SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsmeta-content-audit

meta-content-audit

General & other

View repository

Updated 10-16 08:04

About this skill

Content Audit

A read-only review of a client's published posts from exported platform data: performance by platform against Uganda / East Africa benchmarks, top and bottom posts, pillar coverage, tone and consistency, and five 30-day fixes. Metric findings are paired with narrative, audience-empathy and readability checks.

Use When

  • The client has months or years of posts and wants them rated by reach and engagement: which themes underperformed, which posts earned results and which wasted effort.
  • Posts need sorting into keep, update, retire (delete) or test decisions before anything is reused or a new plan is written.
  • Pillar balance, tone consistency or narrative quality looks uneven and needs rating against the brand voice.
  • A read-only review of a content library is required from exported platform data, without touching live accounts.

Do Not Use When

  • 02-platform-audit for auditing profiles, bios, links and account set-up.
  • meta-competitor-analysis for comparing named rivals.
  • meta-content-repurposing for turning the best posts into new formats.
  • Stop when no dated platform export is supplied; request it rather than estimating performance.

Required Inputs

Artefact Source/provider Required? If absent
Post-level data for the audit period (default last 3 months, at least 30 posts per platform) Native analytics exports completed on the data template Yes Stop; hand over the template and request the export rather than estimating performance. Extend to 6 months where a platform has fewer than 30 posts.
Platforms to audit Client brief Yes Audit only platforms with an export; mark the rest not assessed.
Client name, industry, country/city and primary goal Client brief Yes Default to Uganda / Kampala; ask the goal before prioritising fixes.
Content pillars 10-content-pillars output If established Cluster posts into 3–5 topic groups and recommend them as draft pillars.
Brand voice guide 04-brand-voice-intake output If available Rate tone against observed patterns and state that no guide was supplied.
Representative items with renders, sources and approvals Client content library For the narrative audit Mark narrative, accessibility and permission checks not assessed.

Workflow

  1. Collect intake and the post-level data (audit data and output template § Step 1); stop if no dated export exists, and route profile or set-up audits to 02-platform-audit.
  2. Sort posts by engagement rate descending and build the performance summary per platform against the Uganda / EA benchmarks; flag any platform below benchmark.
  3. Analyse the top 5 posts (ties broken by absolute reach) and the bottom 3 (excluding zero-reach posts), with specific, distinct reasons and one instruction each.
  4. Check pillar coverage against the 10-4-1 rule, or cluster posts into draft pillars when none exist.
  5. Rate visual, tone and posting consistency on 1–10 with evidence-based justifications.
  6. Apply the narrative, audience empathy and content quality audit to representative items and publish the capped audit plus the 95/100 remediation plan.
  7. Write exactly 5 priority improvements for the first 30 days, highest impact first, covering format, consistency and pillar balance.
  8. Run the quality standards and anti-slop gate; correct any finding without a data basis and rerun the check. Withhold release while a metric is invented or a missing check is shown as a pass.

Outputs

Artefact Consumer Acceptance condition
Performance summary per platform Client lead Every metric row filled from supplied data and compared with the Uganda / EA benchmark.
Top 5 and bottom 3 post analysis Content team Each post has distinct reasons and one specific replicate or avoid instruction.
Pillar coverage and tone/consistency ratings Client lead; 10-content-pillars Pillars over 40% or under 10% flagged; each score justified with evidence.
Narrative quality audit (capped score) and 30-day improvement list Client lead and next workflow owner Audit score published as min(raw score, 65); 5 improvements each traced to a finding.

Evidence Produced

Evidence Format Acceptance condition
Completed post-level data table Spreadsheet or table per platform Source export and period stated; engagement rate = engagement ÷ reach × 100.
Not-assessed register Table: check, reason, recovery evidence Missing renders, sources, permissions or native-language reviews listed, never counted as passes.

Capability and Permission Boundaries

Read and search only; analysis is read-only. Publishing, spend, live account changes, outreach and personal-data processing need explicit, action-specific client authority. Retiring or editing a post is the client's decision after the audit; the audit never deletes or changes live content.

Degraded Mode

Without dated post-level exports, return the narrowest qualified result and mark the affected checks not assessed. The data template, the benchmark table and a qualitative tone and pillar review of supplied screenshots can still be delivered, labelled as unscored.

Decision Rules

Condition Action Failure or risk avoided
The requested outcome belongs to meta-competitor-analysis Route there and hand over the verified inputs already collected. Neighbour collision and duplicated work.
A platform has fewer than 30 posts in 3 months Extend the audit period to 6 months. Patterns read from too small a sample.
A bottom-ranked post had zero reach Exclude it and log it as a technical issue, not a content failure. Blaming content for a delivery fault.
A pillar exceeds 40% of the mix or falls below 10% Flag over-reliance or neglect against the 10-4-1 rule (Bodnar and Cohen, 2012). An unbalanced, over-promotional feed.
A consistency dimension scores 1–4 Recommend addressing it before further investment in content production. Scaling inconsistent content.
Engagement is high but the offer or next step is unclear Inspect narrative and information hierarchy before calling the content effective. Rewarding posts that do not move the audience.
A post was paid-boosted Record "Paid boost? Yes" and name paid amplification as a factor when explaining its performance. Crediting paid amplification to content quality.

Quality Standards

  • Uses only data provided by the consultant; no invented metrics or engagement rates without a stated basis.
  • Platform summaries compare against Uganda / EA benchmarks, not just internal averages.
  • Top and bottom post analysis gives specific and distinct reasons, not the same generic factors repeated.
  • Pillar coverage flags imbalances clearly with reference to the 10-4-1 rule.
  • Tone and consistency ratings are justified with evidence from the content, not assigned arbitrarily.
  • The 30-day improvements are prioritised by impact and each traces to a specific audit finding.
  • Output is direct and honest: underperformance is named, not softened.
  • Metric findings are paired with audience empathy, narrative clarity, hierarchy, readability/accessibility, evidence, permissions, AI transparency and learning-value findings.

Anti-Patterns

  • Estimating performance when no export was supplied. Fix: request the dated export and hand over the data template.
  • Repeating the same generic reasons ("good visuals") for every top post. Fix: name distinct factors per post from the list in the template.
  • Softening the bottom-post diagnosis. Fix: state the likely cause plainly; honest diagnosis prevents repeated mistakes.
  • Calling content effective because engagement is high while the audience cannot understand the offer or next step. Fix: inspect the narrative and information hierarchy.
  • Treating a missing render, source, permission or native-language review as a pass. Fix: mark it not assessed, state the risk and name the recovery evidence.
  • Judging YouTube by engagement rate alone. Fix: measure watch-time completion % and click-through rate.

References