What data Search Terms uses

What it reads from Google Ads, and how far back.

Search Terms works from your real Google Ads performance - nothing is invented or assumed.

What it reads

  • Your search-term report for roughly the last 90 days - every term, with its impressions, clicks, cost, conversions, conversion value, match type, and the exact keyword that triggered it.

  • Your existing keywords, negative keywords, and attached shared negative lists, so it never suggests adding a keyword you already have, or blocking a term you're deliberately bidding on.

  • Your targets and bidding signals - used to decide whether each campaign is judged on cost (CPA) or value (ROAS).

Why 90 days

It's long enough to see real patterns and gather enough conversions to trust, but recent enough to reflect how the account behaves now. Terms that haven't yet earned enough clicks, spend or conversions to call confidently are held back as Pending or Monitor rather than guessed at.

info icon
Example. A term with 6 impressions and 1 click won't be judged either way - there isn't enough signal. The same term at 400 impressions, 30 clicks and 0 conversions is a clear negative candidate.

A note on conversion tracking

Because so much of the logic leans on conversions and value, accurate conversion tracking in Google Ads is what makes the recommendations sharp. If a campaign isn't tracking conversions, you'll still get relevance- and engagement-based suggestions, but the money-based ones need that data to fire.


RELATED ARTICLES

What is the Search Term Analyzer?

What it does, and the problem it solves.

CPA vs ROAS

How the mode is chosen and how each judges performance.

Did this answer your question?
😞
😐
😁