How to actually use the Google Ads search terms report

There is one screen in Google Ads that pays back the time it takes to learn it, and it is not the dashboard. It is the search terms report. Everything else is downstream of what shows up there.

If you only had thirty minutes a month to spend on a Google Ads account, you would spend twenty-five of them in the search terms report. The remaining five would be split between checking the daily budget pace and skimming the headline conversion number. Everything else can wait, and most of it will look different next month anyway.

This piece is about working that single screen well. Where to find it, how to read it, what to act on, what to ignore, and how to keep doing it on a cadence that does not eat your week.

Why this report is the one that matters

The search terms report shows you the actual queries people typed when they saw your ad. Not the keywords you bid on. The queries the matcher matched to your keywords. Those are different things, often very different, and the gap between them is where most ad spend leaks.

You bid on plumber phoenix. The matcher matches your bid against plumber salary phoenix, plumber jobs phoenix, and plumber school phoenix az. Three queries you would never have written as keywords, all three eligible to spend your budget, all three appearing in the search terms report so you can see what is happening.

The only way to find out which actual searches your money is buying is to look at the report. There is no other surface in Google Ads that exposes this. The keywords screen tells you what you bid on. The campaigns screen tells you the rolled-up totals. Only the search terms report tells you what was matched.

That ratio sometimes reaches five or six to one in accounts with broad match heavy targeting. The wider the match types, the more the report becomes the only honest record of what your account is doing.

Where to find it

The path moves around occasionally as Google reorganizes the interface. As of this writing, the way to get there is straightforward.

If your interface looks different from this, you are probably in the new Google Ads UX rollout, where some of the navigation has moved. The names stay roughly the same; the indentation may differ. Search "search terms" in the search box at the top if the menu hunting takes more than ten seconds.

What the report actually shows you

The default columns are search term, match type, added or excluded, impressions, clicks, cost, and conversions. The columns you actually use are search term, cost, and conversions. The rest are noise on the first sweep.

The match type column tells you whether Google considered the match an exact, phrase, or broad-match-and-close-variants. Most queries on a broadly targeted account show as "Broad match" in this column. That is not a defect; broad is the default for the match types you have probably chosen. You will not act on the match type column on the first sweep.

The added/excluded column shows whether you have already added the term as a keyword or as a negative. You will use this column rarely, mostly to confirm a negative actually applied. The "None" value, which is most rows, means the term came in via match-type expansion and is up for review.

The "review" flag in that table is the trickiest case. "Plumbing supply wholesale phoenix" looks like a wholesale buyer, which a residential plumber does not serve. But the word "wholesale" is also valid in legitimate contexts; somebody might be looking for the local supply house and accidentally find your ad. Phrase-block "supply wholesale" instead of single-word-blocking "wholesale" and you avoid both the false negative and the false positive.

The first sweep, top to bottom

The first time you do this on an unfamiliar account, expect to spend forty-five minutes. By the third sweep on the same account, you will be at fifteen.

The journaling habit is the one most people skip and then regret. Six months in, when you are wondering why a campaign's reach feels narrow, the answer is sometimes a negative you added in month two and forgot. Without a journal, you have to undo blocks one at a time and watch what happens. With a journal, you find it in thirty seconds.

What to do with the long tail

Below the top fifty rows, the search terms report becomes a long tail of low-cost, mostly-irrelevant queries. The temptation is to read it all. The right move is usually to set it down.

The math on the long tail is unforgiving. Two hundred rows, each spending two or three dollars, is six hundred dollars total. If half of that is wasted, your potential save is three hundred dollars over thirty days. The reading time to triage two hundred rows individually is roughly an hour. Most account owners' time is worth more than three hundred dollars an hour, especially when the same hour spent on the top fifty already returned a much bigger save.

There is a half-measure that is worth doing every quarter. Filter the report for queries with zero conversions and at least three clicks. That filter alone surfaces most of the long-tail leakage without forcing you to read every two-dollar entry. It usually adds another ten or fifteen minutes to the monthly sweep and uncovers a manageable second wave of negatives.

When the report looks empty or wrong

There are four common reasons the search terms report shows less than you expected.

The first is the privacy threshold. Starting in late 2020 and tightening since, Google only reports queries that exceed a privacy floor. Low-volume terms collapse into "(other)" or just do not appear. There is no fix for this; it is the data you have. Plan for the sweep to be incomplete on small accounts and don't try to extrapolate from the missing rows.

The second is a too-narrow date range. Seven days does not have enough data on most accounts to show the patterns the thirty-day window does. Switch to thirty.

The third is a campaign filter that hides what you wanted to see. The report respects the active filter at the top of the page. Clear it, then try again.

The fourth, and the most subtle, is a Performance Max campaign that is eating most of the spend in the account. PMax does not surface a search terms report in the same way Search campaigns do. You will see a "search categories" view instead. The takeaway is that if your account is mostly PMax, the search terms report you do see is showing you only the slice from non-PMax campaigns, and most of your spend is invisible to this workflow.

How often is often enough

Three numbers, three cadences.

Under five hundred a month in spend, monthly sweeps are fine. The volume is not high enough to surface much new pattern between weeks, and the leakage cost of waiting two extra weeks is small in absolute dollars.

Five hundred to five thousand a month, every two weeks. This is where most small-business accounts live, and it is also where the leakage compounds fastest. Every two weeks is the cadence that keeps the obvious blocks fresh without becoming a weekly chore.

Above five thousand a month, weekly. The pattern shifts faster, the leakage in absolute dollars is bigger, and the marginal hour of work pays back inside a week.

There is a fourth case worth flagging: any campaign in its first thirty days. Sweep it weekly regardless of spend. Match types are still settling. Quality scores are still being learned. You will block more obvious junk in the first month than you will in any single later month.

What does not belong in this workflow

Three things are tempting to do during a search terms sweep and almost always wrong to do.

Bid changes. The search terms report is a triage screen. Bid changes need their own time and a different mental model. Doing them mid-sweep is how you end up six items deep into a different problem with the original list half-finished.

Ad copy edits. Same logic. If a row makes you want to write new ad copy, write a one-line note in the journal and come back to it after the sweep is done. Otherwise the sweep ends with two negatives applied and forty rows still un-triaged.

Pausing keywords. A poorly converting search term is not a poorly converting keyword. Different surface, different decision. Pausing keywords belongs in the keywords screen with its own data and its own pace.

The discipline of doing one thing at a time makes the sweep faster and the decisions cleaner. Save the other work for the screens it belongs to.

What good looks like, by month four

After four months of regular sweeps, the search terms report will look quieter. The top rows will be familiar variations on real buyer queries. The obvious junk will be gone. The remaining ambiguity will narrow to the queries where you genuinely cannot tell whether the searcher was a buyer or a tire-kicker.

That quietness is the goal. It does not mean the work is done. It means the work has shifted from finding leaks to finding edge cases, and the cadence can drop accordingly. A campaign that was sweeping every week in month one can move to every two weeks by month two and monthly by month four. Most steady-state accounts settle into a monthly sweep that takes ten to fifteen minutes.

If the report still looks chaotic at month four, the cause is usually one of three things: match types that are too broad, ad copy that misrepresents what the business sells, or a campaign structure that mixes intents that should be split. Each of those is a separate piece of work, and each will be covered elsewhere in this section.