- Google Ads
- Budget
Search terms report: find Google Ads wasted spend in 10 minutes
Your keywords may look tidy, but your ads also show for searches you never wrote, and you pay for every click. A calm, repeatable way to find wasted spend in the search terms report and shut it off.

On this page8
The fastest way to find wasted spend in Google Ads is the search terms report. Set the range to the last 30 days, sort by cost, filter to zero conversions and read the top 20 rows. Give each term one of three decisions: exclude, wait, or move it somewhere better. Then add negative keywords at the right level. It takes about ten minutes a week.
In short
- Your keyword list describes what you want. The search terms report shows what you actually paid for.
- Ask one question: which terms spent money in the last 30 days without a single conversion?
- Every term gets one of three outcomes: exclude, wait, or give it a proper home.
- Add negative keywords at the level the problem lives at: ad group, campaign, shared list or account.
- Check again two weeks later. Leak hunting isn't a one-off clean-up; it's a weekly habit.
What is a budget leak, and why does nobody notice it?
A keyword describes which searches your ad may show for; the search term is what someone actually typed. Broad and phrase match let Google interpret intent, which is how you reach useful searches you never thought of. The same flexibility lets irrelevant ones in. That's the leak: terms that get clicked and paid for but never become a sale, a lead or a call.
Nobody notices because the campaign overview shows averages. A healthy cost per conversion can hide hundreds of wasted clicks when a handful of strong terms carry the average.
A hypothetical: with a $100 daily budget and 15% of spend going to terms that never convert, that's roughly $450 a month. Your number will differ; the real problem is that most accounts never measure it.
Also note that Google leaves some low-volume search terms out of the report for privacy reasons (About the search terms report), so the visible spend may not add up to your total. That's a reason to read what you can see carefully, not to skip it.
How do you find the leak in the search terms report?
Reading the report top to bottom is exhausting and gets you nowhere. Ask a single question instead: which terms spent money in the last 30 days without a single conversion?
- In Google Ads, go to Campaigns → Insights and reports → Search terms.
- Set the date range to the last 30 days. If your sales cycle is long, meaning conversions arrive weeks after the click, use 60 days.
- Sort the table by cost, highest first.
- Filter the Conversions column to "0".
- Turn on the Keyword column. It's hidden by default, but it tells you which of your keywords triggered each term.
- Read the top 20 rows. That's almost always where most of the leak is.
These steps turn several thousand rows into a ten-minute list. Glance at the match type column too: if most zero-conversion terms come from one broad match keyword, the problem may be that keyword, not the terms.
Which date range is right?
Seven days is too noisy: good terms whose conversions haven't arrived yet will look like waste. Very long ranges carry the effect of old bids and ads. Thirty days suits most accounts; for B2B services or high-ticket products, 60 is fairer.
Which decision should each term get?
Every term on the list gets one of three outcomes. Base the decision on the term's intent, not just its cost.
| Intent behind the term | Example search | Decision | What you do |
|---|---|---|---|
| Nothing to do with you | "bookkeeping software jobs" | Exclude | Add a negative keyword |
| Looking for something free | "free bookkeeping software download" | Exclude | Add a negative if you sell a paid product |
| Right intent, thin data | "bookkeeping for small business" (3 clicks) | Wait | Watch it until your conversion lag has passed |
| Great intent, wrong place | "bookkeeping software for contractors" | Give it a home | Add it as a keyword with a matching ad and page |
If the intent is irrelevant (job seekers, students, people looking for a competitor's support line), there's nothing to wait for. If the intent is right but the term has three clicks, the only honest conclusion is that it's too early to conclude anything. And sometimes a term that looks like waste is valuable; it just matched the wrong ad group, ad and landing page.
A negative keyword isn't a punishment. It's a deliberate decision about who your ad should not be shown to.
At what level should you add negative keywords?
When you select terms in the report and click Add as negative keyword, Google Ads asks where to put them: an ad group, a campaign or a negative keyword list (Get negative keyword ideas using the search terms report). That choice matters as much as the keyword itself.
- Ad group: When the problem belongs to one ad group, for example two product groups stealing each other's traffic. Exclude the term from the wrong group only.
- Campaign: When the irrelevant intent shows up in a single campaign.
- Shared negative keyword list: When the same intent appears across several campaigns. Build lists around themes such as "jobs" or "free" and apply them where needed. Google allows up to 20 lists per account, each with up to 5,000 negatives (Add negative keywords to campaigns).
- Account: For terms you never want anywhere. Up to 1,000 account-level negatives are allowed.
Performance Max campaigns now accept campaign-level negative keywords too; they apply to Search and Shopping inventory only.
Choose the match type deliberately
Negative keywords behave differently from regular keywords: they don't expand to close variants or synonyms. Google handles misspellings and capitalization for you, but you need to add singular and plural forms and synonyms yourself (About negative keywords). Excluding "free" does not exclude "complimentary".
One more trap: negatives added from the search terms report default to negative exact match. Exclude "bookkeeping software jobs" that way and tomorrow's "bookkeeping jobs remote" still gets through. To close the intent, add "jobs" as a phrase match negative. Build the list from real terms and the pattern they share, not from guesses.
How do you measure the result?
Two weeks after adding negatives, ask the same report the same question:
- Did zero-conversion spend fall as a share of total spend?
- Did cost per conversion improve while conversion volume held?
- Has a new irrelevant intent appeared?
Track one number weekly: zero-conversion spend divided by total spend. If it falls, you're on track. If conversions fall too, you've probably blocked good intent, so revisit your latest negatives.
You also need to trust the conversion data itself. If Google Ads and Analytics disagree, understanding why GA4 and Google Ads conversions don't match stops you blaming the wrong search term.
And because broad match keeps discovering new searches, leak hunting isn't a one-off spring clean; it's a weekly habit.
What are the most common mistakes?
- Excluding too early. Deciding after three clicks is the fastest way to shut off good intent by accident.
- Mixing in brand searches. Brand searches are almost always cheap and efficient. Keep them separate when judging non-brand performance, or the averages will mislead you.
- Going overboard with one-word broad negatives. A broad match negative on "how" also blocks "how to buy".
- Blocking your own keywords. A negative that conflicts with a keyword in the same campaign stops your ad showing for that keyword's searches. Check for conflicts as lists grow.
How does ad10 do it?
When you ask ad10 "Where did the budget go this month?", the assistant follows the same path:
- Reads. It pulls the period's search terms through the official Google Ads API.
- Ranks. It sorts zero-conversion spend and groups irrelevant intents into themes such as "jobs", "free" or "competitor support".
- Proposes. It drafts a proposal: which negatives go to which campaign, each with its match type. The proposal is checked by the Google Ads API in validation-only mode, so nothing changes in your account yet.
- Waits for you. Nothing is applied until you approve it.
- Logs it. Every approved change is recorded along with who approved it.
ad10 proposes negatives at campaign level; for a theme that repeats across many campaigns, a shared list you set up in Google Ads is tidier.
ad10 asks only for the permissions it needs, and you can remove access at any time (see the security section). Before connecting any tool, read the 7 questions to ask before giving AI access to your ad account. The home page lists the supported platforms and how the plans are priced.
You can of course read the report yourself. What matters is that these ten minutes actually happen every week.
Frequently asked questions
What's the difference between the search terms report and the keywords report?
The keywords report shows the keywords you added. The search terms report shows what people actually typed when your ad showed. Wasted spend hides in the gap.
How often should I check the search terms report?
Weekly is enough for most accounts. Check more often right after launching a campaign or if you rely heavily on broad match.
How many clicks should I wait before excluding a search term?
There's no fixed number. Exclude clearly irrelevant intent straight away. For the right intent, wait at least one conversion lag and until the term has spent about your average cost per conversion.
Why can't I see some search terms in the report?
For privacy reasons, Google omits search terms without enough search volume. So the terms you see can add up to less than the campaign's total cost.
Do negative keywords block synonyms and close variants?
No. Negatives don't expand to close variants or synonyms, so you need to add singular, plural and synonym forms yourself. Misspellings and capitalization are handled automatically.