Every Google Ads account pays for some clicks that were never going to become customers: someone looking for a job rather than a tradesperson, a student searching for a course, a DIYer after spare parts. Negative keywords are how you tell Google to stop showing your ads for those searches. They don’t need a bigger budget, a new website or a change of strategy, just regular attention to what people are typing before they click.
How negative keywords work
A negative keyword tells Google not to show your ad for a search containing that term, even if one of your normal keywords would otherwise match it. Add “jobs” as a negative and your ad for “electrician Gloucester” won’t appear when someone searches “electrician jobs Gloucester” because they want work, not an electrician.
The catch is that negatives behave differently from the keywords you bid on. Your normal keywords match close variants, so your ad can show for plurals, misspellings and searches with the same meaning. Negative keywords don’t. Google’s own guide to negative keywords gives the example that excluding “flowers” blocks “red flowers” but not “red flower”. So if you exclude “job”, you also need “jobs”, and probably “vacancies”, “careers” and “apprenticeship”. Google does take care of capital letters and common misspellings for you.
There are three match types for negatives:
- Negative broad match (the default) blocks any search containing all of the words, in any order. A negative broad “boiler manual” blocks “manual for my boiler”.
- Negative phrase match blocks searches containing the words in that exact order, with anything before or after them.
- Negative exact match blocks only that exact search, with no extra words.
Most everyday negatives work well as single words in broad match (“jobs”, “course”, “free”). Phrase and exact negatives are for the trickier cases, where a word is only a problem in one particular combination.
One thing to watch: if a negative overlaps with a keyword you’re bidding on, the ad for that keyword simply won’t show. It’s an easy mistake to make when adding negatives in bulk.
The searches that quietly drain a budget
The obvious culprits are the ones most accounts already exclude: “free”, “jobs”, “DIY”, “how to”, “salary”. The ones that cost real money are usually specific to the business, and they only become obvious once search data starts coming in.
A fencing contractor bidding on “fencing” can pick up “fence panels for sale” from people who want to buy materials, not hire a fitter. A bookkeeper targeting “bookkeeping services” might attract “bookkeeping course” or “bookkeeping jobs near me”. A removals firm advertising “man and van” can find budget going on “man and van hire” from people who want to drive the van themselves. None of these are careless keyword choices. They’re the kind of overlap you only see once real searches come through.
The same care applies in reverse. “Course” might be irrelevant for a bookkeeper, but a driving instructor who sells intensive courses would be cutting off their best customers by excluding it. Check every negative against what you do sell.
Finding them in the search terms report
The search terms report shows the actual searches that triggered your ads, next to the keyword that matched each one. In Google Ads you’ll find it under Campaigns, then Insights and reports, then Search terms. You can tick an irrelevant term and add it as a negative keyword straight from the report.
Two things are worth knowing. First, it won’t show every search: Google leaves out terms used by only a small number of people, for privacy reasons, as explained in its page about the search terms report. Second, the report is only as useful as your conversion tracking. If you can see which search terms led to calls and enquiries, you can tell the difference between a term that’s irrelevant and one that’s just expensive. Our guide to conversion tracking mistakes that skew your data covers the common problems.
As a rough rhythm, look every week or two for the first couple of months of a new campaign, while you learn what it’s matching to, then monthly after that. If you use broad match keywords, or Google’s AI Max feature for Search campaigns (which finds searches beyond your keyword list), it’s worth looking more often, because the range of searches you can appear for is much wider. Google confirms that negative keywords are still respected with AI Max switched on.
Build your list from your own data
It’s tempting to sit down on day one and try to anticipate every irrelevant search, but an overcautious list built before any data exists tends to block searches that would have converted, simply because they looked risky on paper. Start with the genuinely obvious exclusions, then build the rest from what the search terms report shows you.
For the same reason, be wary of downloading a generic negative keyword list and adding the whole thing in one go. These lists aren’t usually wrong, but they’re written for nobody in particular. A locksmith and a driving school might both sensibly exclude “jobs”, but the rest of their lists should look nothing alike.
Keeping negatives organised
You can add negatives to a single ad group, a whole campaign or the entire account. For terms that apply everywhere, such as “jobs”, “careers” or “free”, a negative keyword list saves adding the same words to every campaign: you build it once, apply it to as many campaigns as you like, and any word you add later applies to all of them. Google allows up to 20 lists per account, each holding up to 5,000 negatives.
There’s also a single account-level negative keyword list, found under Account settings, which holds up to 1,000 terms and applies across all Search and Shopping inventory in the account, including Performance Max campaigns.
What about invalid clicks and click fraud?
When an account spends money with nothing to show for it, click fraud is often the first suspicion: a competitor clicking your ads to drain your budget, or bots clicking at random. It does happen, but it’s worth knowing what Google already does about it, and why the real cause is usually something more fixable.
What Google counts as invalid traffic
Google defines invalid traffic as clicks and impressions that aren’t the result of genuine interest. That includes accidental clicks such as the second click of a double-click, manual clicks meant to increase an advertiser’s costs, and clicks from bots, crawlers and other automated software.
Automatic filtering and credits
Google’s systems filter this traffic automatically, and you aren’t charged for invalid clicks they catch. If invalid activity is found after you’ve already been invoiced, you get a credit rather than a refund. Credits appear as “Invalid activity” under Adjustments in your billing summary, and Google’s automated reviews look back over the previous 60 days.
Where to see invalid clicks in your account
Add the “Invalid clicks” column to your campaigns table: select the Columns icon, search for “Invalid clicks” and apply. A number in that column isn’t a warning sign. It shows clicks that were caught and not charged. There’s also an Invalid Activity Credit Report, which breaks any credits down by campaign.
Why most “fraud” turns out to be something else
Google’s own help pages list the usual reasons for a sudden jump in clicks: a higher budget or bids, new keywords, broader targeting, or a seasonal rise in interest. Several clicks from one IP address can be one person comparing options and coming back, or lots of different people sharing an address through an office, a university or a mobile network.
In account audits, what looks like fraud is usually loose broad match keywords with too few negatives, location settings that reach people outside the service area (see our post on the Google Ads location setting), or a landing page that isn’t turning visitors into enquiries. Don’t forget your own team either. Clicking your own ad to check it’s running costs money like any other click. Google’s Ad Preview and Diagnosis tool lets you see your ads without clicking on them.
IP exclusions and their limits
You can block specific IP addresses from seeing your ads: up to 500 per campaign, or across the whole account under Account settings, which also covers Performance Max (Google’s guide to excluding IP addresses). Excluding your own office network is a sensible use.
As a defence against fraud, they’re limited. Google Ads doesn’t show you the IP address behind each click, so you’d need your website’s server logs to find them. Many home and mobile connections share or rotate IP addresses, so someone set on clicking can appear from a new address, while a broad exclusion can block genuine customers on the same network.
When it’s worth reporting
If you see a repeated pattern over days or weeks that Google hasn’t filtered, such as bursts of clicks at odd hours from places you don’t serve with no enquiries, you can ask Google to investigate the last 60 days using its click quality form. Gather the dates, campaigns, keywords and your server logs (including IP addresses) first. Google says investigations typically take several working days.
For most small businesses, though, a well-kept negative keyword list protects far more budget than chasing suspected fraud, because it stops your ads showing to the wrong searches in the first place. It’s also worth doing before you decide how much to budget for Google Ads, since cutting waste makes the same money go further.
Want a second pair of eyes?
Our free Google Ads audit goes through your search terms, negative keywords and invalid click data, and tells you plainly where budget is leaking and what we’d change first.
← Back to Blog