Search Term Negation
Four-phase search-term irrelevance analysis with surgical negative keyword strategy: the workflow our analysts run weekly.
or just say “run search term negation for <brand>”
Comes with the plugin, one install activates every skill. See the install flow. Needs brand setup first.
Every week someone has to comb the search term report for traffic that is the wrong category, wrong demographic, or otherwise irrelevant, and decide whether to negate it exact, phrase, or at the campaign level. Get the tier wrong and you either keep bleeding spend on obvious waste or block terms that were quietly converting.
A tiered negation review (clean negates, terms to keep, and a bucket for anything genuinely borderline) with the negative keyword change set already built and previewed as a dry run, so you confirm one exact set instead of assembling it by hand.
What your agent can do
- 01
Pull the window search-term report and join lifetime performance at three granularity levels
- 02
Suppress terms already negated and route the rest through semantic relevance classification
- 03
Separate clean negate candidates from review-and-watch buckets
- 04
Generate match-type-appropriate negatives (exact / phrase) with reasoning
Before you run
This skill reads a brand’s context to calibrate its output. Build it once per brand, then every teammate runs on the same shared context.
/mixshift-ai:mx-brand-context <brand>How brand setup works →- ACoS target: the reference efficiency every term is judged against
- Protected terms: brand anchors that are never put up for negation
- Lane rules: the relevant vs mismatch dictionaries that draw the borderline line
- Reporting window: the T-3 to T-7 span the search term corpus is pulled for
As protected terms, lane rules, and prior run history accumulate in brand context, fewer terms land in the review pile as borderline and the tier boundaries hold steadier from run to run, so more of the review resolves itself before an account manager has to weigh in.