Keyword Bid Health
Weekly keyword-level review: surface scale-up candidates with proven conversion volume and pullback candidates running high ACoS.
or just say “run keyword bid review for <brand>”
Comes with the plugin, one install activates every skill. See the install flow. Needs brand setup first.
Deciding which keywords to cut, which to push harder on, and which quietly went dark takes a manual pass through every keyword’s T-30 and T-7 numbers each week, and it is easy to miss the ones that stopped spending entirely.
Three ranked tables (bid-cut candidates, scale opportunities, and dormant keywords), each with a recommended bid, so the weekly bid pass is a review instead of a rebuild.
What your agent can do
- 01
Rank scale-up candidates by demonstrated conversion velocity and headroom
- 02
Surface high-ACoS keywords for bid reduction with revenue-impact estimates
- 03
Detect bid-floor and bid-ceiling drift relative to placement performance
- 04
Output a prioritized work queue ready for execution
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 →- Scale and pullback threshold percents plus ACoS target: the bid math every recommendation runs on
- Brand terms: separates brand keywords from nonbrand headroom rules
- Posture stance and multiplier: how aggressively recommended bids move up or down
- Active structural events (price tests, promotions, stockouts, recent bid changes): annotate a recommendation instead of hiding it
As thresholds get calibrated and structural events accumulate on the brand timeline, the recommended cuts and raises stop reacting to swings the team already explained, and week-over-week drift becomes something the report can point to.