Unleashing the power of generative AI for private equity

In brief

• The rise of generative AI brings huge upside potential to private equity (PE) firms that have invested in their data environment and infrastructure.

• The technology also lends opportunities to portfolio companies (PortCos). Our research estimates a 6–7% productivity gain and a 3–4% revenue uplift.
  

• The challenge is figuring out where to start. We recommend a four-step approach to understanding and successfully implementing generative AI.

When one of Hemingway’s characters in the book The Sun Also Rises is asked how he went bankrupt, he answers “gradually, then suddenly.” Technological change happens much the same way. Small changes accumulate and suddenly the world is a different place!

Some PE firms have been gradually investing in their data and analytics capabilities for the past two decades. They have set themselves up to capture the unprecedented opportunities suddenly presented by the emergence of generative AI. Others will need to catch up quickly, as generative AI has the potential to revolutionize the investment lifecycle for PE, from fund-raising to exit planning, and from driving efficiencies to creating new business models for PortCos.

To wield its full potential, PE firms and PortCos alike have started identifying use cases and building capabilities that contribute to a Total Enterprise Reinvention mindset: a strong digital core, nurturing talent and driving effective change management.

The AI edge for PE firms

Whether accelerating target screening and due diligence through assisted research or enabling general partners to manage limited partner relationships through customized pitch decks, across the investment lifecycle, generative AI is lending PE firms an edge.

Take deal sourcing. When scoping private targets, a combination of robust data, machine learning and generative AI can help PE firms screen thousands of potential targets and shortlist them. This can reduce time spent by an eye-popping 50–60%.

BC Partners is an early adopter. They’re using generative AI for deal sourcing and due diligence, building APIs that will let pre-trained AI large language models work on unstructured proprietary data in ways that a human simply can’t.

Figure 1: Generative AI throughout the PE investment lifecycle

The value potential for PortCos

ChatGPT’s explosive adoption by the public marks a true inflection point for generative AI. Everyone everywhere can experience its innovative potential. While this technology is set to transform every industry and function, not all will be impacted at the same rate or level. Financial Services, Technology, Retail and Life Sciences are particularly ripe for disruption. Generative AI presents both defensive and offensive opportunities, with our research estimating a 6–7% productivity gain and a 3–4% revenue uplift.

Figure 2: Generative AI will transform work across industries (work time distribution by potential AI impact)

Within the PortCos themselves, generative AI has the potential to significantly affect business functions like customer service, knowledge management, sales and marketing, and product management. 40% of working hours can be impacted by large language models that underpin generative AI.

Figure 3: Generative AI will transform work across every job category (work time distribution by potential AI impact)

With so many potential use cases, the challenge is figuring out where to start. We recently worked with a $100B+ AUM PE firm looking across their PortCos to determine their use case prioritization and suitability to generative AI implementation. By assessing urgency, impact and readiness, we identified a combined EBITDA uplift of approximately $460M across four portfolio companies. These results cascaded into a high-level strategy and an implementation plan. It also informed their leadership about the future potential across the broader portfolio.

Entering the race

There are several ways that PE can start harnessing generative AI for the firm or its PortCos:

• At the absolute minimum, go for table-stakes use cases, for example, acquiring Microsoft 365 Copilot for MS Office apps to improve middle/back-office productivity.

• Next, place no-regret bets to drive efficiency, programming and personalization. For example, by adopting GitHub Copilot or AWS Code Whisperer, implementing intelligent knowledge assistants or automating financial reporting.

• Increasingly, PE firms focus on building capabilities to enable differentiated use cases. For PortCos, this can mean highly specialized industry use cases like generative design in Manufacturing, field worker Copilot in Resources, intelligent relationship management in Banking, AI-assisted drug discovery in Life Sciences, etc.

We recommend a four-step approach for private equity to understand and successfully implement generative AI:

1. Educate: Plan and deploy a data and AI learning pathway to bring firm and PortCo leadership up to speed. Generative AI is not the answer to everything! Understanding it and related topics like automation, data strategy and machine/deep learning will help inform better investment decisions.
   
   
2. Explore: Define your overall vision and use cases that align with business needs. Prioritize areas where generative AI can make the biggest impact and drive incremental revenue, cost optimization or risk reduction. Also assess where solutions go beyond generative AI and what broader capabilities are needed. Over half of the 2,300 leaders surveyed by Accenture see a lack of data readiness as the top obstacle in adopting AI.
   
   
3. Experiment: Identify the right ecosystem partner — hyper-scalers like Microsoft, Google, AWS, or specialized players like SambaNova, Nvidia, Anthropic — and conduct rapid proof-of-concepts to validate the planned value delivery of use cases. Take a “fail-fast” approach to test the feasibility and impact of the technology in a controlled environment.
   
   
4. Execute: Develop a plan to scale the chosen use cases across the enterprise. This means integrating generative AI into existing workflows, managing any change or disruptions to business as usual and training talent to drive desired outcomes. Also determine whether to build models in-house, buy ready-to-use products or boost internal efforts through ecosystem partnerships.
   

Gaining upside

While early adoption of generative AI can have huge upside, its successful implementation demands PE firms and PortCos take a strategic approach. Here are some considerations as you start your journey:

• Strengthen the digital core. Models underpinning generative AI need large amounts of curated data to learn. It is an investment area many PE firms have yet to make. Take a strategic and disciplined approach to acquiring, refining, safeguarding and deploying data. Consider requirements for infrastructure, operating models and governance structures to wield generative AI while keeping a close eye on cost and sustainability.

• Dive in with a business-driven mindset. On one level, focus on “table stakes” and “no-regret” opportunities to realize quick returns using consumable models and applications. On another level, focus on reinventing businesses using models that are customized with internal data. Wrap both levels with strong governance structure and leadership buy-in. 
  

• Take a people-centric approach. Concentrate on people as much as technology, ramping up talent investment to create and use AI. Develop AI-related technical competencies and train people across the organization to work effectively with AI-infused processes.
  

• Accelerate ecosystem innovation. Take advantage of industry experience and insights offered by ecosystem partners. Include technology companies, professional services firms and academia.

• Level up responsible AI. Make sure the company’s responsible AI governance is sufficiently robust before scaling applications. Build in controls to assess risk at the design stage. Embed responsible AI throughout the business.

Like Marco Argenti, CIO of Goldman Sachs, noted recently in the Wall Street Journal: “We are way more than just thinking about it [Generative AI], we are really trying to prioritize certain use cases and then starting to invest in those.” To industry players who move beyond thinking to action, the competitive advantage could prove to be unassailable.

The author would like to thank Sambit Banerjee and Amir Waqas for their contributions to this article.