What Is Exact Match in Google Ads? Close Variants and When to Use It

Exact match is the Google Ads keyword match type that gives you the tightest control over which searches can trigger your ads. You wrap a keyword in square brackets, and Google treats it as an instruction: show this ad only when the search carries the same meaning. However, modern exact match no longer means identical words, and that gap quietly shapes your budget.

I have managed search advertising since 2012, and this is the misunderstanding I see most often in account audits. In this guide I explain how the match type behaves today, using Google's own help documentation, real close variant examples and the account structure my team uses. If you want to test match types side by side while you read, keep the keyword match type tool open in another tab.

What is exact match in Google Ads?

Exact match is a Google Ads keyword match type that shows your ad on searches with the same meaning or the same intent as your keyword. You write it in square brackets, for example [womens running shoes]. Of the three match types, it offers the most control and the narrowest reach.

Google's official definition is short: ads may show on searches that have the same meaning or same intent as the keyword. The help page also says this match type gives you the most steering over who views your ad, but reaches fewer searches than phrase and broad match. You can read the wording yourself on the Google Ads keyword matching options page.

Put simply, the key word here is "meaning". A bracketed keyword no longer waits for a letter-for-letter query. For example, [womens running shoes] can also match "running shoes for women" or "ladies running trainers". In short, this match type is a tightly drawn cluster of meaning, not a single string of words.

Which searches can a bracketed keyword match?

The concept that sets the boundary is the "close variant". Google applies close variants to every match type by default, and there is no way to opt out. So when you use square brackets, expect to see the following types of queries in your reports.

• Closely related syntax: misspellings, singular or plural forms, stemmings (Google's example is floor and flooring), abbreviations and accents.
• Reordered words: as long as the meaning stays the same, [shoes mens] and [mens shoes] count as one.
• Function words: prepositions, conjunctions and articles can appear or disappear; "shoes for men" is Google's example for [men shoes].
• Implied words: Google's example is [daydream vr headset], which can match "daydream headset" because the query already implies "vr".
• Synonyms and paraphrases: [bathing suits] can match "swimming suits".
• Same search intent: [images royalty free] can match "free copyright images".

In practice, the list follows the categories on Google's close variants help page. Specifically, the last two categories surprise advertisers the most, because they go well beyond spelling. That is also where most of the budget leakage I find in audits begins.

Why is today's exact match different from the old one?

For a long time, this match type really did mean a letter-for-letter match. A single plural or typo was enough to stop the ad from showing. As a result, advertisers built huge lists with every plural and every misspelling as a separate keyword.

Google widened the behaviour step by step over the years. First came misspellings and plurals, then word order and function words, and finally same meaning and same intent. Today, exact match works like a strict filter that relies on meaning rather than on characters.

In practice, the consequence is simple. Legacy accounts with thousands of bracketed variants now carry a lot of dead weight. Writing ten versions of the same idea repeats work the system already does. Moreover, a fragmented structure spreads data thin, so each keyword collects fewer conversions and automated bidding learns more slowly.

That said, the expansion did not make the match type useless. Google interprets the meaning, but you still draw the outer boundary of that interpretation. That boundary remains far narrower than broad match.

How does it compare with phrase and broad match?

All three match types answer the same question differently: how far can a search drift from my keyword and still trigger my ad? The table shows the difference at a glance. I leave the finer points of phrase match for a separate article; here the comparison only serves to position bracketed keywords.

Also, the "data needed" row is easy to overlook. Broad match relies on conversion history, audience signals and landing page content to pick the right queries. Bracketed keywords depend far less on that data. Therefore, in a new account without reliable conversion tracking, they are the safest starting point.

Which keyword wins when a search matches several?

An account often holds more than one keyword that could serve the same search. In that case Google follows a fixed order of priority, and exact match sits at the top. The keyword prioritization page describes it roughly like this:

1. An exact match keyword that is identical to the search term gets priority. Spell corrected searches count too; plurals and synonyms do not.
2. If no identical bracketed keyword exists, phrase and broad keywords (including AI Max) or search themes that are identical to the search term come next.
3. If nothing is identical, AI based ad group prioritization picks the most relevant ad group, judging meaning, context, keywords and landing pages.
4. Finally, if several candidates are still equal, the one that produces the highest Ad Rank wins.

In other words, this rule gives you a strong steering tool. For example, when you place your best selling query in its own ad group as a bracketed keyword, you decide which ad and which landing page that query sees. Still, remember the exceptions: a limited budget, different location targeting or disapproved ads can all break that priority.

How does exact match work alongside Performance Max and AI Max?

As AI powered campaigns spread, this is the question I hear most. Google's answer is clear. If the search is identical to an eligible bracketed keyword in your Search campaign, the Search campaign takes priority over any other broad or phrase keyword and over Performance Max.

AI Max for Search campaigns adds a nuance. Its search term matching uses broad match and keywordless technology to find new queries. According to Google's help page, the standard priority for bracketed keywords still applies in AI Max. However, there is an important caveat: when you turn off the broad match campaign setting or move to AI Max, broad keywords are not treated like bracketed ones in prioritization. Google therefore recommends adding copies of those keywords to the new AI Max campaign.

In its AI Max announcement, Google says advertisers typically see 14% more conversions or conversion value at a similar CPA or ROAS, and 27% for campaigns still mostly using exact and phrase keywords. These are Google's own figures, so I recommend testing them in a controlled experiment before you rely on them. My practice: the most valuable queries stay locked in brackets, while discovery runs on a separate budget in AI Max or broad match.

When is exact match the right choice?

Building every account around brackets is not the answer. Yet in some situations the match type becomes almost essential. These are the scenarios where my team puts it first:

• Proven queries: high intent searches that bring sales again and again in the search terms report.
• Tight budgets: accounts with a small daily budget, where every click needs a clear purpose.
• Brand searches: protecting your own brand name cheaply and with full control.
• Sensitive industries: health, legal and finance, where a wrong match carries a reputation cost.
• Dedicated landing pages: sending traffic to a page you built for one specific query.
• Weak conversion tracking: accounts where Smart Bidding has no reliable data to learn from yet.

All these cases share one goal: less uncertainty. On the other hand, if you are still exploring a market and do not yet know which queries sell, starting with brackets alone can leave you blind.

Where does this match type hold you back?

The biggest cost is invisible: the searches you never appear on. Reports show what matched, not what you missed. Consequently, accounts that lean heavily on brackets can slowly turn into an echo chamber.

Second, low volume can hurt you. A very niche bracketed keyword may become temporarily inactive because too few people search for it. Google shows this as "Low search volume" in the keyword status column. The keyword comes back once enough searches build up, but you collect no data in the meantime.

Finally, scale is a limit too. More control means less reach, so this match type alone rarely delivers the volume a growth plan needs. In addition, Smart Bidding strategies tend to learn better in a wider pool of queries. Knowing these limits is not a reason to avoid brackets. Instead, it helps you give them the right share of the account.

How do you build a keyword list for exact match?

A good list comes from evidence, not guesswork. This is the order I follow:

1. Pick the source: the search terms report of an existing account is the most reliable. In a new account, start with the keyword suggestion tool and Google's Keyword Planner.
2. Split by intent: separate research queries, comparison queries and ready to buy queries.
3. Merge meaning clusters: keep one keyword per meaning and let close variants cover the rest.
4. Check the landing page: every cluster needs a page that truly answers the query.
5. Prepare negatives: exclude unrelated meanings from day one.

I cover buying intent in detail in my guide on finding keywords that drive sales. For assigning keywords to pages, the keyword mapping guide helps. Together, the two articles give your list a solid foundation.

How should you structure ad groups?

For years, single keyword ad groups were the popular answer: one bracketed keyword per ad group. That made sense when close variants were narrow. Today, queries with the same meaning already merge into one keyword, so such fragmentation usually adds work without adding value.

Instead, I recommend themed ad groups. Each group serves one intent, with ad copy written for that intent and a single landing page that answers it. For instance, price queries sit in one group, while "best" and comparison queries sit in another.

There is one exception: the handful of queries that drive most of the revenue. Keeping them in their own ad group lets you use the prioritization rule and write ads for exactly that query. When you write those ads, the responsive search ad builder helps you check how closely your headlines mirror the search.

How do negative exact match keywords work?

Negative keywords behave differently from positive ones, and that difference causes frequent mistakes. Negative keywords do not match to close variants. So a negative exact keyword blocks only the identical query.

Take the example from Google's negative keywords page. With [running shoes] as a negative exact keyword, the results look like this:

• "blue running shoes": your ad can show, because there is an extra word.
• "running shoe": your ad can show, because the form is different.
• "shoes running": your ad can show, because the order is different.
• "running shoes": your ad will not show.

That is why negative lists need plural forms, misspellings and synonyms added by hand. On the positive side the system interprets meaning for you. On the negative side, you have to do that interpretation yourself.

What should you watch in English language accounts?

English looks simple, but it has its own traps. First, there is spelling. Queries like "colour" and "color" or "optimise" and "optimize" often fall under close variants, yet you should confirm this in your own reports rather than assume it.

Second, regional vocabulary matters. "Trainers" and "sneakers", "flat" and "apartment", "holiday" and "vacation" can carry the same intent. Google may match many of these pairs as synonyms. However, if you target several countries in one campaign, a synonym that fits one market can look odd in another.

Finally, watch short, ambiguous keywords. A two word keyword like [apple repair] can mean a phone, a laptop or even a tree service to a system that reads intent. The shorter the keyword, the more room there is for interpretation. So keep your most important bracketed keywords specific enough to leave little doubt.

How do you audit matches in the search terms report?

The only reliable way to see what your bracketed keywords really catch is the search terms report. In your Google Ads account, select Campaigns in the left panel, then open Search terms under Insights and reports. Google lists the official steps on its search terms report page.

In practice, two columns matter most. The "Match type" column shows how closely the search term relates to your keyword. The "Keyword" column tells you which keyword triggered the ad; it is off by default, so you need to enable it in the column settings under Attributes.

I ask three questions in every audit. Does the keyword really match searches with the same intent? Is any close variant drifting away from the meaning? Which shared words appear in variants that never convert? For the last question, the search terms n-gram analyzer saves hours, because it splits thousands of rows into word fragments and shows where the waste piles up.

How does it fit with your bidding strategy?

Google recommends Smart Bidding with all match types. Bracketed keywords do not conflict with that advice. On the contrary, a narrow pool of high intent queries sends a clean signal to strategies like Target CPA or Target ROAS.

Smart Bidding, however, needs conversion data. Without tracking, or with broken tracking, the strategy optimises toward the wrong goal. In that case it is healthier to start with manual bidding or Maximize clicks and switch once enough data has built up.

Also watch the budget. Competition on these queries is usually intense, because every advertiser wants the same high intent searches. So the cost per click can run higher, although conversion rates are often better too. To estimate how many clicks a budget can buy, try the Google Ads budget calculator, and for the wider framework read how to set a Google Ads budget.

How should you use exact match in brand campaigns?

Above all, brand searches are where this match type feels most at home. Keeping your brand name in brackets inside a separate campaign lets you show the right message and the right page to people who already look for you. You also measure brand traffic apart from generic traffic.

Why does that separation matter? Because brand searches almost always convert well. When they share a report with generic campaigns, they make those campaigns look better than they are. As a result, you cannot tell which campaign actually wins new customers.

Competitor brand names, however, are a separate decision. Those queries usually convert poorly, and Google's trademark policy can restrict another company's trademark in your ad text after a complaint. So keep competitor queries out of your own brand campaign and test them with their own budget and goal.

One practical warning: if your brand name is also a common word, close variants can catch unrelated searches. In that case, check the search terms report often during the first weeks and close off unrelated meanings with negatives.

Does it really lower your costs?

The short answer is: not always. Brackets reduce waste, but they do not automatically lower your cost per click. Cost is the combined result of competition, quality and bidding.

In other words, the effect is indirect. When irrelevant clicks drop, the same budget reaches better visitors, so your cost per customer can fall. On the other hand, auctions for high intent queries are more competitive, so the cost per click may rise.

That is why I measure success by cost per conversion and return on ad spend, not by cost per click. The ROAS calculator is a good first step to see profit contribution. For a quick health check of the whole account, the Google Ads audit tool also gives first hints about your match type mix.

Which metrics should you track?

Reading bracketed keywords through a single metric is misleading. Every month I look at four indicators together.

• Search impression share: how often you appear out of all the searches you were eligible for. A low share on high intent queries means lost opportunity.
• Lost impression share (budget and rank): shows what pulls your share down. Budget losses and rank losses need different fixes.
• Conversion rate: the clearest mirror of how well query and page fit together.
• Cost per conversion: tells you whether control turns into profit.

Above all, these four only make sense together. For example, if impression share is low but conversion rate is high, shifting budget toward those queries is the logical move. To compare your click through rate with a benchmark, the CTR calculator is also useful.

What are the most common exact match mistakes?

The same mistakes show up in account after account. These are the most frequent ones:

• Expecting identical matches: assuming "only this keyword" without accounting for close variants.
• Bloated variant lists: adding dozens of spellings of the same idea as separate keywords.
• Treating negatives like positives: assuming a negative bracketed keyword also blocks plurals and synonyms.
• Never opening the report: spending for months without checking search terms.
• Killing discovery: leaving no mechanism in place that can find new queries.
• Ignoring page fit: sending a high intent keyword to a generic homepage.

All these mistakes share one root: treating the match type as a "set and forget" setting. In reality, it is a control tool that needs regular care. Checking search terms at least once a month surfaces most of these problems on its own.

How can you combine it with broad match?

In the most efficient accounts, match types complement each other instead of competing. The structure I use most has two layers: proven queries in brackets, discovery in broad match or AI Max.

In practice, the flow works like this. Broad match finds new queries. When a query starts converting consistently in the search terms report, you move it into the bracketed layer. Thanks to the priority rule, that query then meets the ad and page you chose.

The point to watch is that the two layers should not split each other's data. Keep the number of layers small and shift budget based on performance. I walk through this setup step by step in Google Ads search campaigns for B2B; the logic applies to other industries too.

Who should build your exact match strategy?

Understanding the logic takes an afternoon; building a structure that works takes months. Reading search terms, merging variants, expanding negatives and aligning everything with your bidding strategy needs steady attention.

My team and I always start account audits from the same place: match type mix, search term quality and the accuracy of conversion tracking. Once those three are right, bracketed keywords become the most predictable part of your budget. If you would like to work through your account together, you can see how we work on the Google Ads management page.

If you prefer to handle it yourself, the steps in this guide give you a solid framework. In short: think of exact match as a cluster of meaning, protect your most valuable queries with it, expand negatives by hand and read the search terms report regularly.