Runaway Spend Check
Acute daily detection at the keyword level: catch T-1 spend spikes and zero-conversion runaways before your client does.
or just say “run runaway spend check for <brand>”
Comes with the plugin, one install activates every skill. See the install flow. Needs brand setup first.
Overnight keyword spend spikes and quiet zero-conversion days get buried in the noise of a full account, so by the time someone spots them the excess spend is already gone.
A short list of the keywords that actually breached the account’s own confidence intervals or died on conversions, ranked by excess spend, so you know exactly what to pause or cut today.
What your agent can do
- 01
Flag keywords whose previous-day spend materially exceeded historical baselines
- 02
Identify high-spend keywords with zero conversions vs their lifetime conversion rate
- 03
Sort by absolute dollar exposure so you act on the biggest bleeders first
- 04
Return surgical recommendations: bid down, pause, or negative
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 →- Pullback threshold percent and ACoS target: the bid-cut ceiling every T-1 breach is measured against
- Confidence interval bounds: the CI used to flag T-1 spend and ACoS breaches
- Active structural events (promotions, price tests, stockouts, recent bid changes): downgrade a flag instead of raising a false alarm
- Posture stance and multiplier: how much bid-cut math to recommend once a keyword is flagged
As the pullback threshold and ACoS target get calibrated and the team logs promotions, stockouts, and price tests on the brand timeline, the check stops flagging spend spikes it can already explain and tightens toward real excess spend.