Keyword match types control how loosely or tightly Google interprets your keywords when deciding whether to show your ad. They don’t determine what you bid on — your keyword list does that. Match types determine how far Google is allowed to stretch from your keywords when matching them to real search queries.
Google Ads has three match types: broad match (widest interpretation), phrase match (moderate interpretation), and exact match (tightest interpretation). Each behaves significantly differently than it did even a few years ago, largely because Google’s AI has changed how queries map to keywords. The 2021 retirement of broad match modifier, the expansion of close variants across all match types, and the introduction of AI Max for Search have collectively reshaped how match types function in practice.
Choosing the wrong match type is the difference between a profitable account and one that burns through budget without generating pipeline. This guide explains how each match type works right now, when to use each one, how they interact with Smart Bidding, and the negative keyword discipline that makes any match type strategy actually work.
How Match Types Work in Practice
Broad Match
Broad match is Google’s default. No special formatting — just type the keyword. Google shows your ads for searches related to your keyword, including queries that share no words in common with your keyword. The system uses your landing page content, other keywords in your ad group, user search history, and location signals to determine relevance.
Syntax: tennis shoes (no symbols)
A broad match keyword like “luxury car” might trigger ads for “fast sports cars,” “premium vehicles,” “high-end automobile dealership,” or even “expensive SUV lease” — queries Google’s AI determines are semantically related to your keyword’s intent.
Adalysis data from nearly 17,000 campaigns shows broad match uses about 25% of keywords but drives roughly 46% of impressions. That means a minority of your keywords controls nearly half your budget when you use broad match. This makes monitoring essential.
The broad match “sticky” problem: Broad match reads your landing page content as a signal for what queries to match. If your landing page says “We don’t offer repairs,” Google may see the word “repairs” and start matching your ads to repair-related searches. Audit your landing pages for words that could trigger unwanted broad match expansion.
Phrase Match
Phrase match shows your ads when searches include the meaning of your keyword, even with additional words before or after. Word order matters when changing it would change the meaning.
Syntax: “tennis shoes” (quotation marks)
A phrase match keyword “moving services NYC to Boston” matches “affordable moving services NYC to Boston” and “NYC corporate moving services to Boston,” but not “Boston to NYC moving services” — because reversing the direction changes the meaning.
Since Google merged broad match modifier into phrase match in 2021, phrase match now captures a wider range of queries than it historically did. Some practitioners note it behaves more like the old broad match modifier than like pre-2021 phrase match.
Phrase match generates roughly 24% of conversions and 35% of revenue in a typical account, sitting in a middle position that balances discovery with control.
Exact Match
Exact match shows your ads for searches with the same meaning or intent as your keyword. Despite the name, it hasn’t been literally “exact” since 2019, when Google expanded close variants to include same-meaning queries.
Syntax: [tennis shoes] (square brackets)
An exact match keyword [tissue box] matches “Kleenex box,” “square tissue box,” and “tissue box dispenser” — queries Google interprets as having identical intent, even when the words differ.
Exact match converts approximately 40% of interactions into conversions while accounting for only 27% of total impressions — the best conversion-per-impression ratio of any match type. It delivers the highest click-through rates and the lowest cost per conversion.
The Nesting Relationship
Broader match types capture all the queries that narrower types would, plus additional ones. Exact match queries are a subset of phrase match queries, which are a subset of broad match queries. This means adding the same keyword in all three match types creates redundancy, not additional coverage. When the same query matches multiple keywords, Google selects the most relevant keyword — usually the most restrictive match type with the highest Ad Rank.
Match Types and Smart Bidding: The Critical Connection
Match types don’t operate in isolation — they interact directly with Google’s Smart Bidding algorithms. This interaction is the single most important factor in modern match type strategy.
Broad match needs Smart Bidding. Without automated bidding, broad match floods your campaign with loosely related queries at uniform bids. With Smart Bidding (Target CPA, Target ROAS, or Maximize Conversions), Google adjusts bids for each individual auction based on the likelihood of conversion. High-intent queries get higher bids; low-intent queries get lower bids or no bid at all. This is what makes broad match viable as a primary match type.
Broad match needs conversion volume. Smart Bidding requires data to learn. Broad match performs effectively when a campaign generates 30-50 or more conversions per month. Below that threshold, the algorithm doesn’t have enough signal to distinguish high-value queries from low-value ones. Campaigns with fewer than 30 monthly conversions should start with exact and phrase match until they accumulate enough conversion data to support broad match expansion.
Google’s own data: Advertisers who switch from exact match to broad match in campaigns using Target CPA see approximately 35% more conversions at a similar cost per acquisition. But this result depends on having sufficient conversion history and a trained Smart Bidding model.
The Decision Framework: Which Match Type to Use When
Rather than treating match types as good or bad, treat them as tools that serve different purposes at different campaign stages.
Use Exact Match When:
You have a limited budget and need every click to count. Exact match gives the highest conversion rate and lowest cost per conversion, making it ideal when there’s no room for exploratory spending.
You’re targeting proven high-converting keywords. For terms with documented conversion history, exact match maximizes return by showing ads only for the specific intent you know converts.
You’re running branded campaigns. Your brand name and product names need exact match to capture high-intent navigational traffic without leaking budget into loosely related queries.
You’re establishing baseline performance data. Exact match provides the cleanest performance benchmarks (CPA, CTR, conversion rate) because the keyword-to-query alignment is tightest.
Watch for: Google now marks low-volume exact match keywords as “Inactive” — they won’t trigger ads if Google determines the search volume is too low. Legacy strategies like SKAGs (Single Keyword Ad Groups) with hundreds of exact match variations are effectively dead. If your exact match keywords are going inactive, consolidate into fewer, higher-volume terms or test phrase match.
Use Phrase Match When:
You need more reach than exact match provides but want to preserve the core meaning of your keywords. Phrase match respects the essential concept while allowing variation.
You’re targeting service-specific or location-specific queries where word order changes meaning. “Emergency plumber” needs the “emergency” context preserved — phrase match handles this better than broad match.
You want a middle ground during campaign scaling. Phrase match often serves as the bridge between the control of exact match and the volume of broad match.
Use Broad Match When:
You have 30+ monthly conversions and Smart Bidding is actively learning. Without this data foundation, broad match will waste budget on irrelevant queries.
You want to discover new converting queries you haven’t thought of. Broad match finds long-tail variations and related searches that keyword research tools don’t surface.
You’re running campaigns where reaching new audiences matters more than maximizing efficiency on known queries. Brand awareness, new market entry, and product launches benefit from broad match’s expansive reach.
You have a strong negative keyword list in place. Broad match without negative keyword discipline is how budgets evaporate.
Important in the current landscape: Broad match is the only keyword match type that accesses certain AI-powered placements, including AI Overviews and AI Mode. Exact match and phrase match keywords cannot trigger ads in these surfaces. For advertisers who want visibility in Google’s AI-driven search features, broad match is required.
AI Max for Search: Beyond Match Types
AI Max for Search represents Google’s next evolution beyond traditional match types. When enabled on a Search campaign, AI Max activates three features simultaneously: search term matching (expands reach beyond your keyword list using AI), text customization (Gemini AI generates dynamic headline and description variations), and final URL expansion (Google selects the most relevant landing page automatically).
AI Max effectively turns your keywords into starting signals rather than strict targeting rules — similar in concept to broad match but with additional AI-driven creative and landing page optimization layered on top.
For advertisers already comfortable with broad match + Smart Bidding, AI Max is the natural next step. For advertisers who rely heavily on exact match for control, AI Max represents a significant shift in how much autonomy you give Google’s systems.
The practical implication: the trend in Google Ads is consistently toward broader matching with AI-driven optimization. Match types aren’t going away, but their role is shifting from “I control exactly what queries I appear for” to “I set guardrails and Google’s AI operates within them.”
Negative Keywords: The Discipline That Makes Match Types Work
Every match type strategy depends on negative keywords. Without them, broad match is reckless spending. With them, broad match becomes controlled discovery.
Negative Match Type Behavior
Negative keywords have their own match type system, and it works differently from positive match types:
Negative broad match (no symbols): blocks searches containing all your negative keyword terms in any order. “Free consultation” blocks “free roofing consultation” and “get free consultation” but allows “consultation pricing.”
Negative phrase match (“quotes”): blocks searches containing your negative phrase in order. “Cheap shoes” blocks “buy cheap shoes online” but allows “shoes cheap to expensive.”
Negative exact match ([brackets]): blocks only that specific query. [used car] blocks “used car” but allows “used car dealership.”
Important distinction: negative keywords do not match to close variants, synonyms, or plurals the way positive keywords do. If you add “running” as a negative, it won’t block “run” or “runner.” You need to add each variation separately.
The 14-Day Review Cadence
A structured negative keyword review every 14 days is the minimum discipline for any campaign using phrase or broad match. The process:
- Pull your Search Terms report for the last 14 days
- Sort by spend (highest first) to find where budget is going
- Identify queries that generated clicks but no conversions — or conversions at an unacceptable CPA
- Add irrelevant queries as negative keywords at the appropriate match type level
- Move high-performing queries into your keyword list as exact match
This review is the feedback loop that trains your account over time. Campaigns that skip it consistently underperform campaigns that maintain it.
Universal Negatives
Most accounts benefit from a shared negative keyword list applied across all campaigns. Common universal negatives for commercial accounts include: “free,” “cheap,” “jobs,” “careers,” “salary,” “training,” “internship,” “DIY,” “how to make,” “definition,” “Wikipedia,” “Reddit” (unless you specifically want Reddit-sourced traffic).
Build this list on day one and add to it continuously based on Search Terms report findings.
Common Match Type Mistakes
Running broad match without Smart Bidding. Manual CPC + broad match = uncontrolled spending. Broad match is designed to work with automated bidding that adjusts bids per auction. Without it, every loosely related query gets the same bid.
Running broad match without conversion tracking. If Google can’t see your conversions, Smart Bidding can’t optimize for them. Broad match + no conversion data = Google optimizing for clicks, not outcomes.
Duplicating keywords across match types in the same campaign. Adding “tennis shoes” as broad, phrase, and exact in the same ad group creates internal competition without improving coverage. The broader match type will serve most queries. If you want to test match types against each other, use separate campaigns with separate budgets.
Ignoring the Search Terms report. This is the single most common mistake across all match types. The Search Terms report shows you exactly what queries triggered your ads. Not reviewing it means you have no visibility into what you’re actually paying for.
Setting and forgetting exact match. Exact match keywords that go “Inactive” due to low search volume are silently stopping your ads from showing. Monitor keyword status regularly and consolidate inactive keywords into higher-volume alternatives.
Over-restricting with only exact match. Accounts that use exclusively exact match miss the 60-70% of search queries that don’t precisely match any keyword in their account. Some of those queries convert well — you just never see them.
Frequently Asked Questions
Should I use all three match types in my account?
Most successful accounts use a combination. A common structure: exact match for proven, high-converting terms (maximum efficiency), phrase match for terms where core meaning matters but you want some expansion (balanced reach), and broad match for discovery and volume when paired with Smart Bidding and adequate conversion data. The mix depends on your budget, conversion volume, and tolerance for spending on unproven queries.
How many conversions do I need before using broad match?
Google recommends Smart Bidding have at least 30-50 conversions per month per campaign to optimize effectively. Below that threshold, exact and phrase match give you more predictable results. Accumulate conversion data with tighter match types first, then expand to broad match once your bidding algorithm has enough signal to work with.
Does match type affect Quality Score?
Indirectly. Quality Score components (expected CTR, ad relevance, landing page experience) are evaluated at the keyword level. Exact match keywords tend to produce higher expected CTR because the query-to-keyword alignment is tighter. Higher Quality Score leads to lower CPCs and better ad positions. But Quality Score is one factor among many — a broad match keyword with strong landing page relevance and good ad copy can still achieve competitive Quality Scores.
What happens when the same query matches multiple keywords in my account?
Google selects the keyword with the highest Ad Rank (combination of bid, Quality Score, and expected impact of extensions). In practice, if you have the same keyword as both exact and broad match, the exact match version typically wins for queries that precisely match it, while broad match captures the expanded queries that exact match wouldn’t cover. Running both isn’t wasteful as long as they’re in the same campaign — Google deduplicates within campaigns.
Is broad match safe to use for small budgets?
It can be, with guardrails. Set a conservative daily budget, pair with Target CPA or Maximize Conversions bidding, build a negative keyword list before launching, and review Search Terms reports every 7-14 days. Start broad match in a separate campaign so it doesn’t cannibalize budget from your proven exact match campaigns. If the first 14 days show mostly irrelevant queries, tighten back to phrase match.
Will Google eventually eliminate match types?
The trend points toward broader matching with AI-driven optimization (AI Max for Search is the clearest signal), but Google hasn’t announced plans to remove match types. They’re more likely to make match types less distinct over time — exact match already includes synonyms and paraphrases, phrase match already absorbed BMM, and broad match already uses AI to interpret intent. The practical advice: learn to work with Smart Bidding and negative keywords, because the value of manually restricting match types is decreasing with every platform update.
