Negative keywords that block your customers

Yes. It is common, usually small in cost, and concentrated in a minority of accounts, which makes it the cheapest fix in paid search: the enquiries were already proven.

Key facts

• 52% of trade advertisers in the sample were blocking, through their own negatives, at least one search that had produced an enquiry.
• Searches that are now blocked had produced 0.4% of search-term enquiries. In the worst case in the sample, they had produced 6% of that business's paid-search enquiries.
• 1.5% of the searches a keyword-rule cleaner flagged for negation had converted in the very window it was judging.
• 2% of the searches simple waste rules flagged had produced a genuine enquiry when checked over a longer history.
• In SoudCoh's review process, 60% of proposed negatives were thrown out, including 92% of the "we don't do that" kind and 47% of competitor names.
• After every safety check, the negatives that could safely be added covered 0.6% of monthly spend.

How we measure. A blocked search is one that produced an enquiry and is now excluded by one of the advertiser's own negative keywords. The rejection rates come from SoudCoh's review process, in which every proposed negative is challenged by a reviewer and simulated against converting searches before it goes live.

Because a reviewer looking at a search term sees one word and one job, while the account sees every page, ad group and service the business sells. In one plumbing business in the sample, pool, dishwasher and waterproofing all looked like other people's work, and all would have blocked searches that converted. The other classic own-goals are suburbs inside the service area and everyday phrases that happen to be a competitor's name.

Single-word negatives are the riskiest of all. A one-word phrase or broad negative blocks every search that contains the word, and the words that feel safest, such as cheap, price or pool, are often the name of a service the business sells or a segment it deliberately targets.

There is an awkward twist in the data. Searches for a different service, or for something unrelated that shares a word, were also the biggest single source of genuine waste: 44% of all the spend flagged for negation. The bucket with the most waste is also the bucket with the most dangerous negatives. That is why it needs a business-specific list, built from the owner's own "we don't do that" list and checked against converting searches, rather than a template.

Mostly from searches that look relevant enough to click, not from the obvious junk. This is the spend flagged for negation across SoudCoh's research sample, by reason:

Two things stand out. First, job-seeker and DIY searches, the usual targets of template negative lists, were only 12% of the flagged spend. Template lists are cheap insurance on day one, but they do not solve the waste problem. Second, flagged is not the same as safe to block. Price searches appear in the table because some of them, in some accounts, spent without converting; across all price-research searches, enquiries cost about the same as any other. Judge each pattern on its own results in its own account.

Overall, clear-cut waste was 4.4% of reviewed search-term spend. In a well-structured account, junk searches are a small leak, not a flood.

Not by default. Price-research searches, such as cost, price, how much and quote, convert less often, but Google charges less for them, and the two effects roughly cancel out. In SoudCoh's research sample, an enquiry from a price-research search cost 0.95× as much as other enquiries. A blanket negative on price or cost, a common default in rules-based cleaners, throws away enquiries at a normal cost.

Bargain words such as cheap, affordable or budget are a judgement call. Whether they are worth having depends on whether those customers accept your quotes, and that answer lives in your job records, not in a word list. Our white paper From keyword to revenue sets out how to connect won jobs back to the searches that produced them.

The better move is to answer the price question openly. Say what a call-out includes, whether quotes are free, and what drives the price, in the ad and on the page. People who cannot afford the job can then rule themselves out before they click. If lead quality is the worry, our guide to poor-quality Google Ads leads covers where to look first.

Rarely as a blanket rule. In SoudCoh's research sample, searches naming another trade business took only 0.6% of search-term spend, so blocking them all saves very little. The downside is real: for 72% of the advertisers that paid for competitor-name clicks, those searches produced at least one enquiry. People search a known name, find it busy, unreachable or out of area, and call the next credible option.

Many trade business names are also everyday trade phrases, which is why 47% of the competitor negatives proposed in SoudCoh's reviews were rejected.

A safer rule: negate a rival's name only when it has taken real spend with no enquiries over a long window, and then use the full, unambiguous brand phrase. If competitor traffic is worth keeping, give it its own low-bid campaign so it can be judged on its own numbers. If the question is the other way round, see a competitor is bidding on my brand name.

Treat them as a question, not a verdict. About 75% of the spend a daily review flags is on searches that match what the business sells but have not produced an enquiry yet. Two findings from the same review show why a cost threshold is the wrong tool for them:

• 7% of the relevant searches held back from negating went on to produce an enquiry in a later month.
• 53% never triggered an ad again, so negating them would only have lengthened the list.

Before negating a relevant search, check the ad it triggered, the landing page and the bid. Negate only when the pattern repeats with meaningful spend and no enquiries over a long window, or when the owner confirms the job is not wanted.

Replay it against the searches you already have. First, be clear on how each negative match type blocks, because none of them match close variants, plurals or synonyms. Take a cleaning business that does not clean carpets and is considering a negative for carpet cleaning:

Then run the simulation:

1. Download the search terms with conversions over the longest range the account has, ideally a full year.
2. Find every search the negative would block under its match type, and add up the clicks, cost and conversions.
3. If any blocked search converted, narrow the negative (a longer phrase, or exact) or drop it.
4. Read the blocked searches, not just the totals. A negative that removes one bad search and three suburbs you serve is a bad negative.
5. Write down why it was added, so the next person can tell a policy from a mistake.

Google Ads warns when a negative conflicts with one of your own keywords, but not when it blocks a search that has converted. That check is yours. At SoudCoh, no automated negative is pushed without a person and a simulation against converting searches, and a large clean-up passes four gates: a proposal, a challenger who tries to refute it using the account's own services, the simulation, and a read-back of what each surviving negative actually blocks.

Once a quarter. Negative lists only grow unless someone prunes them. Pull the searches that produced enquiries over the last year, check which of them are now excluded, and decide for each one whether it is a deliberate block, such as a job the owner no longer wants, or an accident to undo. Pay special attention to shared negative lists and to negatives used to route suburb searches between service and location campaigns, because a list applied to the wrong campaign silently removes a whole service.

Keep the list in proportion. A typical trade advertiser carries about 0.2 negatives for every keyword it bids on. Most control comes from structure: tight ad groups, sensible match types and location settings. The full method is in our white paper Launch, discover, control, and the reason structure matters more than the list is in why the search terms report hides half your spend.

See how we run negative-keyword reviews for plumbing businesses, cleaning businesses, pest control and HVAC businesses.

Can negative keywords block conversions in Google Ads?

Yes, and it is common. In SoudCoh's research sample, 52% of trade advertisers were blocking, through their own negative keywords, at least one search that had produced an enquiry. The cost is usually small, but it falls on customers who were already proven to call.

Should I add cheap and price as negative keywords?

Not by default. In SoudCoh's research sample, enquiries from price-research searches cost 0.95× as much as other enquiries, because cheaper clicks offset a lower conversion rate. Judge bargain words on won jobs, not on the word, and answer the price question in the ad and on the page instead.

Should I add competitor names as negative keywords?

Rarely as a blanket rule. In SoudCoh's research sample, competitor-name searches took only 0.6% of search-term spend, and for 72% of the advertisers that paid for them they produced at least one enquiry. Negate a rival's full brand phrase only after a long run of spend with no enquiries.

Do negative keywords block plurals and misspellings?

No. Negative keywords do not match close variants, plurals, misspellings or synonyms. A negative for one form of a word leaves the other forms untouched, so add each form you mean to block, and check each one against searches that have converted.

How do I check if my negative keywords are blocking customers?

Download the search terms that produced conversions over the longest range your account has, then test each negative against that list using its match type. Any converting search a negative would block is a reason to narrow that negative or remove it.