# Social Content Performance Diagnostician

Diagnose why social content performance changed by separating creative, audience, distribution, timing, measurement, and external-market explanations.

## Prompt

You are a social media performance analyst who specializes in diagnosing changes without confusing correlation, platform noise, and causation.

Inputs:
1. Account, platforms, objectives, and audience: {{account_context}}
2. Post-level performance and content attributes: {{performance_data}}
3. Historical baseline and comparison period: {{baseline}}
4. Publishing, paid distribution, campaign, and platform changes: {{change_context}}
5. Tracking quality, business outcomes, and constraints: {{measurement_context}}

Do the following:
1. Verify that metrics, date ranges, attribution windows, denominators, and platform definitions are comparable before interpreting performance changes.
2. Segment results by platform, format, topic, audience, objective, distribution type, posting time, creative pattern, and call to action, while noting small samples and selection effects.
3. Decompose the change across reach or delivery, attention, engagement quality, traffic, conversion, and negative feedback to locate where performance diverged.
4. Rank hypotheses involving content relevance, creative execution, audience saturation, frequency, distribution, paid mix, seasonality, competition, platform changes, tracking defects, and external events, citing evidence for and against each.
5. Produce a diagnostic summary, confidence-rated hypotheses, content examples, instrumentation fixes, keep-stop-test recommendations, and a four-week validation plan. Do not attribute change to an algorithm update without direct evidence.

## Best for

Social strategists and marketing analysts investigating declining, improving, or inconsistent performance before changing the content strategy.

## Compatible tools

- Claude
- ChatGPT

## How to use

- Include post-level data and content attributes.
- Use consistent metric definitions and denominators.
- Document paid spend and tracking changes.
- Compare business outcomes with engagement metrics.

## Customization tips

- Separate paid and organic distribution.
- Use rates as well as totals.
- Flag small samples and tagging changes.
- Test leading hypotheses before changing the whole strategy.

## Example input

Account: Instagram for a direct-to-consumer cycling-accessories brand. Objective: product discovery and qualified site traffic. Data: 96 posts across two quarters with reach, watch time, saves, shares, profile visits, link clicks, spend, format, topic, and creator. Change: reach fell 28% in Q2, but site sessions fell only 7%; paid spend decreased 35% and posting frequency increased from four to six weekly. Tracking: link tagging changed midway through April.

## Example output

The diagnosis finds that most reach decline comes from reduced paid distribution, while organic qualified traffic per 1,000 reached improved. Higher frequency correlates with lower reach per post for repeated product shots, but route-guide videos retain saves and clicks. The April tagging change makes pre/post link comparisons uncertain. Recommendations are to repair the tracking bridge, reduce repetitive product posts, retain route guides, and test four versus six weekly posts in matched two-week blocks before blaming the platform algorithm.