Skill

Documentation Gap Analysis

Compare existing docs against real user questions to surface ranked documentation gaps that cost your team the most time.

Goal

Identify, categorize, and rank documentation gaps by frequency and time cost, so the team knows exactly what to write or fix first.

Trigger

Run quarterly, after a support volume spike, or before a major product release to audit documentation coverage.

Steps

  1. 1

    Using your connected tools (for example support tickets from Zendesk/Jira, community forum threads, Slack help channels, or chat transcripts), pull all user questions and support interactions from the last 90 days. If no tools are connected, ask the user to paste a representative sample of questions or export a CSV of tickets.

    • At least 50 user questions or tickets collected
    • Source and date range noted
    • Fallback paste confirmed if tools aren't connected
  2. 2

    Pull the current documentation set from the connected knowledge base or docs tool (e.g. Notion, Confluence, GitBook, or a public docs URL). Index every article, its title, and its primary topic so you have a complete map of what currently exists. If not connected, ask the user to paste a sitemap, table of contents, or list of doc titles.

    • Full list of existing doc titles and topics captured
    • Docs source and last-updated date noted
  3. 3

    Cluster the collected user questions into thematic groups (e.g. setup, billing, feature X, error codes). For each cluster, attempt to match it to an existing doc. Label each cluster as: Fully Covered, Partially Covered, or Not Covered. Note the question volume per cluster.

    • Every question cluster assigned a coverage label
    • Question volume counted per cluster
    • Partial coverage cases flagged with specific missing detail
  4. 4

    Estimate the time cost per gap. For each Not Covered or Partially Covered cluster, calculate: (number of questions in cluster) × (estimated average handle time per question, defaulting to 10 minutes if unknown). Rank all gaps from highest total time cost to lowest to create a prioritized gap list.

    • Time cost estimated for every gap
    • Gaps sorted by total time cost descending
    • Default handle-time assumption stated if used
  5. 5

    For the top 5–10 gaps by time cost, write a one-paragraph brief for each: what users are actually asking, what's missing or misleading in the current docs (or what doesn't exist), and a recommended action — create new article, expand existing section, or add an FAQ entry. Include a suggested title and the owning team or persona most likely to write it.

    • Each brief includes user question examples verbatim
    • Recommended action is specific (create vs. expand vs. FAQ)
    • Suggested doc title provided for each gap
  6. 6

    Produce a final ranked Gap Analysis Report: an executive summary of overall coverage health (% fully covered, partially covered, not covered), a ranked gap table, and the top-5 action briefs. Conclude with a 30-day quick-win recommendation — the single highest-ROI doc to create or fix first.

    • Coverage percentages included in summary
    • Ranked gap table present with volume and time-cost columns
    • Single 30-day quick-win recommendation clearly stated

Output format

Deliver a structured report with four sections: (1) Coverage Health Summary with percentages, (2) Ranked Gap Table with columns for topic cluster, question volume, estimated time cost, and coverage status, (3) Top-5 Gap Briefs each with user examples, missing content description, recommended action, and suggested title, (4) 30-Day Quick-Win — the single highest-priority doc to create or fix.

Use it everywhere

Copy this skill into your library to inject it into Claude, ChatGPT, and Gemini — or install your whole library as / slash commands in Claude Code and Cowork.

Get started free →