How to Audit Your Google Ads Search Terms

Your keywords are set up. Your bids look reasonable. But somewhere in the account, money is going to searches that will never produce a lead.

Google does not bill you per keyword. It bills you per search query, the actual words a real person typed before clicking your ad. A keyword like “plumber” or “roof repair” can match dozens of different searches, some from buyers ready to book, some from people looking for a job, researching as a student, or trying to fix something themselves.

In accounts I run, this is one of the most consistent findings in a first audit. The keywords look sensible, but the search terms report tells a different story. The good news: finding the waste takes less than 30 minutes, and the fix is straightforward.

The Short Answer

• Your search terms report shows the actual queries that triggered your ads, not just your keywords. These are often very different.
• Irrelevant searches silently consume budget and give Smart Bidding the wrong training data.
• Pull the report in Google Ads, sort by cost, and look for patterns: DIY intent, job seekers, wrong service, wrong geography, wrong buyer stage.
• Add the wasted-spend patterns as negative keywords to stop them showing up.
• This audit takes 20-30 minutes and should run monthly, or any time leads feel off.

Why search terms are not the same as your keywords

When you add a keyword to a campaign, you are not buying that exact phrase. You are telling Google that you want to show ads to people searching for things related to that phrase. How broadly or narrowly Google interprets that depends on match type, and in 2026, even phrase and exact match expand more than most advertisers realise.

A keyword like “home security installation” can match queries like:

• “home security system install near me” (good)
• “how to install a home security camera yourself” (bad, DIY)
• “home security installation jobs” (bad, job seeker)
• “home security installer salary” (bad, wrong intent entirely)

All of these can result in a paid click. Google’s system is not trying to waste your money; it is trying to find searches that look semantically similar. But similar is not the same as relevant, and the difference between those two things is your wasted budget.

Smart Bidding adds another layer: it uses your conversion data to decide which search patterns to bid up on. If certain irrelevant queries happen to produce page visits that look like soft conversions (a thank-you page that’s accessible without a form, a phone button click), the system learns to target them more. The irrelevant spend compounds over time.

Verdict: your keywords are an instruction. The search terms report is what actually happened. Check both, regularly.

How to pull your search terms report

The report is inside every Google Ads campaign. Here is how to get it.

1. Log into Google Ads and go to your campaign.
2. In the left menu, click “Campaigns” and select the campaign you want to audit.
3. In the secondary menu, click “Keywords,” then “Search terms.”
4. Set your date range to at least the last 30 days. For a first audit, use 90 days to capture enough data.
5. Click “Columns” and add: Cost, Conversions, Conv. value, Clicks, Impressions, CTR. These tell you what each query cost and whether it converted.
6. Download to a spreadsheet or work directly in the UI.

Pro Tip: Sort by Cost descending first. The queries at the top are where most of your budget went. Work top to bottom. You do not need to review every query, only the ones that spent meaningful money. For most lead-gen accounts, the top 20-30 queries by spend cover the majority of the waste.

If you manage multiple campaigns, repeat this for each. Search terms do not cross campaigns, so a query wasting money in Campaign A will not show up in Campaign B’s report.

What to look for: the red flag patterns

Once you have the report sorted by cost, scan for these patterns. They appear in almost every account I audit.

Pattern	Example queries	Why it wastes budget
DIY / how-to intent	“how to fix [service] myself”, “DIY [service] guide”	Person wants to do it themselves, not hire anyone
Job seekers	“[service] jobs”, “[trade] apprenticeship”, “how to become a [service] professional”	Looking for employment, not a service provider
Wrong service	“[related service] near me” where you don’t offer that service	Broad match pulling in adjacent but irrelevant services
Research / early stage	“what is [service]”, “[service] explained”, “[service] review Reddit”	Researching, not ready to buy or book
Wrong geography	Queries from cities or regions you don’t cover	Location targeting sometimes leaks outside your target area
Competitor brand + wrong intent	“[competitor name] phone number”, “[competitor] login”	Looking for an existing provider’s contact, not comparing
Existing customers	“[your brand] customer service”, “[your brand] invoice”	Post-sale queries triggering your ads and costing you money

Not every query in these categories is a miss. “DIY [service] cost” could still be someone who starts researching DIY and decides to hire. Use judgment. A query that spent €40 and produced zero conversions is almost certainly waste. A query that spent €5 and produced zero conversions might just need more data.

Keep in Mind: Zero conversions does not automatically mean a query is irrelevant. A query might need 20-30 clicks before you can expect a conversion. The question to ask is: given the search intent, could this person ever realistically become a client? If the honest answer is no, add it as a negative regardless of conversion data.

What to do with the queries you find

You have three options for a query that is wasting budget.

Option 1: Add it as a negative keyword. This is the right call for queries that are clearly irrelevant (job seekers, DIY intent, wrong service). Add the negative at the right level: campaign-level if the query is only irrelevant for that campaign, account-level if it should never trigger any ad you run. For the mechanics of how negative keywords work and how to structure a negative list, see how negative keywords work in Google Ads.

Option 2: Restructure the keyword. If a broad match keyword is pulling in too many irrelevant variations, tighten it to phrase or exact match. This reduces reach but increases relevance. For lead-gen accounts on Smart Bidding, the tradeoff is often worth it.

Option 3: Do nothing for now. If a query has only a few clicks and zero conversions but the intent seems potentially relevant, add it to a watch list and check again in 30 days. Do not add negatives based on two clicks. Wait for a pattern to emerge.

Verdict: add high-spend irrelevant queries as negatives immediately. Watch low-spend ambiguous queries before acting. Build the list over time.

Common mistake: Adding single-word negatives that are too broad. Adding “jobs” as a broad match negative can block queries like “painting jobs [city]” but also “HVAC jobs done right” or “jobs for plumbers recommended” even when those might be legitimate buyer searches. Use phrase match negatives (“[jobs]” or “employment” or “apprenticeship”) to be precise.

How often to run this audit

For a new account or a new campaign: weekly for the first month, then monthly. New campaigns often match broadly while learning, and irrelevant queries accumulate fast in the early weeks.

For an established account on Smart Bidding: monthly is usually enough. Smart Bidding gets better at filtering irrelevant traffic over time as it learns which queries actually convert. But it is not perfect, and new irrelevant patterns appear as search behaviour shifts.

Any time lead quality drops, lead volume drops without an obvious cause, or cost per lead spikes: pull the search terms report first. This is often where the explanation lives.

If your account is connected to a CRM and you run offline conversion imports, you can add a useful layer: check whether the queries that converted actually produced qualified leads in your CRM, not just form submissions. A query with 10 conversions in Google Ads but 0 qualified leads in the CRM is producing junk. Add it to your negative list or check your conversion tracking setup for that segment.

How I handle this for lead-gen businesses

The search terms report is the first thing I pull in a new account audit and the last thing I check before a monthly report. It is where the gap between “the account looks fine” and “something feels off” usually becomes visible.

In practice: I pull 90 days for a first audit, sort by cost, and work through every query over roughly €20 in spend. I build a running negative keyword list, organised by theme (intent, geography, service category), and add to it each month. Over 6-12 months, this list becomes one of the most valuable assets in the account, stopping irrelevant spend before it compounds.

Ready to stop paying for the wrong searches? If you have not looked at your search terms report lately, there is almost certainly money going to queries that will never become clients. I include a search terms audit in every account review I run.