A search campaign can report a healthy click-through rate and still waste thousands of dollars on the wrong intent. The usual culprit is not the ad copy. It is the gap between the keyword you added and the searches Google was allowed to match. Google Ads match types determine how wide that gap can be.
For founders and marketing teams, this is not a technical setting to hand off and forget. Match types shape query quality, conversion volume, reporting clarity and the speed at which a campaign learns. Used well, they give Google enough room to find incremental demand while protecting budget for searches that can move revenue.
Key takeaways
- Exact match offers the tightest control, but it does not mean only identical searches can trigger an ad.
- Phrase match protects the meaning of a query while allowing useful variation around it.
- Broad match is an expansion tool, not a default setting for every campaign.
- Negative keywords are the control layer that makes all positive match types more efficient.
- The right match type depends on conversion data, budget, product complexity and how costly an irrelevant click is.
How Google Ads match types work now
Google has three positive keyword match types: broad match, phrase match and exact match. Advertisers can also use negative keywords to prevent ads appearing for unwanted searches.
The old mental model of match types was mostly punctuation: brackets meant exact, quotation marks meant phrase, and no symbols meant broad. The mechanics remain useful, but the practical reality has changed. Google interprets meaning, intent, search context and other auction-time signals. This includes the user’s device, location, language and recent search activity.
As Google’s own Ads Help documentation makes clear, close variants apply across positive match types. Plurals, misspellings, reordered words and searches with the same intent may still trigger an ad. That means a keyword labelled as exact should be treated as high control, not literal control.
This is commercially significant. A retailer selling premium office chairs may want demand around “ergonomic office chair”, but not research-led queries for free design plans, repairs or cheap second-hand furniture. Keyword selection sets the direction. Match types decide the tolerance for interpretation.
The three Google Ads match types compared
| Match type | Example keyword | Best use case | Primary trade-off |
|---|---|---|---|
| Broad match | ergonomic office chair | Finding additional relevant demand at scale | Can spend on loosely related queries without strong controls |
| Phrase match | “ergonomic office chair” | Capturing variations that retain the keyword’s meaning | Less reach than broad match and still requires query reviews |
| Exact match | [ergonomic office chair] | Protecting high-intent terms and clean performance analysis | Can limit volume and may miss valuable adjacent demand |
Broad match: the scale setting
Broad match gives Google the widest latitude to match a keyword with searches it considers related. A broad keyword such as ergonomic office chair could potentially match searches for lumbar support seating, desk chairs for back pain, or office furniture packages.
That flexibility is valuable when a campaign has reliable conversion tracking, enough budget to learn, and a bidding strategy that optimises toward a meaningful business outcome. In these conditions, broad match can identify demand that a manual keyword list would never anticipate.
It is a poor fit when measurement is weak. If the account optimises to page views, form starts or other soft signals, broad match may efficiently find people who complete those low-value actions rather than people who buy. The same problem applies when a business has long sales cycles but does not import qualified leads or revenue outcomes into Google Ads.
A practical rule: do not judge broad match by search relevance alone. Judge it by marginal business value. If broad match drives incremental qualified leads at an acceptable cost, it is working even if some individual queries look less obvious than your core keywords.
Phrase match: the intent guardrail
Phrase match is useful when word meaning matters. It provides room for modifiers before or after a core concept, while retaining the intent of the phrase.
For example, “commercial solar installation” may capture searches such as “commercial solar installation Melbourne” or “commercial solar installation for warehouses”. It is generally better suited to a focused offer than broad match, particularly when the product has a distinct audience, application or margin profile.
Phrase match is often the most practical starting point for newer search campaigns. It can collect enough search-term data to reveal how real buyers describe the problem, without immediately opening the door as widely as broad match. That said, it is not automatically safer. Google’s interpretation of intent can still produce surprises, so regular search-term reviews remain necessary.
Exact match: the precision layer
Exact match is designed for searches that closely align with a keyword’s meaning or intent. It is the right home for proven commercial queries, branded terms and high-value product or service categories where message alignment matters.
Use it to ensure your best ads, landing pages and bidding approach are applied to the queries most likely to produce value. For an eCommerce business, that could mean exact-match campaigns for specific product categories with strong conversion rates. For lead generation, it may mean terms that consistently lead to sales-qualified enquiries rather than generic information requests.
The trap is assuming exact match is automatically the best-performing option. Tight targeting can reduce data volume, make automated bidding less stable and cap profitable growth. Exact match protects efficiency; it does not create scale on its own.
Negative keywords are the budget protection layer
Negative keywords prevent your ads from showing for particular words or phrases. They are one of the most underused levers in paid search because they require judgement, not simply more automation.
A business selling new accounting software might exclude terms such as “jobs”, “training”, “template”, “free download” or competitors it has no strategic reason to target. An online retailer may exclude “second hand”, “repair”, “manual” or “DIY” where those searches cannot lead to a sale.
Do not build a massive negative list from assumptions. A word that is irrelevant in one context can be commercially useful in another. “Free”, for instance, may signal low purchase intent for some brands but a legitimate free-trial query for SaaS. Review search-term reports, identify repeat patterns of waste, then add exclusions at account, campaign or ad-group level according to their relevance.
Negative match types are more literal than positive matching. A negative broad keyword behaves differently from a positive broad keyword, so test carefully before applying broad exclusions across a whole account.
A match-type framework built for measurable growth
The best structure is rarely broad versus exact. It is a controlled portfolio where every match type has a job.
Start by separating keywords according to commercial intent. High-intent, proven queries belong in exact match, where you can assess conversion rate, cost per acquisition and return on ad spend with greater confidence. Use phrase match for qualified variations around those themes. Introduce broad match where conversion tracking is mature, and there is a clear need to discover incremental volume.
Then set the measurement standard before scaling. For lead generation, the formula should move beyond cost per lead:
Cost per qualified lead = ad spend / qualified leads
For eCommerce, look beyond platform-reported revenue where possible and assess contribution margin, repeat purchase behaviour and stock availability. A keyword that appears efficient on revenue can be less attractive once low-margin products, returns and discounting are considered.
Finally, review the search-term report on a deliberate cadence. Weekly reviews make sense for high-spend or fast-moving campaigns. Smaller accounts may need fortnightly or monthly analysis. The goal is not to eliminate every unexpected query. It is to find recurring patterns: wasted spend, emerging commercial language, new negative candidates and themes that deserve their own campaign or landing page.
Common mistakes that distort performance
The first mistake is using only exact match because it feels safe. This can produce neat reporting while starving the account of learning and growth opportunities.
The second is switching on broad match before offline conversions, enhanced conversions or reliable purchase values are established. Automation can only optimise towards the signals it receives. Give it low-quality signals, and it will find more of them.
The third is treating all search terms as equally valuable. A click from “best software for small business” may be useful early research, while “buy accounting software for 20 employees” suggests a very different commercial moment. Campaign structure, ad messaging and landing-page experience should reflect that difference.
The fourth is making match-type decisions in isolation. Bidding strategy, audience signals, creative quality, product feed health and landing-page conversion rate all affect whether broader targeting becomes profitable. Match types are a control mechanism inside a wider performance system.
The practical decision
Use exact match to protect what already works. Use phrase match to capture meaningful variations with manageable risk. Use broad match to find incremental demand when tracking and bidding are strong enough to distinguish a valuable customer from a cheap click. Build negatives from evidence, not fear.
The most effective search accounts do not chase perfect keyword control. They create a disciplined feedback loop between search behaviour, conversion quality and commercial outcomes – then let the evidence decide where Google should have more room to work.
