If the searches in your report do not add up to what the campaign spent, nothing is broken. In SoudCoh's research sample of Australian trade and home-service advertisers, Google's search terms report accounted for about 52% of the money a typical Search campaign spent, and named the search behind roughly 51% of enquiries. The rest went on searches Google does not show you.
That changes how a campaign has to be controlled. Below: why the gap exists, what it means for negative keywords, the settings that leak spend the report never lists, and a check you can run on your own account.
How much of your Google Ads spend does the search terms report show?
About half. SoudCoh's analysis of reported trade campaign data found the report names the searches behind a little over half of the spend, and the share is no better for enquiries.
Key facts
- The search terms report accounts for about 52% of the money a typical trade Search campaign spends.
- It names the search behind roughly 51% of a typical campaign's enquiries.
- Only 4.4% of the visible search-term spend was clear-cut waste to negate.
- 75% of the spend a daily review flags is on searches that look relevant but have not converted yet.
- A typical trade advertiser carries about 0.2 negative keywords for every keyword it bids on, or roughly one for every five.
- 21% of trade advertisers in the sample had campaigns labelled Search also serving ads on the Display Network, which the search terms report does not list at all.
| What we measured | Median |
|---|---|
| Share of Search spend behind named search terms | 52% |
| Share of enquiries behind named search terms | 51% |
| Clear-cut waste in the visible spend | 4.4% |
| Flagged spend that was relevant but not yet converting | 75% |
How we measure. Visibility compares the spend, or the enquiries, attributed to named searches in the search terms report with the campaign's total for the same dates. An enquiry is a recorded phone call or a submitted form.
Why does Google hide some search terms?
Google says the search terms report only includes searches made by a significant number of people. Rare searches are held back, and trade demand is full of rare searches: 54% of enquiries across the sample came from searches that received no more than two clicks in total. The searches most likely to be withheld are exactly the one-off phrasings a trade customer types in the middle of a problem.
So the gap is a threshold, not a fault. It is not a sign of click fraud or a broken account, and there is no setting that removes it. What it does mean is that every search terms report is a sample of demand, not the whole of it. Decisions about match types and negatives have to be made knowing that the hidden half exists.
Is the hidden half of your spend wasted?
Not by default, and the data points the other way. If the hidden searches were mostly junk, they would cost money without producing enquiries, and the report would name a much larger share of enquiries than of spend. It does not: the named share of enquiries (51%) sits almost exactly on the named share of spend (52%). The hidden searches behave like the long tail you can see.
The visible half is also cleaner than most owners expect. In SoudCoh's research sample, clear-cut waste was 4.4% of reviewed search-term spend. It varied by trade:
| Trade | Clear-cut waste in reviewed search-term spend |
|---|---|
| Other trades and home services | 4.5% |
| Cleaning | 1.5% |
Most of what a review flags is not rubbish but relevant searches that have not paid back yet. That is why a promise of big savings from cutting wasted spend alone usually describes an account nobody has looked after. The large gains in trade search come from structure, bids and landing pages; cleaning protects them.
A negative list can only police the part of the budget you can see. Structure polices all of it.
If you cannot see the searches, what controls them?
Structure. A negative keyword does apply to every matching search, named or not, but you can only write one for a pattern you have already seen. Almost every other lever in a Search campaign acts on all searches, visible or hidden:
| Lever | What it controls | Acts on searches the report hides? |
|---|---|---|
| One service per ad group | Which ad and page a search is answered with | Yes |
| Match types | How far each keyword can reach | Yes |
| Location set to presence | Whether someone outside the service area can see the ad | Yes |
| Search network only | Whether the ad can appear on apps, sites and partner search | Yes |
| Ad schedule | When the ad can show | Yes |
| Conversion tracking and bidding | Which searches automated bidding chases | Yes |
| Landing page | Whether a click becomes an enquiry | Yes |
| Negative keywords | Patterns you have already found | Yes, but only for patterns you can see |
That is why SoudCoh treats structure as the control and the negative list as the finishing layer. A typical trade advertiser carries about 0.2 negatives per keyword. A negative list that grows much faster than the keyword list is a sign the structure is too loose, and every negative added is one more thing that can block a customer by accident, which we cover in negative keywords that block customers.
Which settings leak spend before any search term is read?
Three settings decide where your ads can appear before any search term is matched, and none of them is fixed by reading search terms:
- The Display Network inside a Search campaign. 21% of trade advertisers in the sample had campaigns labelled Search also serving on Display. Where it ran, Display took 23% of those campaigns' impressions but only about 1.5% of their spend, and a Display click produced a real enquiry at about 4% of the rate of a Google Search click. Our piece on the Display Network in Search campaigns covers the fix.
- Search partners. Partner clicks cost less, but converted at only 16% of the Search rate, so each enquiry cost more.
- Presence or interest. Google's default location option also reaches people who have shown interest in your area without being in it. For a business that can only serve a set area, that is spend on people who cannot become customers.
Every Search campaign created through SoudCoh's campaign builder has Search Partners and Display expansion switched off and targets people physically in the service area (100% of builds). For an account you inherit, check all three before reading a single search term.
How should you use the search terms report, then?
As a sample of demand, read for patterns:
- Weekly: look for repeated words, services and places, not individual rows. A pattern is worth a negative; a one-off rarely is.
- Monthly: promote searches that have converted more than once into exact keywords, so they can be bid on and written for. See most trade enquiries come from searches you never chose.
- Always: judge the campaign on enquiries and cost per enquiry, not on how tidy the visible terms look.
Cleaning does pay off over time. Each negative removes a pattern for good, so the junk that reaches an account shrinks as the list matures: two months into daily cleaning, the share of reviewed spend that needed negating had fallen by 43% across the sample. The first month of a new account is when review matters most; after that it becomes maintenance. The full method is in our white paper Launch, discover, control.
The search terms report is a sample of demand, not the whole of it.
How can you measure the gap in your own account?
You can work out your own visibility share with the report and a spreadsheet:
- Pick one Search campaign and a date range of at least a month, and note its total cost.
- Open the search terms report for the same campaign and dates, download it and add up the cost column.
- Divide the named cost by the total. That is the share of spend your report can explain. Repeat with conversions.
- Segment the campaign by network. Any cost on the Display Network is spend the search terms report will never list, and search-partner cost deserves its own look, because partner clicks convert far less often.
- Check the location option in campaign settings, and look at the user location report for clicks from outside the area you serve.
A low visible share is not a fault in itself, but it tells you how much weight the report can carry. The lower it is, the more your control has to come from ad group themes, match types and settings. See how we set these up for plumbing businesses, electricians, roofing businesses and moving companies, or read how we run Search campaigns.
Common questions about the search terms report
Why don't my Google Ads search terms add up to my total spend?
Because Google only names searches that enough people have made, and leaves the rest out of the report. In SoudCoh's research sample, the search terms report accounted for about 52% of the money a typical trade Search campaign spent. The gap is expected, not an error.
Is the spend on hidden search terms wasted?
Not by default. Enquiries were no more visible than spend: the report named the search behind roughly 51% of enquiries. The hidden searches produce enquiries in about the same proportion as they cost, so treat them as the long tail you cannot see, not as waste.
Do negative keywords work on search terms I can't see?
Yes. A negative keyword applies to every matching search, whether or not the report names it. The limit is that you can only write negatives for patterns you have seen, which is why structure, location and network settings have to do most of the controlling.
Can I get Google to show me the hidden search terms?
Not one by one. Google's Insights page groups searches into themes, which can hint at what sits behind the gap, but it does not list every search. Read the visible report as a sample and look for patterns rather than single rows.
How often should I check the search terms report?
Weekly for patterns worth negating, monthly for converting searches worth promoting to keywords, and daily in the first month of a new account, when cleaning matters most.
Questions and answers
Should I pause a campaign if I can't see most of its searches?
No. Judge it on enquiries and cost per enquiry. A campaign with a partial report and a healthy cost per enquiry is working; tidy search terms with few enquiries are not.
What should I bring to a search terms review?
Bring the campaign totals, the downloaded search terms for the same dates and a short list of the jobs you do and do not want. That is enough to separate patterns from noise.