Log in / Create free account🌐 ES☀️

Basic training · Module 3 — Search I — keywords and match types

The search terms report: the weekly routine that pays for the account

⏱️ 10 min read · 🔎 Search 🥷 Scripts · updated on 2026-09-02

🎧 Listen to the lesson · ≈ 5 min🔒 Subscribers only

In short: the search terms report shows what people actually typed to trigger your ads. It is the only place where you can see whether your keyword hypothesis works. For each term there are only three possible decisions: add it as a keyword, add it as a negative, or leave it alone. Twenty minutes a week there is worth more than any other optimisation.

The whole module has been pointing here. Keywords are your hypothesis, match types widen it, negatives trim it back... and the search terms report is where it all gets checked against reality: what people typed, how much it cost you and whether it was any use. It is the most profitable report in Google Ads per minute spent on it.

Where is the search terms report and what does it show?

Campaigns → Insights and reports → Search terms (also from a specific campaign or ad group, to see only its own). Each row is a real term with its metrics (impressions, clicks, cost, conversions) and two key columns:

Two warnings for 2026:

  1. You do not see every term. Google hides low-volume ones for privacy reasons. In small accounts that can be a noticeable slice of the spend ("other search terms"). There is no fix; you just need to know.
  2. In Performance Max the report exists but is more limited (categories and terms with volume). We cover it at the Advanced level.

What should you do with each search term?

For every row worth your attention there are exactly three ways out:

Decision When Action
Add as a keyword It converts (or has a good CTR and cost) and is not on your list Add it to the right ad group, in exact or phrase match. You gain control and, often, a better QS
Add as a negative Wrong intent, a product you do not sell, a pattern that spends without converting Add the root as a negative at the right level (list, campaign or ad group)
Leave it Relevant, not enough data to judge Nothing. Come back next week

The third one is the hardest to accept and the most important: most terms do not need a decision.

How much data do you need to decide on a term?

A term with 3 clicks and 0 conversions says nothing: with a 5% conversion rate it is perfectly normal for 3 clicks to produce none. Practical rules to stop you reacting to noise:

How often should you review search terms?

  1. Open the report with the last 7 days date range (large accounts) or 30 days (small accounts), sorted by cost.
  2. Work down the highest-spending rows until you have covered ~80% of the cost. For each one: is the intent right? → if not, add a negative on the root.
  3. Filter for conversions > 0 and sort by conversions. Are there terms that are not keywords yet? → add them.
  4. Filter for low CTR with plenty of impressions: these are usually tangential terms (people see the ad and are not interested). Candidates for a negative.
  5. Note the new roots in the shared list if they apply across the whole account.

Every week the report gets a little cleaner, the terms look more like your keywords and the CPA drops. It is cumulative: six months of the routine turn a mediocre account into a clean one.

What does the report teach you about your own business?

Beyond optimisation, the report is free market research: what people call your product, what doubts they have before buying, which competitors they look up, which features matter to them. Plenty of FAQ pages, plenty of product names and more than one catalogue decision have come out of reading search terms.

💡 Ninja trick: the 20-minute routine is exactly what the Search Query Optimizer automates: every night it reads the terms, applies your scenarios (spend with no conversions above X times the CPA, low CTR with impressions, semantic irrelevance using AI), adds negatives for whatever meets the rules and proposes as keywords the ones that convert. You look at the spreadsheet on Monday and approve with a checkbox. And the rules from this lesson enforce themselves: it never touches anything that has converted and it demands volume before adding a negative on performance.

What you should remember

That is the end of Module 3. In Module 4 we move on to the other half of the ad group: writing ads that win the click (responsive ads and their assets).

📎 Sources and further reading

⚠️ Free training with no support. Ninja Scripts support channels (email and Telegram) are only for the use of the scripts, not for Google Ads questions or questions about this training.

Pick up here

← BeforeNegative keywords: the control you do keepSearch I — keywords and match typesRelacionadaAI Max Analyzer in practice: install it, read the verdict and actAI Max — the black box of automatic Search expansionRelacionadaNegativisation scenarios: rules that decide for you (and how to calibrate them)Advanced Search — search terms, negatives and traffic controlRelacionadaThe white list: controlling traffic the other way roundAdvanced Search — search terms, negatives and traffic controlRelacionadaArchitecture principles: how to design an account that can grow without being rebuiltAccount architectures for scale — real cases

Ver el temario completo

🎓
You're reading the Ninja Academy in the open

Create your free account (no card) to save your progress, take the graded quizzes, earn the Basic level certificate and unlock part of the Intermediate and Advanced training.

Create my free account →
🥷

Subscriber feature

This option is part of the Ninja Scripts Suite subscription.

See the subscription →