact

Match types are a control dial

A keyword in Google Ads is not a literal string. It is an instruction about intent, and the match type tells Google how generously to interpret that intent. Tighten the dial and you see fewer, more predictable searches. Loosen it and you see more searches, some of which you would never have thought of, and some of which you would never have wanted.

Over the past several years Google has loosened all three match types. Each now relies on meaning rather than exact wording, so 'exact' no longer means character-for-character identical. Anyone still using rules learned a decade ago is likely to be surprised by their search terms report. Check current documentation, because the definitions continue to evolve.

The three match types today

Match type	Notation	How it behaves	Typical use
Exact	[running shoes]	Searches with the same meaning or intent as the keyword, including close variants	Proven, high-value queries where you want tight control
Phrase	"running shoes"	Searches that include the meaning of the keyword, possibly with words before or after	Mid-level control with room to find related queries
Broad	running shoes	Searches the system judges related, using context such as the landing page and other keywords	Discovery and scale, usually paired with smart bidding

The notation is a convention in the interface and in bulk uploads. What matters is the behaviour. Note that each type matches close variants: misspellings, plurals, abbreviations and reordered words that keep the same meaning. That is why even exact match needs negative keyword hygiene.

Fig. 01 · Hierarchy

Tap to explore

Precision versus reach

01 · Exact
Highest precision, smallest reach.
02 · Phrase
Balanced; the meaning must be present.
03 · Broad
Widest reach; relies on the system's judgement and your data.

Apex first. Moving down the pyramid widens reach and narrows your certainty about what you are buying.

Broad match is not what it was

Historically, broad match was the setting that wasted money. Today it behaves differently, because it reads signals such as the other keywords in the ad group, the landing page and the searcher's recent activity. Paired with automated bidding and good conversion data, it can find profitable queries that tighter match types would miss.

The condition is important. Broad match without smart bidding gives you loose matching with no system checking whether the extra queries convert. Broad match with weak conversion data gives the system the wrong goal to chase. Both combinations reliably produce waste.

Loose matching is a loan of trust to the machine; only good conversion data keeps the loan from going bad.

Choosing a match strategy

Compare scenarios

Three common match strategies

Mostly exact and phrase, with broad used sparingly or not at all.

Suits low budgets and thin conversion data
Easy to read and audit
May miss new queries and cap growth

There is no universally correct choice. The right answer depends on how many conversions you record, how well those conversions reflect real value, how much budget you have for exploration, and how much risk the business can tolerate in an unfamiliar query.

Fig. 02 · Matrix

Tap to explore

Which match strategy fits your account

HighAppetite for explorationLow

Thin or noisyConversion data quality and volume →Rich and reliable

Order: top left, top right, bottom left, bottom right.

Keeping looser match types honest

Whatever you choose, the search terms report is your audit trail. Read it weekly at first, then at least every fortnight. Look for three things: irrelevant queries to add as negatives, high-performing queries to add as their own keywords, and themes that need a separate ad and landing page.

Build negatives in tiers. Account-level for things you never want (jobs, free, competitor names if you choose), campaign-level for intent mismatches.
Watch cost per conversion by match type, not just volume. Broad match that adds conversions at double the cost may still be worth it if margins allow; decide deliberately.
Do not judge too quickly. Automated systems need a learning period. A few days of poor performance is noise; a few weeks is a signal.

A worked illustration

Consider a firm that sells accounting software to small businesses and bids on the keyword 'accounting software'. The searches each match type might reach differ sharply, and so does the work required to manage them. The examples below are illustrative, not a guarantee of how any platform will match on a given day.

Exact [accounting software]: searches such as 'accounting software' or 'software for accounting'. Predictable, but it misses people who describe the need differently.
Phrase "accounting software": adds searches such as 'best accounting software for small business' or 'accounting software with GST invoicing'. Most of these are valuable, some are not.
Broad accounting software: may add searches such as 'how to do bookkeeping', 'invoice app' or 'accountant near me'. Some are excellent prospects; others are looking for a person, a course or a free spreadsheet.

The broad match list contains both the best new opportunities and the worst waste. That is its nature. Whether it pays depends on whether bidding can tell the two apart, which in turn depends on whether your conversion data distinguishes a trial sign-up from a paying customer.

The practical lesson is to decide match types after deciding measurement, not before. An account that can only see form fills should be cautious with broad match. An account that imports paying customers can afford to let it explore.

Common mistakes

Self-diagnostic

0/5

Match type health check

Answer for your main non-brand campaign.

01Do you use broad match only alongside automated bidding?
If yes: Good. That is the combination broad match is built for. If no: Either switch to automated bidding or tighten to phrase and exact.
02Have you read the search terms report in the last two weeks?
If yes: Good. You know what you are paying for. If no: Do it now. Expect to find queries to exclude.
03Is the same keyword running in several match types in the same ad group?
If yes: This is often redundant now; consider consolidating. If no: Good.
04Do you know cost per conversion for each match type?
If yes: Good. You can judge what looser matching adds. If no: Segment your keyword report by match type before changing anything.
05Does your primary conversion reflect real business value?
If yes: Broad match has a meaningful goal to pursue. If no: Fix conversion tracking before loosening match types.

Match types beyond Google

Microsoft Advertising uses similar match types, and imported Google campaigns usually carry them across, though matching behaviour is not identical. Amazon Ads uses broad, phrase and exact match for keyword targeting in sponsored ads, with its own definitions. Social platforms do not use keywords for targeting at all. Whichever platform you use, the same discipline applies: check what the platform actually matched, not what you assumed it would.

How match types fit with the rest of the account

Match types are one lever among several. A clean structure, honest conversion tracking and relevant ads matter more. If the ad and landing page do not match what people search, no match type will rescue the economics. Our Google Ads guide shows where match types sit in the wider build, and Quality Score explains why relevance keeps paying.

Key takeaways

01Match types control how loosely Google interprets your keyword, from exact to broad.
02All three match types now work on meaning and close variants, not literal wording.
03Broad match performs best only with automated bidding and reliable conversion data.
04The search terms report is the audit trail for every match type, including exact.
05Choose a match strategy based on data quality and appetite for exploration, not habit.

Frequently asked

What is the difference between phrase and broad match?: Phrase match requires the meaning of your keyword to be present in the search, possibly with extra words around it. Broad match is looser and can show your ad for searches the system judges related, even if they do not include the keyword's terms, using signals such as your landing page and other keywords.
Does exact match still mean exact?: No. Exact match now covers searches with the same meaning or intent, including close variants such as misspellings, plurals and reordered words. It is still the tightest option, but you should check search terms for exact match keywords as well.
What happened to broad match modifier?: Google retired broad match modifier and folded its behaviour into phrase match some years ago. Existing modified broad keywords were treated as phrase match. If you see older guides recommending it, the advice is out of date.
Should I use broad match for a small budget?: Usually with caution. Broad match explores widely and needs conversion data to steer it. On a small budget with few conversions, exploration can consume most of the spend. Start with phrase and exact, then test broad in a ring-fenced campaign once you have reliable data.
Do match types apply to negative keywords?: Yes, but negative keywords behave differently. They do not match close variants or synonyms in the same way, so you may need to add plurals and misspellings explicitly. Check current Google documentation for the rules on negative broad, phrase and exact match.

Published by Fabulous.Media, a network of specialist marketing agencies. Updated 9 October 2026. Platform features change often; check current official documentation before acting on platform-specific detail.