SuperAPI AI Directory

social-performance-review-codex

Marketing & growth

View repository

Updated 10-16 10:35

About this skill

Social Performance Review For Codex

Analyze performance and feed learnings into next month.

Inputs

Accept:

  • CSV exports
  • Screenshots
  • Manual metrics
  • Platform summaries
  • Previous context/best-performers.md
  • context/content-calendar.md

Optional sources:

  • Scheduling tool exports from Blotato, Buffer, Later, Metricool, or Hootsuite.
  • Native platform exports from Instagram, LinkedIn, Facebook, TikTok, Threads, X, or YouTube.
  • Manually maintained reporting spreadsheets.

Metrics

Use available metrics only. Common fields:

  • Reach/impressions
  • Engagements
  • Engagement rate
  • Saves
  • Shares
  • Comments
  • Clicks
  • Follows
  • Leads or conversions

Recommended CSV columns when the user can choose an export format:

date, platform, post name, format, topic, reach, impressions,
engagements, engagement rate, saves, shares, comments, clicks,
follows, conversions, post URL, media file

Normalize comparisons by platform. Do not compare Instagram reach directly with LinkedIn impressions as if they mean the same thing. If data is incomplete, state the limitation and focus on directional patterns.

Output

Save report to outputs/reviews/:

# Social Performance Review - [Month]

## Executive Summary

## Top Performers
| Rank | Platform | Post | Why it worked | Repeatable pattern |
|---|---|---|---|---|

## Bottom Performers
| Platform | Post | Likely issue | Fix |
|---|---|---|---|

## Patterns
Pillars:
Formats:
Hooks:
Visuals:
Timing:

## Recommendations
1.
2.
3.

## Next Month Inputs

Update context/best-performers.md with repeatable patterns, not just winning post names.

Also update context/review-history.md if it exists, or create it when this is the first review.