Knowledge Audit to Copilot Opportunities Mapper
Quickly inventory internal knowledge and surface high-ROI copilot use cases by function, with a de-risked, prioritized backlog and next-step specifications.
Prompt Overview
Featured AI Partner
Tips For You
- Limit scope to top 3 roles and their most repeated tasks for a focused first sprint. - Treat unknown update cadences as risks and add owners before building. - Require explicit acceptance criteria for each backlog item to avoid scope creep. - Use cost-of-delay to break ties on near-equal priorities.
From Operations TeamNexusAi TechnologyProblem It Solves
Teams don’t know where to start or which copilot to build first. This prompt runs a structured audit and ranks opportunities with clear scope and ROI.
Scored Opportunity Backlog
Ranks copilot ideas by value, feasibility, and risk with cost-of-delay.
Traceable Sources
Citations, owners, and update cadences for every recommendation.
Actionable Specs
Inputs, retrieval, and acceptance criteria per backlog item.
Risk Flags
Highlights compliance and data-quality risks early.
AI Prompt Instructions
Act as: An AI Product Analyst and Knowledge Architect specializing in converting organizational knowledge into safe, high-ROI copilots.
Why this task matters: Most teams have scattered SOPs, docs, and tribal knowledge. A quick, structured audit identifies where a copilot removes the most friction and risk, so we build the right thing first.
Important boundaries:
- No speculative features; propose only use cases grounded in existing knowledge assets and measurable workflows.
- Require citations, ownership, and update cadence for all sources.
- Flag compliance, privacy, or IP-sensitive areas for governance review.
User inputs:
- Repositories/locations (links, paths)
- Target roles and top 5 recurring tasks per role
- Business goals (e.g., reduce handling time, increase first-contact resolution)
- Constraints (PII, regulated content, languages)
- Available tools (ChatGPT, Claude, Gemini, Cursor, Notion, n8n)
Objectives:
1) Inventory knowledge and classify by reliability, freshness, and authority.
2) Map tasks to copilot patterns (FAQ, SOP executor, tool-augmented agent, RAG explainer).
3) Estimate ROI, risk, and cost-of-delay.
4) Produce a prioritized backlog with crisp acceptance criteria and metrics.
Analysis workflow:
1) Catalogue sources with owner, update frequency, and trust level.
2) Cluster tasks by frequency, variability, and failure impact.
3) Match each task cluster to a copilot type and minimal viable scope.
4) Score options (value, feasibility, risk, dependency) and compute priority.
5) Draft specs: inputs, prompts, retrieval, tool calls, guardrails, KPIs.
6) Identify missing data, canonical source of truth, and governance needs.
Required output format:
- Section A: Knowledge Inventory (table)
- Section B: Task→Copilot Mapping (table)
- Section C: Scored Prioritized Backlog (JSON)
- Section D: Next-step Specifications (bullets with acceptance criteria)
Quality controls:
- Every proposed use case must cite exact sources and owners.
- Include red flags and assumptions per backlog item.
- Include a simple ROI and cost-of-delay rationale.
Verification checklist:
- Are top 3 time sinks addressed?
- Are compliance-sensitive areas flagged?
- Are acceptance criteria testable?
Final instruction: Produce the full output with explicit citations and owners. End with 5 clarification questions that would most improve accuracy before execution.
Expected Outcome
Section C: Backlog (excerpt) 1) Support FAQ Copilot (Priority 9.2): Sources: Zendesk macros, Product Handbook v3. Owner: CX Ops. KPI: FCR +15%. 2) Onboarding SOP Copilot (Priority 8.1): Sources: HR SOPs. Owner: HR Ops. KPI: Time-to-productive -30%.
Implementation Journey
Run the audit in ChatGPT or Claude
Open ChatGPT or Claude. Paste links or pasted excerpts from your SOPs, handbook, support macros, and wiki pages plus your top 5 recurring tasks per role. Ask the model to execute the full prompt and return the Knowledge Inventory, Task→Copilot Mapping, and Prioritized Backlog with owners and KPIs.
25-40 minStructure the backlog in Notion and Sheets
Copy the Backlog JSON into Google Sheets to sort by priority and cost-of-delay. Push the final list into a Notion database with properties for owner, KPI, acceptance criteria, and due date. Link each item to its source doc for traceability.
20-30 minAutomate source freshness checks with n8n
Use n8n to check file timestamps or Notion update dates weekly. If a source is stale or owner missing, create an issue in GitHub or notify the owner in Slack. This keeps your backlog valid as content changes.
30-45 min