Skip to main content

​
What is an AI Readiness Assessment?

An AI readiness assessment is a structured evaluation of whether a company — or group of portfolio companies — is prepared to adopt AI tools and where AI can create the most economic value. For PE firms, this has become a critical component of the value creation toolkit. AI-driven automation and intelligence can reduce operating costs, improve customer experience, accelerate revenue, and enhance exit multiples. The key insight that separates effective AI deployment in PE from hype-driven initiatives is a focus on practical readiness over technological ambition. A portfolio company does not need a data science team or a sophisticated tech stack. It needs three things: (1) clean data inputs for the specific use case, (2) a management team champion who will own implementation, and (3) the ability to run a pilot in 30 days using off-the-shelf tools. Companies that meet all three criteria are “Go.” Companies that fail any gate are “Wait” until the specific blocker is resolved. The portfolio-level view adds another dimension: replays. If one portfolio company successfully deploys AI for invoice processing, every other portfolio company with significant AP volume can copy that playbook. This is the highest-leverage move in PE AI strategy — one implementation, multiple deployments.

​
Why It Matters

AI is increasingly a factor in PE exit valuations. Companies that have successfully adopted AI tools are perceived as more operationally sophisticated, more scalable, and better positioned for future growth. Conversely, companies that have not adopted AI where competitors have may face valuation discounts. For operating partners, the AI readiness assessment provides a prioritized action list. Instead of guessing which company should get AI investment first, the assessment ranks every opportunity across the entire portfolio by dollar impact, speed to value, and probability of success. At the portfolio level, AI readiness directly impacts two things PE firms care about deeply: EBITDA improvement and exit narrative. A company that can demonstrate successful AI adoption — even in back-office processes — signals operational maturity to prospective buyers. During exit, this can translate to a 0.5-1.0x multiple premium for demonstrating a modern, scalable operating model.

​
Key Concepts

TermDefinition
Go / Wait GateA three-question framework to determine if a company is ready for an AI pilot: is the data there, is there an owner, can you pilot in 30 days
ReplayA successful AI deployment at one portfolio company that can be copied to others with similar characteristics
Off-the-shelf ToolingAI tools that can be purchased and deployed without custom development (e.g., document processing, CRM automation, AP/AR matching)
EBITDA ImpactThe estimated annual earnings improvement from an AI initiative, net of tool costs
Speed to ValueThe number of months from pilot launch to first measurable financial result
Hold Period UrgencyCompanies near exit need quick-hit AI wins that show up in LTM EBITDA for the CIM; companies early in hold can afford foundational projects
AI ChampionThe management team member who will personally drive adoption; without one, initiatives stall within 90 days
Data ReadinessWhether the company can produce clean, structured input data for a specific AI use case without a multi-month data project
Vendor LeverageWhen multiple portfolio companies buy the same AI tool, the PE firm can negotiate better pricing and terms

​
How It Works

1

Connect to Portfolio Data

Ingest quarterly updates, board decks, and financials for the portfolio (or a subset). For each company, extract: sector, revenue, headcount by function, tech stack, and any AI/automation initiatives already in flight. Ask about hold period remaining per company.If the user provides materials via MCP servers (data room, SharePoint, Google Drive), local files, or file uploads, work with whatever is available. If only a single company is provided, run the scan but skip the cross-portfolio ranking.Ask up front if not obvious from materials: hold period remaining per company, and whether any portfolio company has already deployed something that worked.
2

Per-Company Scan

Answer three gate questions for each company. All three yes = Go; any no = Wait with a note on what unblocks it.Gate 1: Is the data there? Can the company produce a clean input for the use case — customer list, invoice feed, contract repository — without a 6-month data project first?Gate 2: Is there an owner? Someone on the management team who will drive this — not a sponsor who will “support” it, but an owner who will be accountable.Gate 3: Can we pilot in 30 days? One team, one workflow, off-the-shelf tooling. If the answer starts with “first we would need to…”, it is not a quick win.Then identify the top 2-3 leverage points from these patterns:Back Office (usually fastest to pilot): Invoice processing, AP/AR matching, expense categorization. Contract abstraction — vendor agreements, leases, customer MSAs. Month-end close: reconciliations, flux commentary, lender reporting first drafts.Revenue / Front Office: RFP and proposal first drafts (big lever if revenue is project-based). Sales call summaries and CRM hygiene. Customer support ticket triage and first-response drafting. Quoting for configured or complex products.Operations (sector-dependent): SOP and quality documentation generation. Scheduling and dispatch (field services, logistics). Code generation and review (software portfolio companies).For each leverage point, capture in one line: what it replaces, FTE-hours per week saved (assume 30-50%, not 100%), and whether it is buy-off-the-shelf or needs a light build.
3

Rank Across the Portfolio

Stack every opportunity from every company into one list. Rank by dollar impact (annualized EBITDA contribution), speed to value (months to first result), and probability (discounted for data quality, change management, and management capability). Tiebreaker: favor companies with less hold period remaining.
4

Find the Replays

Scan for playbooks that work at multiple companies: same sector/same function, same tool/different company, and shared vendor leverage (multiple companies buying the same tool creates pricing power).
5

Output

One page for the operating partner: top 5 across the portfolio with owners and 30-day first steps, replays (2-3 playbooks hitting multiple companies), Go/Wait by company with blockers, what you are NOT doing (saves relitigating), and aggregate EBITDA contribution (Year 1 quick wins vs. Years 2-3 scale).

​
Worked Example: Five-Company Portfolio Scan

Below is a simplified numerical walkthrough showing how the assessment works across a five-company PE portfolio.

​
Portfolio Setup

CompanySectorRevenueEBITDAHeadcountHold Remaining
AlphaWidgetsIndustrial Mfg$80M$16M40018 months
BetaSaaSB2B Software$25M$5M1204 years
GammaCareHealthcare Svcs$60M$9M3503 years
DeltaLogisticsLogistics$45M$5.4M2802 years
EpsilonServicesBusiness Svcs$35M$6.3M2003.5 years

​
Step-by-Step Gate Assessment

AlphaWidgets (Industrial Mfg, $80M revenue)
  • Gate 1 (Data): Invoices are in QuickBooks, but the AP team uses paper purchase orders scanned to a shared drive. Data exists but is semi-structured. Partial yes.
  • Gate 2 (Owner): CFO is enthusiastic about automation, has budget authority. Yes.
  • Gate 3 (Pilot in 30 days): An AP automation tool can ingest scanned POs with 2 weeks of setup. Yes.
  • Verdict: Go (with note: data cleanup for POs needed in parallel)
  • Top opportunity: AP automation. 3 FTEs spend ~60% of time on invoice matching. At 40% automation, that saves 1.2 FTEs = ~72K/year.Toolcost 72K/year. Tool cost ~15K/year. Net EBITDA impact: $57K annualized.
BetaSaaS (B2B Software, $25M revenue)
  • Gate 1 (Data): Customer data is in Salesforce, product usage data is in their own platform. Clean. Yes.
  • Gate 2 (Owner): VP Engineering wants to use AI for code review. CRO wants AI for sales call summaries. Two potential owners. Yes.
  • Gate 3 (Pilot in 30 days): Code review tools and sales call AI are both off-the-shelf. Yes.
  • Verdict: Go
  • Top opportunity: Sales call summaries + CRM hygiene. 8 reps spend ~5 hours/week on notes and CRM updates. At 50% reduction, that is 20 rep-hours/week redirected to selling. Estimated revenue lift: 500K−500K-750K annually (assuming even modest conversion improvement). Tool cost: ~20K/year.EBITDAimpact:∗∗20K/year. EBITDA impact: **100K-$150K** (conservative, contribution margin on incremental revenue).
GammaCare (Healthcare Svcs, $60M revenue)
  • Gate 1 (Data): Prior authorization data is in an EHR system but exports are messy. No clean API. Partial.
  • Gate 2 (Owner): COO is interested but overwhelmed with a regulatory audit. No real bandwidth. No.
  • Gate 3 (Pilot in 30 days): Prior auth automation requires EHR integration that takes 60-90 days minimum. No.
  • Verdict: Wait — unblock by (a) completing regulatory audit, (b) identifying a simpler first use case like clinical documentation.
  • Fallback opportunity: Clinical note drafting from visit templates. Requires less data integration. Could pilot in 45 days once COO is free. Estimated savings: $120K/year (reduced transcription costs).
DeltaLogistics (Logistics, $45M revenue)
  • Gate 1 (Data): Route and scheduling data is in a TMS system with an API. Clean. Yes.
  • Gate 2 (Owner): VP Operations is a strong champion, already evaluated two vendors. Yes.
  • Gate 3 (Pilot in 30 days): AI-powered scheduling can run parallel to existing system for testing. Yes.
  • Verdict: Go
  • Top opportunity: AI-optimized routing and scheduling. Current manual dispatch handles 200 routes/day. AI optimization can reduce miles driven by 8-12%, saving fuel and driver overtime. Estimated annual savings: 180K−180K-270K. Tool cost: ~$40K/year.
EpsilonServices (Business Svcs, $35M revenue)
  • Gate 1 (Data): SOPs are in Word documents on a shared drive. Proposals are in a mix of formats. Messy. No.
  • Gate 2 (Owner): CEO is excited about AI but has no specific use case in mind. Vague. No.
  • Gate 3 (Pilot in 30 days): Without a defined use case, no pilot is possible. No.
  • Verdict: Wait — unblock by running a 2-day workshop with the CEO to identify one specific, data-ready use case.

​
Portfolio Ranking

RankCompanyOpportunityEst. EBITDA ($)Months to ValueGateFirst Step
1DeltaLogisticsAI route optimization$180-270K3GoVP Ops to run 30-day parallel test with vendor
2BetaSaaSSales call AI + CRM$100-150K2GoCRO to pilot with 3 reps for 4 weeks
3AlphaWidgetsAP automation$57K2GoCFO to onboard AP tool, digitize PO backlog
4GammaCareClinical note drafting$120K6Wait (audit)Revisit after Q2 audit completion
5EpsilonServicesTBDTBDTBDWait (no use case)Schedule CEO workshop in 2 weeks

​
Replays Identified

  1. AP automation replay: AlphaWidgets (57K)andDeltaLogistics(estimated57K) and DeltaLogistics (estimated 30K — smaller AP volume but same tool). Shared vendor = pricing leverage for both deployments.
  2. Sales call AI replay: BetaSaaS (100−150K)andEpsilonServices(onceusecaseisidentified,100-150K) and EpsilonServices (once use case is identified, 35M revenue company likely has 5+ reps who could benefit). Same tool, second deployment.

​
Aggregate Impact

  • Year 1 quick wins (Go companies): 337K−337K-477K incremental EBITDA
  • Year 2-3 scale (Wait companies + expansion): Additional 200K−200K-400K as GammaCare and EpsilonServices come online
  • Total portfolio AI opportunity: 537K−537K-877K annualized EBITDA

​
Daily Workflow for Operating Partners

Monday Morning (15 min): Review the portfolio AI dashboard. Check which pilots are in-flight, which are stalled, and whether any new blockers have emerged. Update the Go/Wait status for each company. Weekly Check-In (30 min with each pilot owner): Ask three questions: (1) Is the pilot on track for the 30-day milestone? (2) What is the biggest obstacle this week? (3) Have you seen any early data on adoption or savings? Document answers and escalate blockers. Monthly Portfolio Review (60 min): Re-run the ranking. Has any Wait company unblocked? Has any Go company’s pilot failed or succeeded? Update the EBITDA impact estimates with actual data from live pilots. Present the updated one-pager to the partnership. Quarterly Board Materials (30 min prep per company): For each portfolio company board deck, include a single slide on AI status: Go/Wait classification, active pilots with early results, and next quarter’s AI priorities. This keeps the board informed and signals to management that AI is a tracked priority. Replay Execution (as pilots succeed): When a pilot at Company A shows positive results, immediately schedule a call with the AI champion at Company B to discuss replication. Share the playbook: which vendor, which workflow, what worked, what did not. Target 2-week setup for replay deployments since the learning curve is already flattened.

​
Practice Exercise

You are an operating partner at a lower middle market PE firm with four portfolio companies:
CompanySectorRevenueEBITDA MarginHold Period LeftAI Status
FastTrack CouriersLast-mile delivery$30M12%2 yearsNone
ClearView AnalyticsData analytics SaaS$15M20%4 yearsUsing AI for QA
Pinnacle Dental GroupMulti-location dental$50M18%18 monthsExploring
GreenBuild MaterialsBuilding products dist.$90M8%3 yearsNone
Tasks:
  1. Run each company through the three-gate framework. Identify which are Go vs. Wait.
  2. For each Go company, identify the single highest-leverage AI opportunity and estimate the annualized EBITDA impact.
  3. Identify at least one replay opportunity across the portfolio.
  4. Produce a one-page summary ranking all opportunities and listing recommended 30-day first steps.
  5. For the company with 18 months of hold period remaining, determine whether AI is worth pursuing or whether the time-to-value makes it impractical for exit.

​
Common Mistakes

Rank by dollars, not excitement. A boring AP automation that saves 400Kata400K at a 40M revenue company beats a flashy customer-facing chatbot every time.
  1. Starting with the technology instead of the problem. Teams get excited about a new AI tool and go looking for a use case. Start with the cost structure and operations, then ask where AI can help.
  2. Overestimating automation rates. Assume 30-50% automation of a task, not 100%. AI augments humans; it rarely replaces them entirely for complex workflows. If your EBITDA estimate assumes 80% FTE reduction, cut it in half.
  3. Ignoring the ownership gate. A quick win with no internal owner dies in 90 days. If no one on the management team wants it, mark it Wait regardless of the dollar size. The operating partner cannot be the day-to-day owner at the portfolio company level.
  4. Confusing interest with readiness. A CEO who says “we should do AI” is not the same as a CFO who says “I will own our AP automation rollout and have the vendor on-site next Tuesday.” Interest without ownership is just enthusiasm.
  5. Skipping the data readiness check. The binding constraint is almost always data, not models. If a company cannot produce a clean customer list, AI is not the first project — a data cleanup is. Say so plainly.
  6. Proposing custom builds to companies without engineering depth. Off-the-shelf first. Custom builds are slow, expensive, and fragile for companies without a technical team. Favor tools they can buy and deploy within 30 days.
  7. Ignoring hold period urgency. A company 3 years from exit can afford a foundational data project. A company 12 months out needs something that shows up in LTM EBITDA for the CIM — or skip it entirely. Do not waste the last year of hold period on a project that will not be measurable before exit.
  8. Not tracking replays. The highest-leverage move in a portfolio is one implementation deployed multiple times. If you solve AP automation at one company and do not immediately assess which other portfolio companies have the same pain point, you are leaving value on the table.
  9. Relitigating failed pilots without understanding why they failed. If management already tried something and it did not stick, find out why before proposing the same thing again. The failure may have been a vendor issue, a change management issue, or a signal that the company is not ready.
  10. Presenting too many opportunities at once. The best plans have 3-5 major initiatives, not 15. Each portfolio company should focus on one AI pilot at a time. Operating partners who present a menu of 10 options to a management team get none of them done.

​
How to Add to Your Local Context

claude plugin install private-equity@financial-services-plugins
Customize for your portfolio: 
## Portfolio Company List
| Company | Sector | Revenue | Hold Period Remaining | AI Status |
|---------|--------|---------|----------------------|-----------|
| [Company A] | SaaS | $25M | 3 years | No initiatives |
| [Company B] | Industrial | $80M | 18 months | Exploring AP automation |

## Preferred AI Vendors
- Document Processing: [vendor]
- CRM/Sales: [vendor]
- Finance/Accounting: [vendor]

## Quick Win Threshold
- Minimum EBITDA impact: $100K annualized
- Maximum time to pilot: 30 days
- Maximum implementation cost: $50K
Connect to portfolio company data rooms, SharePoint, or Google Drive for automatic document ingestion.

​
Best Practices

  • The binding constraint is almost always data, not models. If a company cannot produce a clean customer list, AI is not the first project — a data cleanup is. Say so plainly.
  • Off-the-shelf first. Custom builds are slow, expensive, and fragile for companies without engineering depth. Favor tools they can buy and deploy.
  • Ownership is the real gate. A quick win with no internal owner dies in 90 days. If no one on the management team wants it, mark it Wait regardless of the dollar size.
  • Hold period drives urgency. A company 3 years from exit can afford a foundational data project. A company 12 months out needs something that shows up in LTM EBITDA for the CIM — or skip it.
  • Failed pilots are signal. If management already tried something and it did not stick, find out why before proposing the same thing again.
  • Track replays relentlessly. After every successful pilot, ask: which other portfolio companies could use this exact same playbook? The marginal cost of the second deployment is a fraction of the first.
  • Measure in dollars, not adoption rates. “85% of reps are using the tool” is interesting. “$140K in annualized cost savings from reduced CRM admin time” is what matters for EBITDA and exit.
  • Run quarterly re-scans. The AI landscape evolves rapidly. A use case that was not feasible 6 months ago may now have an off-the-shelf solution. Re-run the portfolio scan every quarter to catch new opportunities.