workflow-pe-ai-readiness

# PE AI Readiness Assessment ## What this skill covers A structured diagnostic that identifies, gates, and ranks AI opportunities across PE portfolio companies by annualised EBITDA impact. Runs as a single-company assessment or a full cross-portfolio diagnostic on a quarterly cadence. **Core principle**: rank by dollars, not excitement. A back-office automation saving $400k beats a flashy customer chatbot. Hold period and data quality together determine urgency. ## Three-Gate Evaluation Apply all three gates before scoring. A single "No" stalls the opportunity to "Wait" status. | Gate | Question | Pass Condition | |------|----------|----------------| | G1 — Data Readiness | Can clean, structured inputs be sourced without a multi-month data-engineering programme? | Usable data exists today or within 30 days at negligible cost | | G2 — Organisational Readiness | Is there a named manager with authority, budget, and motivation to drive this to completion? | Owner identified; role confirmed; incentive aligned to outcome | | G3 — ROI Readiness | Does a defensible EBITDA case clear the hurdle for hold period and resourcing? | Payback inside hold period; impact >=$100k annualised | G2 is the binding constraint in the majority of portfolio situations — a quick win with no internal owner dies in 90 days. ## Opportunity Scoring Rubric Score 1-5 on four dimensions. Multiply scores to produce raw score (max 625). | Dimension | 1 (Low) | 3 (Medium) | 5 (High) | |-----------|---------|-----------|---------| | Annualised EBITDA Impact | <$100k | $100k–$500k | >$500k | | Time-to-Value | >6 months | 2–6 months | <60 days | | Hold-Period Fit | Payback ≥ remaining hold | Payback < remaining hold | Payback < 25% of remaining hold | | Data-Quality Dependency | Significant cleansing required | Partial — some work needed | Clean and accessible today | Use Ownership Strength (1-5) as a tiebreaker, not a multiplier. ## EBITDA Impact Quantification To compute the EBITDA impact estimate: - Call `variance_analysis` to decompose the cost or revenue line the use case targets - Call `peer_benchmarking` to anchor achievable margin level (gap to top-quartile = upper bound on impact) - Call `dupont_analysis` to identify whether the lever flows through margin, asset turnover, or leverage - Call `working_capital` for back-office automation use cases (impact on DSO/DIO/DPO) - Call `sensitivity_matrix` to stress-test impact across optimistic/base/conservative adoption assumptions ## Common Opportunity Archetypes | Archetype | Typical EBITDA Impact | Data Prerequisites | |-----------|----------------------|--------------------| | Sales productivity (proposal drafting, CRM, lead scoring) | $200k–$1.5m | CRM with ≥12 months clean activity data | | Customer support automation (ticket deflection, agent assist) | $300k–$2m | Ticket history ≥90 days; resolution outcomes | | Finance / back-office automation (AP, GL, contract abstraction) | $150k–$800k | Invoice scans; chart of accounts; AP system export | | Predictive maintenance (industrial/fleet assets) | $400k–$3m | Telemetry ≥6 months; maintenance log; asset register | | Fraud / anomaly detection | $200k–$1.5m | Transaction ledger ≥18 months; known-fraud labels | | Demand forecasting (consumer/retail) | $250k–$1.5m | POS history ≥24 months; promotional calendar | | Code generation (software portcos) | 10-25% engineering productivity uplift | Repo access; CI/CD telemetry | ## Cross-Portfolio Diagnostic Cadence **Quarter-long cadence (5-20 companies)**: - Week 1: scope, intake, confirm company list, pull quarterly financials, request operating-team use-case memos - Weeks 2-6: per-company gating and scoring (half-day per company) - Weeks 7-8: portfolio synthesis, rank GO opportunities, identify replay candidates - Week 9: top-3 deep-dive memos (2-3 pages each) - Weeks 10-11: challenge session, revise scores - Week 12: deliverable lock, WAIT-item watchlist, next-quarter rollover **Replay identification**: two or more companies in the same sector where the same tooling can be deployed with configuration-only changes. Document combined EBITDA impact and shared implementation cost saving. ## Output format **Ranked portfolio table** (one row per GO opportunity): | Company | Use Case | Gate Status | EBITDA Impact (ann.) | Speed | Score | Priority | Owner | |---------|----------|-------------|----------------------|-------|-------|----------|-------| Gate status: **GO** (all three gates pass), **WAIT** (one or more gates fail — record blocker), **PASS** (declined — record reason). **Top-3 deep-dive memos** (2-3 pages each): business context, scoped pilot, data prerequisites, owner and resourcing, 30/60/90-day plan, sensitivity grid output, success metric. **Portfolio roll-up**: total addressable EBITDA impact if all GO opportunities execute; apply 60-70% capture probability to GO items; 0% to WAIT. ## Quality gates - [ ] Three-gate evaluation documented for every assessed opportunity - [ ] Scores computed from four-dimension rubric (not qualitative judgement alone) - [ ] EBITDA impact anchored to `peer_benchmarking` and `variance_analysis` outputs - [ ] Replay candidates identified where applicable - [ ] GO items with EBITDA impact >$250k reflected as VCP line items - [ ] Failure modes checked before finalising the ranked table ## Common failure modes to check - Data-quality blockers presented as solvable (force specific owner/date/budget) - Change-management debt (recent ERP, leadership turnover) — haircut capture probability - Vendor lock-in without exit clause — score lock-in as downside scenario - Pilot-to-production gap — apply 60-70% capture probability, not 100% - EBITDA double-counting with active VCP initiatives — reconcile before submission ## Related skills - `workflow-pe-value-creation-plan` — AI opportunities scoring GO with EBITDA >$250k become VCP line items - `workflow-pe-portfolio-monitoring` — tracks GO opportunity delivery quarter-by-quarter - `workflow-pe-ic-memo` — AI readiness assessment cited as a VCP lever in Section V (Investment Thesis)