How much spend the search terms report covers

You open this report from Campaigns, then Search terms inside Insights and reports. It presents itself as the record of what people typed before your ad showed, and its layout implies that the rows in front of you are the query activity behind the campaign, full stop.

The gap shows up the moment you try to reconcile the report's spend column against the campaign's actual spend for the same date range. On the large UK Shopping account behind this figure, the rows with a click covered under half of spend, not the full total a reader would assume from the report's framing.

Google states plainly that some search terms are left out. Its own help page says some search terms that do not have enough query activity are omitted from the search terms report in order to keep with its standards on data privacy, and that the report only surfaces search terms used by a significant number of people that triggered impressions and clicks.

That is the entire documented mechanism. Nothing there says how the omission threshold is set, how many terms fall under it, or how the omitted spend is distributed across keywords. Riibon does not guess at that distribution. It computes the coverage percentage directly from the account's own spend and click totals and reports that number instead of a story about why rows are missing.

Coverage here is partial, not complete. Whatever share of spend and clicks sits outside the search terms report is unmeasured by that report. No documentation explains how that outside share behaves, only that a reader working from the visible rows is working from part of the account, not all of it.

That is true for both spend and clicks. The report's coverage of spend and the report's coverage of clicks are two separate figures, and neither one implies the other. Printing both, for your own account, is the only way to know how much of what happened is actually sitting in front of you when you read the rows.

The report is still worth reading. Partial coverage is not the same as no coverage, and the terms it does show are real queries that triggered a real ad and a real click.

What changes is the confidence attached to any conclusion drawn from it. A negative keyword added because of a pattern in the visible rows is still a reasonable decision, it just needs to be labelled with the coverage percentage it was made on, so nobody later mistakes a partial read for a complete audit.

This is not the same question as how many rows a search terms pull returns. A different measurement on this same report looks at row counts: most rows in a large pull carry no clicks and no spend at all, and pruning them is a separate exercise from asking what share of spend the clicked rows account for. A pull with thousands of zero click rows can still have high or low spend coverage on its clicked rows, the two numbers do not move together.

It is also not a benchmark for any other account. The coverage level behind this fact describes one large UK Shopping account over one 90 day window. A different account, vertical or match type mix will land on a different coverage figure, which is exactly why that figure has to be computed per account rather than assumed from this one.