# How to investigate a sudden ad performance change with an AI agent

> Tier: deep. Full contract: quantified decision rules and evidence requirements.

Spend stopped, CPA doubled, conversions vanished: a fixed investigation order that finds the cause instead of the first plausible story.

**Intent.** Name the cause of a sudden change, with evidence, in minutes, checking causes in the order of their base rates.

## When to use

- Any 'why did X suddenly change' question.
- An alert fired and the dashboard offers no reason.

## When not to use

- Slow drifts over weeks: use account-audit or creative-fatigue; anomaly hunting on a trend finds coincidences.

## Sources

Required: Google Ads
Optional: Meta Ads, TikTok Ads, Google Analytics 4

## Preconditions

- The anomaly is defined: metric, direction, date, scope, before any tool call.

## Inputs

- `anomaly` (required): What changed, when, on what scope.
- `platform` (required): Where it happened.

## Procedure

1. Confirm the anomaly in the data: exact date, magnitude, scope. Rule out a reporting artefact first. — tools: `google_ads_get_insights`, `meta_get_insights`, `tiktok_get_insights`
2. Check humans first: the change log around the date. — tools: `google_ads_get_change_events`, `meta_get_ad_activity`
   Most anomalies are edits. Attribute and date them before theorising.
3. Check the platform: delivery status, disapprovals, learning resets. — tools: `tiktok_get_delivery_status`, `meta_get_delivery_diagnostics`
4. Check measurement: did tracking move rather than performance. — tools: `ga4_run_report`, `ga4_get_key_events`
5. Check the market: CPM and auction pressure on the same window. — tools: `google_ads_get_insights`
6. Stop at the first cause that explains magnitude and date, and say which checks were not needed.

## Decision rules

- **When** A logged change lands within 24h of the anomaly and touches the affected scope → **conclude** Primary suspect; verify magnitude fits before closing.
- **When** Conversions fell while clicks and spend held → **conclude** Investigate measurement before performance: the funnel rarely breaks that cleanly on its own.
- **When** CPM rose sharply across campaigns at once → **conclude** Market pressure, not account error; name the affected inventory.
- **When** The anomaly date is within the last 72 hours and the metric involves conversions → **conclude** Attribution lag is a candidate cause; re-check after the lag before escalating.

## Evidence requirements

- The confirmed anomaly: metric, dates, magnitude, scope, before any cause is discussed.
- The cause names its evidence: a change event, a delivery status, a diagnostic, never only a correlation.

## Human approval boundary

- Reverting someone's change is a human decision, then a Safe Write.
- Policy appeals and billing issues route to the platform; the skill identifies, humans escalate.

## Failure modes

- The first plausible story ends the search while the real cause survives: the order exists to prevent exactly this.
- Timezone mismatch between platforms makes 'the same day' off by one; align before matching events to effects.
- Google's change history stops at 30 days; older anomalies lose their best evidence source.

## Output

- **Anomaly confirmed**: Metric, date, magnitude, scope.
- **Cause**: The finding with its evidence, and the checks that ruled alternatives out.
- **Remedy**: Proposed fix, its owner, and the Safe Write if one applies.

## Related

- Skill: [tracking-audit](https://www.getmcpads.com/skills/tracking-audit.md)
- Skill: [weekly-performance-review](https://www.getmcpads.com/skills/weekly-performance-review.md)
- Guide: [campaign-audit](https://www.getmcpads.com/guides/campaign-audit.md)