Before you add another negative keyword, answer one question: what fraction of this campaign’s spend can you actually see in the search terms report? Most practitioners never compute it, assume it is most of their spend, and mine the report as if it were a complete ledger. It is not. A blunt contrarian summary of the problem puts it this way — “Google buries most of your clicks under ‘Other search terms’. It’s quite common to see actual search terms for less than a third of your ad spend.” If that is true of your account, then mining harder is aimed at the third you can see while the expensive two-thirds stays invisible.
This post gives you the number that settles the argument: the search terms visibility ratio. It is one division you can do in five minutes, and it turns a vague sense that “the report is getting worse” into a concrete go/no-go rule for whether term-by-term mining still earns its hour, or whether the real fix is upstream in match types and campaign structure. The library already covers why terms are hidden and when mining hits diminishing returns; this is the measurement that connects them.
What the search terms visibility ratio is
The visibility ratio is the share of a campaign’s spend that appears against a named query in the search terms report, expressed as a percentage of total campaign spend. Put simply: visible spend divided by total spend. If a campaign cost £10,000 over the last 30 days and the named rows in its search terms report add up to £3,000 of cost, its visibility ratio is 30% — every query you could negate accounts for less than a third of what you paid, and the remaining £7,000 sits in the anonymous “Other search terms” bucket where no negative you can write will reach it.
This is a diagnostic you build, not a metric Google hands you. Nowhere in the interface does it say “you can see 30% of this campaign.” That absence is the problem: because the number is never surfaced, most accounts run for years without anyone noticing that their most-used optimisation tool is looking at a shrinking slice of the account. The ratio matters because it changes the meaning of everything downstream. A negative keyword strategy that is excellent at 80% visibility is close to cosmetic at 20% — not because the negatives are wrong, but because the queries doing the damage never appear where you can act on them. Compute the ratio first, and every later decision about mining, match types, and structure has a number behind it instead of a hunch.
Why so much of your spend never shows as a term
Google groups any query that falls below a privacy volume threshold into “Other search terms,” and that bucket has swollen as automated matching generates more near-unique queries that each appear too few times to clear the threshold. Independent analyses of the hidden share are consistent about the scale: reviews of aggregate account data commonly find that only about half of spend is tied to visible terms, and for broad-match-heavy campaigns the invisible portion can climb well past 80%. The exact figure is less important than the direction: the report is partial by design, and getting more partial.
Two forces compound the volume threshold. First, broad match and AI Max expand one keyword into a long tail of one-off queries, most of which are individually too rare to report — so the more automation you run, the more of your spend disappears into the anonymous bucket. Second, for some AI-interpreted queries the report now shows Google’s read of intent rather than the literal string a user typed, a shift we cover in how AI-interpreted terms change negation. The combined effect is that visibility is not a fixed industry constant you can look up — it swings from maybe 80% on a tight exact-match brand campaign to under 20% on a broad-match prospecting campaign in the same account. That variance is precisely why you have to measure your own campaigns rather than assume the average applies to you.
How to compute your visibility ratio
The calculation takes five minutes per campaign. Open the campaign, set your date range (30 days is a reasonable default), and go to the search terms report. Download or sum the Cost column across every visible row — that total is your visible spend. Then read the campaign’s total cost for the identical date range from the campaigns view. Divide visible spend by total spend and multiply by 100. That percentage is the visibility ratio. Do it per campaign, never account-wide, because a high-visibility brand campaign will paper over a low-visibility prospecting campaign if you blend them, and the blended number hides the exact place you most need to look.
Two refinements make the number more useful. Segment by match type where you can — run the ratio separately for the broad-match and exact-match portions of a campaign, because the gap between them tells you how much of your blindness is self-inflicted by match-type choice. And weight by what you care about: if you have conversion data, also compute the ratio on converting spend, since a campaign where you can see the terms behind your conversions is in far better shape than one where the winners are invisible even if the raw cost ratio looks similar. For Performance Max, the equivalent exercise runs through the search terms and channel views covered in the Performance Max search terms report, where the “visible” portion is smaller still and worth measuring before you assume negatives are doing much.
Reading the number: thresholds and what they mean
Once you have the ratio, it maps to a decision. Above roughly 70%, the report is still a usable to-do list: most of your spend is attached to queries you can read and act on, so term-by-term mining pays and the hidden bucket is a rounding error. Between 30% and 70%, you are in mixed territory — mining still catches genuine waste, but the majority of spend is invisible, so you treat the report as one input and pair every mining session with an upstream check. Below 30%, the report is mostly blind: adding single-query negatives is theatre, because the leak lives in the two-thirds you cannot see, and no negative you are able to write will touch it. These are practitioner cutoffs, not Google thresholds, so calibrate them to your own account rather than treating them as law.
The value of the thresholds is that they redirect effort to where it can actually work. A low ratio is not a signal to mine harder — that is the treadmill the same source describes, where you could add another hundred negatives today and need another hundred tomorrow. It is a signal that the report has stopped being the right tool for this campaign, and that the money is escaping through a door negatives do not close. When the ratio is high, mining is your lever; when it is low, the lever is upstream. The number tells you which, so you stop spending hours on the tool that cannot reach the problem.
What to do when the ratio is low
A low visibility ratio is fixable, but not by mining — you fix it by changing where spend flows so more of it runs through queries above the reporting threshold. The first lever is match type: broad-match spend that is generating one-off junk is exactly the spend disappearing into “Other search terms,” so moving it toward phrase or exact concentrates budget on fewer, higher-volume queries that clear the threshold and reappear in your report. The match-type decision is therefore also a visibility decision: tighter matching buys back the ability to see what you are paying for. The second lever is structure — segmenting a sprawling campaign so each ad group chases a tighter intent raises the per-query volume and pulls more terms into view.
Where you genuinely cannot raise visibility, shift to negatives that do not depend on seeing the query. Pattern and n-gram negatives block an intent class across every future phrasing, seen or hidden, which is why an n-gram analysis outperforms row-by-row mining on a low-visibility campaign: it acts on the recurring fragment behind hundreds of invisible one-offs. Brand and channel exclusions do the same at the structural level, stopping Performance Max and Search from cannibalising cheaper branded traffic regardless of what the report shows. Keep this disciplined: the move on a low ratio is a match-type change, a campaign split, a pattern negative, or an exclusion — not a rebuild of your whole account. You are buying back visibility and blocking categories, then re-measuring to confirm the ratio moved.
Re-running the audit as a cadence
Treat the visibility ratio as a metric you track, not a one-off curiosity. Compute a baseline for every campaign you actively manage, then re-check monthly and immediately after any structural change — a match-type shift, a campaign split, or a forced event like the September 2026 AI Max auto-upgrade that switches search-term matching on by default and can quietly drop your visible share. A falling ratio is a leading indicator: it tells you the report is going dark before your wasted spend visibly climbs in the cost-per-conversion numbers, which buys you time to act while the fix is still cheap. Log the number alongside your other campaign health checks so a slow decline is impossible to miss.
The habit this builds is the point. Instead of opening the search terms report and reflexively adding negatives, you open it and first ask whether it can even see the problem — and you have a number that answers. When the ratio is high, mine with confidence. When it is low, spend the same hour on match types, structure, and pattern-level negatives that work in the dark. Combined with a proactive negative keyword list built before launch and the pillar workflow in reading the Google Ads search terms report, the visibility ratio turns negative-keyword work from an act of faith in a partial report into a measured decision about where your money actually goes.