How Do I Know If My Google Ads Are Showing for the Wrong Searches?

Your Google Ads keyword list looks right, yet the enquiries are off-target. They come from job seekers, DIY questions and requests for services you don't offer. You paid for every one of those clicks, and the campaign dashboard won't flag them, because keywords record what you bid on, not what anyone typed.

You find wrong searches in the search terms report, which lists the real queries that triggered your ads. Pull the last 90 days, sort by cost, and ask of each expensive term whether the person typing it could buy what you sell. Any term that fails is a wrong search. Then check your location settings and Performance Max, because the report alone won't expose either.

Why don't your keywords match your searches?

Match types used to be a promise about characters. They are now a judgement about meaning.

• Broad match serves your ad on searches Google considers related to your keyword, including searches that don't carry its direct meaning. It also weighs signals beyond the keyword: the searcher's recent activity, your landing page and assets, and the other keywords in the ad group.
• Phrase match serves on searches that include your keyword's meaning.
• Exact match can still show on searches judged to share the same intent.
• Close variants apply to every match type. These include misspellings, plurals, abbreviations and reordered words.

So your keyword list is a set of instructions Google interprets, not a filter it applies. A plumber bidding on "water heater" can appear for "water heater recall" or "water heater DIY." Both contain the keyword, and neither comes from a customer.

Broad match stretches furthest, especially before an account has enough conversion data to tell good traffic from cheap traffic. No match type gives certainty, though. If you need control over which searches you appear on, negatives do that job. Match types only shape it.

Where can you see the searches that ran?

The search terms report sits in the Keywords area of a Search campaign. For each query that triggered an ad, it shows clicks, cost and conversions. You can view it at account, campaign or ad group level and download it. Google has moved the menu over the years, so if it isn't where you remember, search the interface for "search terms."

How you read the report matters more than where you find it. The default view rarely surfaces the waste. Use a window long enough for patterns to show. Ninety days is a sensible audit range. Then sort by cost, highest first, and judge the top terms one by one, because that is where a wrong match costs most. You can handle the long tail of one-click queries in batches once you know what the bad ones share.

One caveat: the report is not complete. Google withholds queries it considers too low-volume to display, so some of your spend is attached to searches you'll never see. Treat the report as the best evidence available, not a full ledger.

What does a wrong search look like?

A search is wrong when the person typing it isn't trying to buy what you sell. That test works better than any keyword rule because it makes you read each query as intent, not text. Wrong searches usually fall into a few recognisable types:

• Employment: "jobs", "salary", "career", "vacancy". These searchers want to work in your industry, not hire from it.
• Do-it-yourself and free: "DIY", "how to", "free". These searchers are trying not to pay anyone.
• Research: "reviews", "X vs Y". These are sometimes worth buying, but often not on Shopping ads, where you pay product-ad prices for someone still reading comparisons.
• Competitor brands: broad match will readily put a running-shoe advertiser on "Nike Pegasus trail review." Appearing there can be a legitimate strategy, but it should be a decision you made.
• Services you don't offer: the query is commercial and local, just for the wrong job.
• Problem queries: recalls, complaints, outages.

Check intent first and conversions second. A relevant term that doesn't convert is a different problem from an irrelevant one. When you do weigh performance, compare conversion value against cost rather than counting conversions, since a few high-value sales can outearn a pile of cheap ones.

What will the search terms report miss?

Some wrong traffic looks perfectly right in the report.

Location. Google's default location setting can include people "interested in" your area, not just people located there. A Houston dentist can pay for clicks from someone in another state researching Houston. The query is flawless, but that searcher can never become a patient. Check the location report, set targeting to people in or regularly in your service area, and exclude places you don't serve.

Performance Max. Running alongside Search, Performance Max can absorb branded queries from people who were going to find you anyway. To check, compare branded Search impression share and branded conversions before and after it launched. If Performance Max search conversions rose while branded Search volume fell, it is taking credit for demand it didn't create.

Hidden queries. You can't negate a term the report never shows you. That is one more reason tightly themed ad groups and broad negative lists matter more than reacting row by row.

What if the report comes back clean, with matching intent from people in your area, and the phone still doesn't ring? Then targeting isn't the problem, and the next question is whether the landing page is losing them.

What do you do once you find them?

Add the wrong searches as negative keywords, but know what the interface does by default. Negatives added straight from the report are typically created as exact match and attached only to the ad group or campaign where the query triggered. Other campaigns stay exposed until you add the negative there too. For recurring patterns, build a shared negative list and apply it account-wide.

Also extend each negative beyond the single query. If "job" was a problem, "vacancy", "career" and "employment" will be too. Brainstorm the variants once rather than paying to discover each one separately.

The harder part is the habit. This waste rarely announces itself, because ads keep running and some leads still arrive. Put the search terms review on a fixed schedule instead of waiting until someone notices cost per lead drifting. A daily brief that flags shifts in spend and performance, like the one SproutMe Companion builds from live account data, tells you when to open the report sooner. If wrong searches are only part of a budget going nowhere, the wider diagnosis of spend with no leads covers the other causes.

Conclusion

Your keyword list can't tell you whether your ads are showing for the wrong searches, because every match type now works on meaning and Google interprets your keywords for you. The search terms report is the evidence. Sort it by cost and judge each expensive query on whether the searcher could buy from you. Then check location targeting and Performance Max, since both can send traffic that looks fine in the report. Turn what you find into negatives that cover every campaign and every variant, and review on a schedule rather than waiting for a crisis.

See how Companion's daily brief surfaces shifts in spend and performance across your accounts here.