THIRD ORDER
AI engines gather steam across PE portfolio companies
KEY TAKEAWAYS
Firms are now exploring second and third order outcomes from AI
ROI can now be measured in terms of cost-revenue ratios per minute
It’s no secret that the private equity landscape and the ability to drive returns has changed for GPs. Strategies centred on quick cost wins and short hold periods, fuelled by low interest rates, were long in favor. Now, with elevated rates, impatient LPs, and longer hold times, this formula has broken down. Leading funds realize the imperative to change – to truly deliver meaningful portfolio company improvements and sustainable value capture.
In this search for value levers, private equity firms are finding AI’s transformational power to be of landmark importance – more effective than new go-to-market strategies, talent and even inorganic growth (see Fig.1). Cost savings are by far the biggest benefit, cited by more than three-quarters of our 100+ survey respondents (see Fig.2.). Other high-impact areas include service excellence (45%), revenue optimisation and product innovation (39%).
Amit Shah, Founder and CEO of Instalily AI, a platform that provides industry-trained AI teammates called InstaWorkers™, explains these outcomes as three orders of value creation. He does so in the context of the average sales team at a PE portfolio company. The journey starts with systems of record like Salesforce or SAP, advances through systems of learning, and ultimately culminates in systems of action, where AI teammates learn and act directly in the same company systems as human employees.
First-order: From systems of record to automation
At this stage, AI reduces manual effort for the sales team by automating repetitive work locked inside company systems. It’s about compressing time-tooutput: generating quotes, logging calls, entering orders, updating pipelines, and handling routine customer follow-ups. “These first-order benefits are obvious,” Shah says. “Most companies stop here, but because these improvements are widely replicable, the value creation ends once cost savings are realized. There is no lasting moat.”
Second-order: From automation to systems of learning
Here the sales use case becomes clearer. The best salespeople already know which signals to prioritize — cross-sell, upsell, or churn risks — and how effective those signals are. Shah explains: “If this judgment is fed back into the AI model as simple up or down votes, the system learns through reinforcement and human feedback. With every sale, the model gets
Figure 1 Most important value creation levers for PE firms
smarter. That is when a true moat emerges, because the system learns how to sell better, not just cheaper, with higher close rates and faster revenue realization.”
Third-order: From learning to systems of action
At this stage, companies equip their staff with AI teammates that can work on their own or under human guidance. These teammates both learn and act inside the same company systems used by employees. Routine signals are resolved automatically, while people focus on closing deals, solving issues, and building relationships. “Systems of action make every process more potent,” Shah explains. “Each action feeds back into the system so it learns and acts more effectively. Revenue per employee steadily rises, and the organization embodies a dynamic value creation engine rather than static value creation plans.”
THE NEXT PHASE
For years, there has been a disparity between the potential of technology and its implementation. Common blockers include management buy-in for digital transformation, employee adoption and change management, and the overall measurement of returns on investment (ROI). Our survey reflects the same challenges persist in PE portfolios (see Fig.4).
Still, according to Shah, the fundamentals have shifted. “Most PE portfolio companies are built through M&A, which leaves them with fragmented systems and uneven data. That used to be a blocker, but AI teammates have changed the game.”
The conversation has moved on. “The PE operators we work with are no longer stuck in proof-of-concept mode,” Shah adds. “They want AI at scale, capable of delivering second- and third-order outcomes.”
ROI has shifted from a question of “if” to “how much.” Shah says: “Companies in our universe are no longer counting marginal savings – they are unlocking hundreds of millions in additional sales by making every interaction profitable and more targeted. And every executed action becomes a training event, compounding results.” As he puts it, this is “ROI on steroids” – measured not in tasks saved, but in profit generated per minute of effort.
“The smarter PE operators we are working with are leveraging AI to build value creation engines, not just value creation plans. They are working to redefine how businesses create and capture continuous value.
That is the power of third-order value creation.”