How Private Equity Links AI Returns To Connected Data And Operational Change

Private equity doesn't want AI experiments; it wants measurable business outcomes. Higher capital costs and shorter investment horizons have left sponsors with little patience for multi-year technology bets. The AI initiatives that survive are those directly tied to revenue growth, EBITDA improvement, cost optimization, or valuation creation. Some of what gets labeled enterprise AI is a dashboard sitting on top of disconnected data, able to report what already happened but little else. Connect that enterprise data underneath, and the same investment starts predicting what comes next, which is where the payoff separates from the theater.

Vishal Vallabha is a Global Chief Technology & AI Officer and Value Creation Leader who advises private equity firms and their portfolio companies on AI, digital transformation, enterprise technology, and value creation. His career spans enterprise architecture, data and AI leadership across capital-intensive industries. He frames every architecture decision as a financial one and reduces what makes an AI investment worth the money to a short equation.

"The winning formula is simple. Connected data + AI-driven intelligence + operational execution = measurable impact," Vishal said. For a sponsor, that impact is the whole case for the spend.

• Moving the numbers: Once operational data is linked across systems, teams can act on it, and only then do the returns surface.The same pressure now extends to public companies, where investors increasingly expect AI initiatives to deliver measurable improvements in revenue growth, productivity, operating margins, operating leverage and not merely incremental efficiencies. For private-equity sponsors and public boards alike, the gains show up in the same operational places: pricing, procurement, working capital, and field productivity. Those are the levers where linked data and AI can add revenue or take out cost fast enough to move a valuation. "It all boils down to value creation," Vishal said.

• Sensors and silos: The problem is rarely a shortage of data, since companies already hold project, asset, workforce, and cost data, and few have defined how those sets relate. Until they do, the predictive models and digital twins sponsors are paying for have nothing to work with. Wiring those sets together is an organizational job, and it is where most programs stall. Operational technology (OT) data including vehicles, factories, warehouses, industrial equipment, and IoT sensors often remain disconnected from enterprise systems. “Until those worlds are connected, AI lacks the context needed to generate meaningful predictions and autonomous decisions,” Vishal said.

Part of why companies skip the work of connecting their data is timing. Boards and executives want quick proof that AI spending is paying off, and a reporting dashboard is the easiest thing to put in front of them. When the deadline is short, that is usually what teams build, presenting early activity while the data behind it is still too scattered to support much.

• Past the dashboard: Whether a system deserves the AI label comes down to the questions a company puts to it. Much of what carries the label just returns the top five accounts by revenue. "For me, that's descriptive analytics and not artificial intelligence. A dashboard could have given you the same result," Vishal said. The value shows up when someone can ask a harder question, why revenue is slipping with a particular customer, and get an answer that pulls together how sales and cost data relate. That only works when the systems are defined to work together, so one dataset knows how it maps to the next. The next evolution is Agentic AI where systems are capable of reasoning, planning, and autonomously executing tasks, but those capabilities only become reliable when they're built on trusted, connected enterprise data.

• Build on what runs: Reaching that point rarely means building a new data setup, because most of the data a company needs already sits in the systems it runs on every day. Most organizations already have trusted systems of record in SAP, Oracle, Salesforce, Microsoft Dynamics, ServiceNow, Snowflake, Databricks and others. Building AI on top of those enterprise platforms is where sustainable value begins. In MIT's widely-cited 2025 study, 95% of enterprise generative-AI pilots delivered no measurable value, a shortfall the researchers traced to how companies integrated and adopted the tools, with model quality largely beside the point. That matched his own read, that the failures came from choosing the wrong use cases and running them poorly. "Building artificial intelligence on top of your trusted system of record is the most critical aspect. That's where the value starts coming in," Vishal said.

• Common core, custom fit: Building on systems a company already runs gives a sponsor with a full portfolio a second advantage. Once the approach works at one company, a version of it can move to the next. Every company needs the same groundwork, its data readied for AI and disciplined control over how the tools get built and used. How far along each one already is varies, so a large, established firm and a smaller, looser one need those same pieces on different timelines. The principles remain consistent, but every transformation roadmap should be tailored to the organization's business model, operating maturity, and strategic objectives. Vishal said.

The role of today's CIO, CTO, and Chief AI Officer is no longer to deploy AI—it is to create measurable enterprise value. Every AI investment should be traceable to growth, productivity, resilience, customer experience, or EBITDA improvement. When organizations start there, AI shifts from an innovation initiative to a competitive advantage. Framed that way, AI gets pointed at the specific operations where it can change an outcome. The mindset is simple: ask how AI can improve each business function rather than implementing it simply to satisfy a board mandate. “When organizations start there, AI shifts from an innovation initiative to a sustainable competitive advantage,” Vishal said.