How you audit an account you cannot see
Search the way the customer would, then search the way the customer would not.
Type the service. Then type the service plus "jobs". Then plus "salary", "course", "free", "government hospital". If the ads keep appearing, the account is running broad match with no negative list, and you have learned something specific about how the money is being spent.
The arithmetic of a missing negative list
Broad match is not the enemy. It finds searches you would never have thought to bid on, and some of those convert well.
Broad match without maintenance is the enemy, because it also finds every adjacent search that shares a word with your business. Somebody looking for a job at a clinic and somebody looking for treatment at a clinic use overlapping language and have nothing else in common.
At healthcare click prices, a modest share of irrelevant traffic represents a meaningful monthly loss. Not catastrophic. Just permanent, and invisible unless someone opens the search terms report.
The report nobody opens
There is a distinction inside Google Ads that costs Indian advertisers a lot of money: keywords are what you bid on, search terms are what people actually typed.
The keywords list looks tidy. The search terms report is where reality lives — and where you will find "free", "jobs", "near me government", competitor names, and searches in languages you do not serve.
Reviewing it weekly and adding negatives is unglamorous, takes thirty minutes, and is the single highest-return activity in most search accounts.
The landing page problem underneath
Even with clean traffic, one page for a dozen services underperforms. A searcher who typed a specific procedure and lands on a general clinic homepage has to re-find what they came for. Many will not.
Google notices too. Relevance between keyword, ad and landing page feeds Quality Score, and Quality Score directly affects what you pay per click. Improving relevance lowers cost without touching bids — one of the few genuinely free wins in paid search.
What we would change first
- Pull ninety days of search terms and build the negative list from what is actually there.
- Add standard negatives immediately: free, jobs, salary, course, training, government.
- Split ad groups by service, with a matching landing section for each.
- Add call extensions and, for a clinic, consider call-only campaigns on mobile.
- Restrict location targeting to presence, not interest.
Caveat
We have no account access. Ad appearance on a search is strong evidence of matching behaviour but not proof of spend distribution, and some of what we saw could be a deliberate test.
The reason we publish it: the pattern is close to universal in accounts that have never had a proper search terms review. If you have not opened that report this month, you are probably paying for the word "free" too.