> ## Documentation Index
> Fetch the complete documentation index at: https://docs.nerve.mattemena.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Review creative performance and fatigue

> Compare creative results over time, investigate possible fatigue with enough context, and turn findings into a test plan without automatic campaign changes.

Use a creative review to identify assets worth investigating or testing. A decline in one metric is a signal to examine, rather than proof that an ad is fatigued.

## Run a focused review

1. Choose the account, objective, conversion result, and period.
2. Ask for ad-level performance and a comparable earlier period.
3. Include available spend, impressions, frequency, clicks, conversion volume, and efficiency metrics.
4. Ask which creative information the assistant can actually inspect.

```text theme={null}
Review creative performance for Example Brand on Meta over the last
14 complete days versus the previous 14. Compare ads by spend,
impressions, frequency, CTR, purchases, and CPA where available.
Flag possible fatigue only with supporting trends and enough volume.
Separate observed metrics from hypotheses about the creative itself.
Recommend three tests. Do not pause, duplicate, or edit ads.
```

## Read the findings

Check whether a poor result reflects enough spend and conversions to judge. A new ad with little delivery is different from a high-spend ad whose efficiency deteriorated. Compare like objectives and result definitions.

Frequency can provide context, but no universal frequency threshold establishes fatigue for every campaign. Consider audience size, placement, time period, and response trends. Ask the assistant to identify alternative explanations, such as tracking changes or shifts in delivery.

## Build a test plan

For each recommendation, ask for the asset or message to test, the hypothesis, success metric, budget constraint, and review period. If the assistant cannot inspect the image or video content, supply it through an external assistant that supports attachments and distinguish that review from connected performance data.

Use [campaign-change review](https://docs.nerve.mattemena.com/campaign-management/review-changes.md) before implementing any pause, duplication, or budget recommendation.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.