Within private equity fund portfolios and across the economy lie treasure troves of valuable financial information. GPs are feeding it into algorithms and prediction models to bolster their strategy…
Within private equity fund portfolios and across the economy lie treasure troves of valuable financial information. GPs are feeding it into algorithms and prediction models to bolster their strategy…
The founders of the world’s largest private equity funds relied on personality and connections to build and execute their first leveraged buyouts. The relentless march of technology is forcing their now gargantuan organisations to take a different approach: they are turning inwards to unearth value from the decades of data in their porfolios.
In a survey by Private Equity Wire, over 50 industry stakeholders were asked ‘In what one area could GPs increase or improve their use of data, technology or AI?’ The most popular response was “value creation within portfolio investments”. The third most prevalent choice (after the more obvious “back office and accounting”) was in “sourcing investments”.
Leveraging recent developments in data science to add or find value from their portfolio is the new secret sauce for GPs, say sources at, or close to, some of the oldest and largest funds. And the strategy is being increasingly used alongside in-person networking and old-school origination to find an edge.
“Data is creating competitive advantage,” says a US-based source working with one of the world’s largest private equity funds.
AI advantage
“There’s more competitive pressure in terms of sourcing deals and driving value creation for investors, so you have to be able to identify those opportunities faster than others.”
More traditional data collection is being supplemented with AI-powered tools, algorithms and prediction models to provide estimates for financial health and revenue growth at target and already-owned firms.
The trend also looks set to drive change within the organisational structure of many private equity firms as new data-orientated teams are built and hires are made (see Chapter 4 in this Report).
“We’re seeing a lot more private equity firms invest in their own data science capabilities, internally, and trying to partner with alternative data providers, to supplement their internal data science efforts,” says a technology consultant advising private equity firms on the subject.
“I think most of the firms that we’ve seen will have a data team in place, or a lead data scientist and not much of a team beyond that.”
How private equity funds use this new capability is dependent on the software they are using or bringing in from third-party providers but also the amount of data they can leverage.
“The pure existence of data, whether it be through providers like Pitchbook, or Cap IQ, or otherwise, means there’s more and more data out there. And there are more tools out there for firms to collect data internally on their own operations,” says a source at an AI-based service provider to the industry.
The volume and granularity of the data available on private assets today would have been unimaginable a few decades ago. The number of alternative-data providers is more than 20 times larger now than it was 30 years ago – with more than 400 currently active providers compared to just 20 in 1990, according to a report by the Alternative Investment Management Association.
Discipline and third-party advice is required to combine statistics and company information with cloud computing and machine learning to analyze large pools of structured and unstructured information, say sources. The insights and predictions unlocked can in turn drive new forms of decision-making as well as faster and potentially more accurate due diligence and valuations.
Private equity firms are leveraging real-time data about consumers from car parking utilization rates as a measurement of foot traffic, seeking out the frequency of keywords in search terms, social media activity, phone usage, demographics, and census data and satellite feeds to improve sales and marketing at their portfolio companies, say sources.
As identified by the Private Equity Wire survey, GPs are applying new data management technologies in two ways. Firstly, in the front office, deal origination and the sourcing of new investments is being disrupted.
“We are seeing more firms try to do proprietary sourcing, where they might be evaluating businesses that aren’t even on the market, but that fit their portfolio strategy,” says the technology consultant. “This includes using publicly available information and data to better drive and find companies that they’re interested in and getting a perspective on that before they might even hit the market.”
More data, please
Research as a Service (RaaS) has evolved from Software as a Service (SaaS) with AI, machine learning and natural language processing (NLP) setting the scene for personalized recommendation algorithms so GPs can identify investment opportunities before others.
“As GPs in the private markets bring in more third-party data, it gets complicated to manage and extract value from information coming from so many sources,” says Cesar Estrada, head of the private markets segment at Arcesium. “We help origination teams harmonize those disparate sources and get to a single source of truth.”
Such an approach works well for fund managers with a specialist sector strategy, for example in healthcare. They may seek insight from their database on the number of healthcare exits in their portfolio that have achieved a certain multiple and use that information to price their next acquisition in a new market. A more generalist private equity fund manager might leverage internal data to better predict whether they will be overweight or underweight a specific sector in their wider portfolio base before making an acquisition or launching a new fund strategy.
Software can also be used to track failed acquisitions by region or sector and interrogate the reasons why a bid was unsuccessful. Were the reasons market-based or internal, and can they be avoided in the future?
“What percentage of the deals you originate do you actually close on,” asks the AI-based service provider. “It’s a lot easier now to collect information on origination and combine that data with internal systems. If you do a robust job of tracking the activities of an origination team, or even a fundraising team, you can actually be smarter around having operational efficiencies [in other parts of the business].”
The second application of data science here is being seen through increased value creation on assets already owned or acquired.
“With a data layer in place, our clients can better organize their information for deal origination purposes and enrich portfolio monitoring to draw new insights,” says Estrada.
According to sources, new value creation teams are being built within GPs, combining experienced origination professionals with often younger data scientists. The approach is not new, but the process is evolving quickly with new software solutions.
Team building
One example is Blackstone which has been compiling historically unconnected data streams to identify value creation initiatives across the business for some time already. Another is KKR Capstone, which has one of the most well-known ‘value creation’ teams in the industry, with more than 90 full-timers located in Europe, US and Asia. Described by Forbes as KKR’s “in-house consulting team”, it works with deal teams during due diligence and with boards and management teams post-acquisition to develop 100-day value creation plans, drawing on the scale and scope of the KKR portfolio.
Investment banks advising on sell-side M&A with access to pools of financial information are also more frequently using data science to promote figures on low customer churn, up-sell/cross-sell opportunities, bolt-on opportunities and digital transformation when they pitch assets for private equity investment, say sources. However, this approach may favour the sale of companies in some sectors over others – for example fintech and retail (where customer data is more readily available) as opposed to sectors such as manufacturing or infrastructure.
“GPs are getting very good at [using data] to look at historical performance and find they’re very good at buying businesses with good talent and management but they lost money by misreading a market evolution, so they will quantify that and create a go-forward solution with a sourcing algorithm for example,” says the technology consultant.
“Value creation is being much more informed by desired outcomes,” he says. “And I think that sort of starts with very simple views around KPIs and how those KPIs might benchmark to other companies or issues inside the portfolio, then using that to determine which are the areas that we want to go deeper from a data analysis perspective.”
Confidence in such an approach will build quickly, as KPIs around digital transformation, online marketing or ESG link to financial performance and financial models add years of data and transactions.
But sources agree the GPs that keep core metrics in their sights will be the ones winning the arms race.
“Lots of teams have got very sophisticated with analysis of customer segmentation and digital marketing [within portfolio companies] but people that have a discipline around asking targeted questions in a particular deal will drive more advantage versus trying to find unique insights from wide swaths of data.”
Owning insight
This could ultimately drive fund managers deeper into their specialized strategies where they can find an edge over rivals with their proprietary data.
“I think you will see more and more GPs that crop up with insights in a handful of sectors. And that’s their special sauce, that they know the sector inside and out,” says the US-based source.
“Can they skin the cat before others do, though?” asks the technology consultant.
If the value creation advantage is being delivered through a third-party solution, rather than developed in-house, using insight as competitive advantage might eventually become the new business as usual.
