Google Ads Exact Match Negative Keywords: The Safest Block

An exact match negative like [plumber salary] blocks that exact query and nothing else. It won't block "plumber salary texas", "plumbers salary" or "salary of a plumber". That's why it's the safest negative. It can't hit a search you never looked at.
The cost is reach. An exact negative only saves money if the query comes back. Build them from search terms that repeat, check each one against your own keywords and conversions, and use phrase match for patterns. A pile of exact negatives on one-off queries looks busy and saves almost nothing.
What an exact match negative actually blocks
An exact match negative blocks a search only when the query is exactly the same as the negative: same words, same order, nothing added. If you add [plumber salary] as a negative, your ad stops showing for "plumber salary". It keeps showing for everything else.
Here is what that one negative does and doesn't stop:
| Search query | Blocked by [plumber salary]? |
|---|---|
| plumber salary | Yes |
| plumber salary texas | No, extra word |
| salary plumber | No, different word order |
| plumber salaries | No, plural |
| plumer salary | No, misspelling |
| average plumber salary | No, extra word |
The last three rows trip people up. Google's help documentation says negative keywords don't match close variants. Plurals, misspellings and near-synonyms that your positive keywords happily match are not caught by your negatives. That holds for all three negative match types, not only exact. With exact, though, the gap is easier to see, because the negative covers so little to begin with.
Compare the three types on the same words:
- Exact
[plumber salary]: blocks only "plumber salary". - Phrase
"plumber salary": blocks any query with those words in that order, so "plumber salary texas" and "average plumber salary" too. It does not block "salary of a plumber". - Broad
plumber salary: blocks queries that contain both words in any order, including "salary of a plumber". In practice broad negatives behave unpredictably across match types, which is one reason many account managers rarely use them.
Most of the pages that rank for this topic stop here: a table of match types and a line saying exact is "the most restrictive". True, but that's not the useful part. The useful part is what that restriction means for how you build and maintain a negative list.
Why exact is the only negative that can't misfire
An exact negative can only block a query you have already read. That's the whole case for it. Phrase and broad negatives block whole families of queries, and some of those families include searches you've never seen and would have wanted.
Take a plumbing company that spots a few wasted clicks on "how to fix a leaking tap" and adds the phrase negative "how to". It reads as clean intent filtering: DIY searchers don't hire plumbers. Then a month later someone searches "how to find an emergency plumber tonight". That query is blocked before it ever shows up in a report, so nobody notices it was lost.
The same thing happens with words that look like junk on their own:
"cheap"blocks "cheap boiler repair", which converts for plenty of service businesses."free"blocks "free estimate roof repair", one of the most commercial queries a roofer can get."diy"blocks "diy vs hire a plumber cost", which is a person deciding to hire."jobs"blocks "plumbing jobs done same day" for a company whose ad copy literally says that.
We've covered the test-before-you-block approach for these in question searches. The point here is narrower. A phrase or broad negative is a guess about queries you haven't seen. An exact negative isn't a guess. You read "how to fix a leaking tap", you judged it, you blocked that string. The worst it can do is block one query you were wrong about, and you'd see that query in your own change history.
That's why exact negatives are the right default when:
- You're not the person who knows the business best (an agency in its first month, a freelancer on a new client).
- The account sells something where junk-looking words often show up in buyer searches.
- You're blocking at scale and can't hand-check what each phrase pattern might catch in the future.
- The negative goes into a shared list that many campaigns use, so one bad pattern spreads.
There is one way an exact negative can still hurt you, and it's worth checking every time. If you exact-negate a query that is also one of your own keywords, the negative wins. Adding [emergency plumber] as a negative to a campaign that bids on [emergency plumber] switches the keyword off. It happens more than you'd think, usually when someone blocks a term in one ad group to push it into another, and blocks it at campaign level by mistake. If a keyword suddenly stops getting impressions, check your negatives first.
The hard part: exact negatives only save money on queries that come back
An exact negative saves nothing on the clicks you already paid for. It only stops future clicks on that same string. If the query never shows up again, the negative did nothing.
This is the part most guides skip. They tell you to go through the search terms report and add the bad ones as exact negatives. Fine for the repeat offenders. But in the accounts we audit, the long tail is long: often more than half of the distinct search terms in a 30-day window appear only once. Exact-negating a one-off like "plumber who fixed my neighbour's sink on elm street" costs you a minute and saves you nothing, because nobody will type it again.
Run the numbers on two terms from the same month:
- "plumber apprenticeship" got 9 clicks at $4 each, so $36. It shows up every month. Blocking it exact saves about $36 a month, or roughly $432 a year.
- "plumber who came to my house tuesday" got 1 click at $6. It will probably never appear again. Blocking it exact saves $0.
Both look the same in the report: irrelevant, cost money, zero conversions. Only one is worth an exact negative. The second one isn't harmless, though. One-off junk adds up, and we've written about how low-volume terms add up. But the fix for that spend isn't exact negatives. It's a phrase negative on the pattern behind it, or a tighter keyword, or both.
Three more limits to know before you commit to an exact-only strategy:
- Variants slip through. Because negatives don't match close variants,
[plumber salary]leaves "plumber salaries", "plumbers salary" and "plumber salery" open. If a junk term is recurring and expensive, add its common variants as separate exact negatives, or step up to phrase. - You can only block what you can see. Exact negatives are built from the search terms report, and the report doesn't show every query. Google hides low-volume terms for privacy, so some of your spend sits behind rows you can't read. See search terms hidden by privacy and where the missing cost went.
- Lists get big. Exact negatives don't generalise, so you need many more of them to cover the same ground as a few phrase negatives. In most Search campaigns that's just untidy. In Performance Max it runs into a hard limit (more on that below).
So exact negatives aren't the whole answer. They're the safe tool for one specific job: blocking queries that are known to be bad and known to repeat.
When to use exact, phrase or broad
Use exact for a specific query you've seen and judged. Use phrase for a word or short sequence that's bad in every context you can think of. Use broad rarely, and only when word order really doesn't matter and you've thought about what else it could catch.
A quick decision table:
| Situation | Match type | Example |
|---|---|---|
| A specific recurring junk query | Exact | [plumber apprenticeship] |
| Junk query whose words could appear in a buyer search | Exact | [how to fix a leaking tap], not "how to" |
| A word that's never a buyer in your business | Phrase | "salary", "apprenticeship" for a plumbing company |
| A competitor brand you never want to show on | Phrase | "competitorname" |
| Routing a query to another campaign | Exact | [acme plumbing] in the non-brand campaign |
| Two words that are bad together in any order | Broad, with care | salary plumber |
The routing row deserves a note. Exact negatives are the cleanest way to send a query to the right campaign without blocking anything around it. Add your brand name as an exact negative to a non-brand Search campaign and the brand query goes to the brand campaign, while "acme plumbing reviews" can still match wherever your keywords allow. Brand routing in Performance Max is a separate problem with its own options. We cover it in splitting brand and non-brand in PMax.
If you've decided a pattern really is safe to block with phrase, we have phrase match negative examples you can copy, each with the buyer queries it might catch.
A practical rule many good account managers follow: start exact, promote to phrase on evidence. Block "plumber salary" exact. If next month you see "plumber salary texas", "plumber salary per hour" and "apprentice plumber salary", that's evidence the word "salary" is never a buyer here. Now promote it to a phrase negative "salary" and you can retire the exact ones it covers. You didn't guess about the pattern. The report showed it to you.
How to build exact negatives at scale from the search terms report
Pull search terms over a long enough window, keep only those that repeat, check each against your keywords and conversions, and add them at the narrowest level that does the job. Here's the process step by step.
1. Pull a window long enough to see repeats
Thirty days is enough to read. It's often too short to tell whether a query repeats. Pull 90 days and split it by month, or compare the last 30 days with the 30 before. If you use the API or a script, the source is campaign_search_term_view, which reports search terms for Search, Shopping and Performance Max. Display, Video and App campaigns report none, so there's nothing to build exact negatives from there.
2. Keep terms that recur and cost something
Filter for terms that appeared in at least two of your periods, or that have enough clicks in one period to suggest a pattern. Sort by cost. The top of that list is where exact negatives pay off. The one-click strays at the bottom are better handled with phrase patterns or not at all. Terms with impressions and no clicks are a different question, covered in search terms with no clicks.
3. Remove anything that converted
Before any term goes on the list, check conversions over a window longer than the one you're judging. A query with 12 clicks and no conversions this month might have brought in a $2,000 job two months ago. That's not junk, it's a slow buyer. Also check conversion value, not only conversion count, if you track value.
4. Check against your own keywords
Compare each candidate against the keywords in the campaigns where it'll be added. If the exact string is also a keyword you bid on, either you're about to switch your keyword off or the keyword itself is the problem. Decide which on purpose.
5. Judge intent, not words
This is where speed is tempting and mistakes happen. "Plumber near me open now" and "plumber near me jobs" share four words and have opposite intent. Read each term as a sentence someone typed. Our walkthrough for adding negatives from search terms without guessing goes through the judgment calls in detail. If you're short on time, there's also a 20-minute version.
6. Add variants for the expensive ones
For the few terms that cost the most, add the obvious plurals and misspellings as separate exact negatives. Don't do this for the whole list. It's only worth it where the money is.
7. Place them at the right level
- Ad group level when the query is fine for the account but wrong for this ad group.
- Campaign level when it's wrong for the campaign but might be right elsewhere.
- Shared list when it's wrong for the whole account. Shared lists work for Search and Standard Shopping. They do not work for Performance Max.
8. Write down why
A negative list with 800 exact entries and no notes is impossible to audit six months later. At minimum, keep a sheet with the term, the date, the cost that triggered it, and who added it. When a client asks why "boiler service plan" is blocked, you want an answer that isn't "it looked wrong".
Why this matters beyond the wasted clicks
Every recurring junk query that gets clicks and never converts is data your bid strategy learns from. Blocking recurring junk exactly cleans that signal without risking the good queries next to it. The mechanism is explained in how junk queries skew Smart Bidding.
Exact negatives in Shopping and Performance Max
Standard Shopping takes exact negatives and shared lists just like Search. Performance Max takes exact negatives too, but only at campaign level, up to 10,000 per campaign, and with no shared lists.
Shopping is where exact negatives earn their keep. You don't choose keywords in Shopping, so the product feed decides which queries match, and Google matches broadly. A store selling "running shoes" will see "running shoes for kids", "running shoe repair", "running shoes meme". Product titles share so many words with junk queries that phrase negatives get risky fast. "repair" sounds safe until you sell repair kits. Exact negatives let you remove the specific recurring junk without touching the product language.
Performance Max is where an exact-only strategy hits its limit. Three things to know:
- No shared lists. Every negative has to be added to each PMax campaign on its own. If you run four PMax campaigns, an account-wide exact negative is four entries to add and maintain.
- The 10,000 cap. Ten thousand sounds like a lot, until you build exact negatives one query at a time across a big catalogue with a long tail. Because exact negatives don't generalise, they eat the cap far faster than phrase. If a PMax campaign is heading toward the limit, that's a sign to move the clear patterns to phrase and keep exact for the borderline cases.
- What you can see is limited. Search terms for PMax appear in
campaign_search_term_view, but the PMax insights view in the interface groups queries into categories. Those categories are not queries, and you can't build an exact negative from a category. See PMax search term insights are not queries, and if your negatives seem to be ignored, why PMax negatives sometimes don't work.
Where PPCleaner fits, and where it doesn't
PPCleaner does the repetitive part of this process every night. It reads the search terms each Search, Shopping and Performance Max campaign paid for, judges which were never going to convert, and adds them as negative keywords. It never blocks a term that converted in the last 90 days. It never touches budgets, bids, ads or keywords. Every block can be undone in one click for 90 days, which matters more than it sounds when you're the one explaining a block to a client.
What it can't do: it can't block a query Google never reports. Queries hidden for privacy stay hidden from us too. It can't add shared lists to Performance Max, because Google doesn't allow them there. And it doesn't replace your knowledge of the business. If "cheap" really is your best-converting word, that shows up in your conversion data, and that data is what keeps a term off the block list.
The method in this article works by hand. Pull the terms, keep the ones that repeat, check conversions and keywords, block them exactly, promote to phrase on evidence. The only question is whether you want to do it every week.
Frequently asked
Do exact match negative keywords block plurals and misspellings?
No. Google's documentation says negative keywords don't match close variants. `[plumber salary]` won't block "plumber salaries" or "plumber salery". For expensive recurring terms, add the common variants as separate exact negatives, or use phrase match if the word is never a buyer in any context.
Does word order matter for an exact match negative?
Yes. An exact negative only blocks the query in the exact same order with no extra words. `[plumber salary]` doesn't block "salary plumber". Phrase negatives also respect word order. Only broad negatives ignore it, and they behave unpredictably across match types.
How many exact negatives can a Performance Max campaign take?
Up to 10,000 campaign-level negative keywords, across all match types. Performance Max doesn't accept shared negative keyword lists, so each campaign needs its own negatives. Exact negatives use up that cap faster than phrase because each one covers a single query.
Should I add every irrelevant search term as an exact negative?
No. An exact negative only saves money if the query comes back. One-off queries that never recur save nothing when you block them exactly. Focus exact negatives on terms that repeat and cost money, and handle recurring patterns behind one-off junk with phrase negatives.
Can an exact match negative block my own keyword?
Yes. If you add an exact negative that matches a keyword you bid on in the same campaign or ad group, the negative wins and the keyword stops serving. Check every candidate against your keyword list before adding it, especially at campaign level or in shared lists.
Do exact match negatives work in Shopping campaigns?
Yes. Standard Shopping accepts negative keywords and shared lists, the same as Search. Exact negatives are especially useful there, because product titles share many words with junk queries, so broader phrase negatives can block real product searches.
Clean your search terms. Save time and money.
Every night, PPCleaner reviews the searches your campaigns actually paid for, groups them by intent, and excludes the ones that never convert. No spend threshold to wait for, nothing touched but negatives — and one click undoes any of it.
Start 7 days freeSearch, Shopping and Performance Max · Cancel in one click
Related articles

Negative Keywords for Lead Generation That Cut Junk Leads
In lead gen, the costliest search terms are often the ones that convert: job seekers, vendors and existing customers filling your quote form. Here is how to trace each junk lead to its search term and block it without losing good leads.

How to Find Negative Keywords Before You Launch
A new campaign has no search terms report, so every negative you add before launch is a guess. Here is how to base those guesses on your own site, your competitors' ads and Google autocomplete, and how to catch what the list misses in the first week.

Phrase Match Negative Keyword Examples You Can Copy Safely
Most guides explain phrase match negatives with one tidy example and stop there. Here are real-looking cases where they block buyers by accident, with the search terms before and after and safer versions to use instead.
