Picture two private equity (PE)–owned companies, each trying to capture benefits from AI. Company A uses AI tools to automate administrative and accounting tasks such as generating invoices and scheduling meetings. Company B also uses AI to boost productivity—but doesn’t stop there. It develops an agent to handle client onboarding and renewals, as well as an AI-powered sales management tool as a new product offering for its existing customers. Over time, which company is likely to generate a higher valuation?
You guessed it: Companies that embed AI across operations, products, and new business ventures demonstrate higher growth potential. The difference is material. In our analysis of PE-backed companies, those that broadly embrace AI trade at a median revenue multiple1 approximately 130 percent higher than that of companies that primarily use the technology in an opportunistic way (see sidebar, “Research methodology”).
With the PE industry facing a considerable exit backlog and registering modest returns, revenue growth and operational efficiency are increasingly central to generating portfolio returns. Against this backdrop, AI is emerging as a powerful but still underutilized catalyst for value creation. While adoption has accelerated across PE-owned companies, many stakeholders are still waiting for a financial impact. Productivity use cases—the most common application of AI—have not translated into durable revenue and exit opportunities across their portfolios.
Our analysis of 471 PE-backed companies across 31 industries (including software, consumer nondurables, financial services, insurance, and infrastructure), combined with research on AI use in the PE industry, indicates four AI capability levels across portfolio companies: opportunistic adoption (level one), operating-model enhancement (level two), embedding AI in products and services (level three), and business building (level four). Each successive level reflects deeper technological integration, higher AI maturity, and stronger value creation outcomes. In particular, we focused on the impact of AI adoption on revenue efficiency and revenue multiples at these different levels. We found that companies at level four traded at a median revenue multiple of 31x between 2023 and 2025—the highest across all four levels (Exhibit 1). Companies at this level also observed a $180,000 increase in median revenue per employee—a 52 percent jump from level three.
This article explores how PE companies can capture revenue growth from AI. Although our analysis focuses on a subset of PE firms, portfolio companies, regardless of sector or geographic focus, can use the framework to generate meaningful impact from their AI initiatives.
AI as a structural driver of revenue growth
Investor demand for AI-enabled business models has led AI start-ups to receive half of all global venture capital funding in 2025.2 The opportunity for PE firms, however, extends beyond simply investing in AI-native companies. When implemented and scaled effectively, the technology can strengthen portfolio companies’ competitive positioning and value. An analysis by QuantumBlack, AI by McKinsey, reveals meaningful differences in the total shareholder return CAGR between PE-backed companies that lead in digital and AI initiatives versus those that lag behind. The analysis shows that between 2018 and 2022, digital and AI leaders delivered TSR CAGR that was 6.1 times higher in insurance, 2.9 times higher in the consumer sectors, and 2.3 times higher in energy and materials than their less digitally advanced peers. These findings suggest that the value creation opportunity for PE firms lies not only in backing AI-native businesses, but also in accelerating AI adoption and scaling across portfolio companies.
To systematically harness this potential and replicate it across their portfolios, PE firms can view their portfolio companies through the four-level AI value creation ladder. In our experience, businesses advance through the levels gradually, unless they leapfrog from level one or two directly to level four to pursue new business building.
Level one: Opportunistic adoption
At this stage, AI use is either absent or limited to isolated experiments. Companies may be running ad hoc pilots or using individual productivity tools, but AI is not meaningfully embedded in workflows, decision-making, or the value proposition. Usage is opportunistic rather than strategic, and there is little evidence of true economic impact from developing new products, changing business models, or reconfiguring value chains.
Many portfolio companies are earlier in the journey than their PE owners assume. Before pursuing more ambitious AI moves, PE firms need to establish a clear baseline on where the portfolio companies are using AI and whether those use cases are generating sustainable value.
Level two: Operating-model enhancement
At level two, companies are embedding AI into their operating models to augment workflows, increase operating speed, enhance decision-making, and expand output without proportional head count growth. For example, they may use AI to streamline the sales process, automate customer service tasks, generate marketing content, and automate client onboarding through agents.
At this level, AI begins to influence not only productivity and margins but also commercial performance. Although the company’s business model may remain unchanged, its operating model becomes more scalable, responsive, and less constrained by labor challenges. Sales and operations teams can see improvements in key commercial metrics such as average revenue per user, customer lifetime value, customer retention, and repayment rates.
A PE-owned global industrial-materials supplier, for example, used AI to maximize the growth of its e-commerce platform. The company introduced an AI-driven sales agent to reengage inactive customer accounts by analyzing their past purchasing behavior and delivering personalized outreach. During the pilot, the approach achieved a 30 percent engagement rate among previously inactive customers. Another PE-backed edtech company, meanwhile, used AI to enhance lead generation. The pilot doubled the average order value across an initial set of prospects.
Moving toward AI-augmented workflows can improve a company’s bottom line and overcome talent constraints that hold back slower adopters. Companies at this level can start to see material gains as they generate approximately 20 percent higher revenue per employee (after accounting for inflation) than those at level one (Exhibit 2). They also trade at a revenue multiple of 14x, compared with 13x for level-one companies. To capture these opportunities, portfolio companies need to establish guardrails defining where AI can operate autonomously and maintain sufficient human oversight and compliance controls to manage exceptions.
Level three: Product transformation
Companies operating at level three are embedding AI within their products or services. They can grow revenue by optimizing pricing, personalizing customer experiences, improving win rates, and reducing churn.
Retail companies, for example, can use AI-powered features to personalize customer experiences and re-create elements of the in-store experience online, such as virtual mirrors for trying on makeup or clothing. Another software-as-a-service-based platform uses AI agents that help finance teams reduce manual work, identify risks and cash flow variances, and recommend strategic actions. The tool provides clients with capabilities typically associated with finance teams at large enterprises without the need for additional employees. The company’s PE owner realized a sizeable return on its investment when it sold the tool in less than three years.
The transition to level three appears to have meaningful implications for valuation. Our analysis shows that companies at this level trade at a median revenue multiple of 20x—43 percent higher than the median revenue multiple of 14x for level-two companies. By contrast, the difference between the median revenue multiple for level one and level two companies is negligible (Exhibit 3). This suggests that markets do not materially differentiate between companies that use AI for productivity and those that integrate it into their operating models. Valuations rise significantly only when AI is embedded in the offerings, changing what the company sells—not just how it operates.
To achieve such gains, the product, data, and technology teams within portfolio companies must work in closer collaboration to develop differentiated products for users—more than at the prior levels. The revenue gains will also likely take longer to materialize compared with the prior two levels.
Level four: Business building
At this level, portfolio companies move beyond applying AI to improve existing economics. Instead, they use it to develop new business lines or businesses that can create new revenue streams and data monetization opportunities.
Portfolio companies that face a structural industry disruption or have an opportunity to lead market innovation can particularly benefit from acquiring level-four capabilities. They could, for example, build AI-powered businesses in adjacent markets (such as a logistics company launching an AI-based route optimization platform), develop data products based on operational telemetry (such as an industrial company selling subscription-based equipment performance data and predictive maintenance insights), or create AI-native service lines that can be layered onto traditional offerings (for example, an accounting firm offering AI-driven financial-forecasting and scenario-planning tools alongside its core services).
One technology company expanded beyond its core business by launching AI-enabled customer solutions and new digital services that generated advertising, referral, and subscription revenue. Combined with the automation of its legacy operations, the shift helped the company reduce head count by nearly 40 percent, while nearly doubling the revenue per employee within two years.
Three findings stand out when examining companies that reach level four. First, building AI-driven businesses correlates with higher revenue multiples; companies that advance from level three to four add an 11-point increase in median revenue multiples (from 20x to 31x).
Second, both software and nonsoftware companies in our sample set of industries generate higher revenue multiples at each successive level above level two—particularly as they move from level three to four. Software companies, however, trade at revenue multiples of 33x at level four versus 24x at level three, while nonsoftware companies trade at 22x and 15x, respectively.
Third, revenue streams anchored in AI are fundamentally more scalable than those generated by embedding AI to improve productivity or enhance workflows (as companies do at the earlier levels). Revenue efficiency also rises sharply at level four: The median revenue per employee increases to $180,000 from $118,000 at level three—a 52 percent jump that significantly exceeds the 19 percent increase observed between levels one and two.
However, capturing this value often requires capabilities beyond those most PE firms possess. PE firms are structured to improve existing businesses, not build entirely new ones. For portfolio companies to succeed at level four, they may need to treat AI initiatives as stand-alone businesses—with dedicated teams, separate go-to-market strategies, pricing models, operating structures, and profit-and-loss accountability.
How PE firms can get started
The role of a PE firm is expanding from asset selection to building repeatable, AI-enabled initiatives to create value across its fund’s investment portfolio. By using AI in a consistent way, firms can improve their portfolio companies faster, achieve more predictable results, and build an edge they can reuse in future investments.
Talent scarcity is a key factor hindering PE firms from climbing the AI value creation ladder, despite the availability of low-code/no-code platforms. At the same time, we find many portfolio companies reluctant to build their own AI teams, particularly at levels one and two, where needs are variable.
This creates a “rent versus buy” decision for PE firms: centralize platforms, redeploy assets and experts at the fund level (rent), or embed dedicated teams in portfolio companies, where proprietary data or product transformation can justify sustained investment (buy). Many leading firms adopt a hybrid model, creating centralized infrastructure and expertise for common use cases and using dedicated in-house teams for product innovation.
Once the right team is in place, PE firms can consider three moves to integrate AI across their portfolios:
- Leverage levels one and two across the portfolio: PE firms can assess their portfolios for areas where AI can enhance productivity and improve operating models. These foundational levels are applicable to most companies. Once they have identified target companies, PE firms can focus on finding quick wins. For example, they could automate or optimize processes with AI, leading to cost savings and efficiency gains without altering the core business model. After successfully executing these levels in a select group of portfolio companies, PE firms can develop a repeatable playbook for their entire portfolios. Many of the initial hurdles relating to AI adoption at the lower levels may be overcome as firms build experience and capabilities.
-
Explore AI’s transformative potential with levels three and four: These levels can capture meaningful revenue upside, provided that the portfolio company has the right structure and capabilities.
Companies that are ready for level three tend to exhibit two characteristics. First, they have a proprietary data moat and have accumulated unique data on their products or services that competitors cannot easily access or replicate. Business leaders can use this data to develop predictive or personalized features. Second, they have built an integrated operating model in which product development, data science, and engineering teams collaborate closely.
Companies at level four distinguish themselves in a different way. They are willing to fundamentally reshape their business models to stay ahead of disruption and create new revenue streams. They are also adept at becoming the disrupters themselves by positioning their business at the forefront of market innovation. Once PE firms are ready to move to this level, they can focus on using AI to solve customer needs in ways that weren’t feasible before, accelerate product development, and create more personalized customer experiences.
- Prioritize and rank: Assessing the potential of AI to create value in portfolio companies can help PE-backed portfolio companies prioritize their AI investments. For example, they can select five to ten companies with the highest potential and rank them based on their readiness to progress through the four levels. One approach could be to prioritize companies in sectors facing AI-driven structural disruption (such as legal services); another could be to focus on the software companies in their portfolio, where the increase in revenue multiple from level three to four is the most pronounced.
Markets are already pricing in the difference between portfolio companies that use AI for efficiency and those that use it to build new businesses. For PE firms, the implication is clear: The greatest opportunities to create value are likely to come not from isolated productivity gains but from embedding AI into products, operating models, and entirely new businesses. The window to capture these gains is narrowing as more and more companies develop AI capabilities. Firms that move early to scale these capabilities across their highest-potential assets will likely be best positioned to capture significant value at exit.