Turning AI Into a Repeatable Value-Creation Lever Across a Portfolio

The Challenge

Our client, a mid-market private equity firm, held a portfolio of companies spanning industrial services, distribution, business services, and consumer sectors. The spread in AI maturity across those holdings was as wide as the spread in industries. One portfolio company had a capable data team and a handful of models in production; several others were still running core operations on spreadsheets. Each management team was pursuing AI on its own terms, with its own vendors, and with no way for the firm to compare progress, share what worked, or direct capital toward the initiatives most likely to move enterprise value.

The results reflected that fragmentation. Across the portfolio, one-off pilots proliferated (a chatbot here, a forecasting proof-of-concept there), but almost none of them showed up in EBITDA. Operating partners lacked a consistent lens for evaluating AI opportunities during ownership, and the deal team had no structured way to assess AI readiness or exposure in targets during diligence. The firm didn’t need eight bespoke transformations; it needed one repeatable playbook it could run at every holding and every future acquisition.

The Approach

We started with a portfolio-wide AI maturity assessment, evaluating each holding against a common baseline: data foundations, technology estate, talent, and, most importantly, the specific operational levers where AI could plausibly move margin within the hold period. That assessment gave the firm something it had never had: a comparable, portfolio-level view of AI opportunity, ranked by expected EBITDA impact and implementation risk. From it, we codified a reusable transformation framework: a sequenced playbook covering opportunity selection, data readiness, vendor and build decisions, deployment, and measurement, designed to be run by portfolio company leadership with operating-partner oversight rather than by consultants indefinitely.

To keep economics sensible at mid-market scale, we designed shared AI platform infrastructure that all eight companies could draw on, built around common data tooling, model deployment patterns, and governance, rather than eight parallel technology stacks. We ran structured enablement for the firm’s operating partners so they could sponsor and pressure-test AI initiatives with the same rigor they apply to pricing or procurement. And we extended the same discipline upstream into the deal process, building an AI-driven due diligence workstream that assesses a target’s data assets, automation potential, and AI risk as a standard part of underwriting. Every initiative launched under the playbook was tied to the value creation plan, with EBITDA attribution tracked quarterly at the firm level.

Adoption was not uniform, and pretending otherwise would flatter the story. Two management teams pushed back hard in the first quarter, reading the playbook as the firm's process being imposed on their operations; both came around only after seeing quarterly numbers from peer companies they benchmarked themselves against. And one early initiative, a pricing model at a distribution portco, failed against its baseline and was shut down at the first quarterly review. Killing it publicly turned out to matter. It told every other management team that the EBITDA attribution was real, not a scoreboard rigged to justify the program.

What We Delivered

• Developed reusable AI transformation framework for portfolio companies
• AI-driven due diligence process for new acquisitions
• Shared AI platform infrastructure across 8 portfolio companies
• Value creation tracking tied to AI initiatives

The Results

Across the eight portfolio companies where the playbook was deployed, AI initiatives tied to the value creation plan delivered an average EBITDA improvement of 25%, driven not by exotic technology but by disciplined application of AI to pricing, demand forecasting, back-office automation, and service operations, measured against agreed baselines. Because impact was tracked initiative by initiative, the firm can point to exactly where the improvement came from, which matters both for governance during the hold and for the equity story at exit.

The upstream gains compounded the operational ones. The AI-driven diligence process shortened the firm’s due diligence cycle by 40%, letting deal teams evaluate more targets with the same resources and walk into negotiations with a clearer view of each target’s AI upside and risk. Most durably, the firm now owns the playbook itself: every new acquisition enters ownership with a proven first-hundred-days AI agenda, access to shared platform infrastructure, and operating partners equipped to hold it accountable.