How AI is Transforming Private Equity

How AI is Transforming Private Equity

The big picture

• Ai is becoming ubiquitous in enterprise software
• Industries known for being conservative, paper-based, intermediary heavy, have long thought that they would not be disrupted, they could not be
• Deals are being made in cigar smoke-filled rooms, whiskey in hand, no room for software here
• Think legal, tax, alternative asset investments
• Private equity is one of them
• Deal-making is personality-based, portfolio monitoring is excel, email, faxes, and people still send capital calls and notice of distributions by regular mail (I am serious)
• Funds structures are complex, involve multiple law firms, tax specialists, asset valuations are manual
• Software and services are back-office focused, fund administration and fund services focus on reporting, accounting,

Source: Drexel

How did we get there?

In any events, it’s 2021 and sadly things do change

Deal Sourcing

• Heavily competing teams need to source differently, an incredible amount of capital battle for a finite number of opportunities

As a result multiples of deals valuations are skyrocketing (but this is another debate)

Scarcity driving prices to record high levels, think 2007, all over again, induces a need for fresh, non crowded deal flow

• Several companies have blossomed in the AI-based market research field
• AI-based market research tools can deliver results in near real-time, and automatically classify and mine the text for key insights, all in a matter of hours or at most a day.
• Improving the capabilities of analysts by sifting through large amounts of unstructured data, companies like Thinknum and Ventureradar offer algorithmic-based databases.
• The goal here is to not only review more companies but also go deeper, look at web traffic, number of employees on LinkedIn, score on glassdoor, followers on Twitter, reviews on yelp, job openings on indeed, financials mentioned in executive interviews, and cross that with perceived risk in the sector, regulatory scrutiny, technology stack, customer behavior, sentiment analysis of Facebook posts, Twitter mentions, Instagram comments
• Other companies specialize in smart CRM, augmenting your contacts with aggregated data from web scraps think affinity.co
• At the other end of the spectrum, you have established players like CB Insights, Pitchbook and Preqin, offering both data and research
• AI can be used to monitor the key happenings, media mentions, events, signals, catalysts, and sentiments surrounding your portfolio companies’ industries. This allows you to keep track of your portfolio in real-time, create accurate market forecasts,

AI-powered due diligence

• A key characteristic of AI solutions is their ability to learn from a data set and use that machine learning to identify similar data in a new data set
• This is particularly useful during due diligence when looking for patterns of fraud, financial misdeeds, hidden accounts, creative accounting,
• Human reviewers get tired, make mistakes, and cannot review as much data as a machine
• Companies like Ansarada offer tools with AI-based due-diligence features
• Ansarada’s AI tools can analyze the real-time flow of data from interactions between bidding parties in a deal, including huge volumes of information. This enables dealmakers to garner value from tens of thousands of data points in seconds. They can also automate hours of manual work to save significant time and cost inefficiency alone.

Portfolio Monitoring

• As with investment research, AI can be used to monitor key events, media mentions, social, sentiment, risk signals of PE portfolio companies.
• The idea here is to help with market forecast, growth areas, risks of bankruptcies, declining demographics.
• For example, Parabole AI is a solution that (i) allows PE investors to define, in their own words, the types of risk they want to proactively manage; (ii) scans a wide range of sources (e.g., business news, investment research, social media) to develop a score for each risk category defined by the PE investors; and, (iii) enables users to quickly navigate to the specific paragraphs in the sources that are driving risk scores.
• Large PE firms are now looking at portfolio companies reporting consolidation. How can one automate the financial reporting for a whole fund?
• Another example is 2 six capital, and its Intellio™ Predict, which uses large-scale engineering to quickly ingest billions of data points and accurately project business drivers that drive investment decisions and value creation.

Portfolio company reporting.

• AI has also emerged as a potential way to more efficiently process and consolidate portfolio company reporting, which is often not standardized. For example, the portfolio company value creation team of a large Canadian institutional investor used AI to automate 92%+ of the process to create a consolidated financial view across the portfolio. The team was also able to leverage AI to quickly identify the key metrics or business areas to focus on in each portfolio company. Among other benefits, early adopters spend up to 30% more time thinking about specific issue areas versus identifying them.

Why it matters

• Growing popularization of AI-powered investment research solutions in the PE space.
• Digital transformation of investment firms
• Investment in platforms consolidations technology
• Quality reporting, cost management, regulatory pressure all drive adoption of sophisticated tools to better monitor portfolio companies and create value

Yes but…

• However, the risks of using AI technology underpinned by huge data sets cannot be underplayed, with maximum fines under the EU General Data Protection Regulation (GDPR) now being the higher of €20 million, or 4% of an entity’s global turnover. The cost of potential liabilities for breach/misuse can exceed the purchase price on a deal, and if issues are identified, the value of the acquisition will be reduced, as GDPR fixes are expensive to action. Deal teams must carefully scrutinize whether AI businesses are compliant, or risk assuming significant liabilities.

The catch

• It is challenging to gather large amounts of PE data to train algorithms, as the number of transactions is limited and information is often confidential
• PE funds have a 10 to 12 years horizon, so performance measurement comes slowly and is impacted by companies that never sell, or at a large discount, resale in the secondary market.
• In the absence of a public market benchmark, what matters are unstructured sets of disparate data, driven by each player appetite
• PE Firms, often small and focused, have to look at the complexity, costs, contracts, and talent needs for these projects

By the numbers

• Affinity raises $26.5 million to manage relationships with machine learning

What people say

• AI projects are hard, they often fail
• They demand talent, leadership, clear roadmaps, understanding of what can be done
• It’s hard to define “What success looks like”

• Source HBR

What I think

• Change is inevitable
• Private Equity will not avoid it
• It’s an exciting space because there is so much to do
• Think automation of the fund management, secured payments, real-time monitoring, powerful dashboards that build a consolidated view and suggest mergers and alliances
• Predictive analytics to prepare exit investments and highlight areas of investment to accelerate growth and build value

Dig deeper

What to listen

• Will and Neil join Alex Agran on The Private Equity Technology Podcast to discuss Artificial Intelligence’s impact on private equity investing.

What to read

• Private equity steps up its technology adoption
• The tech behind it: Machine Learning and GPT-3: Applications for Private Equity (must read)
• Do Algorithms Make Better — and Fairer — Investments Than Angel Investors?

This post is part of Convergences by Melvine. A series exploring how software is changing every corner of human activities. Melvine Manchau