Building a Proprietary Investment Engine Using Public Data for a Top PE Firm

Tribe

A leading PE firm came to Tribe with a specific challenge: to build an ML-driven toolkit that would give them unique insights into one particular vertical of interest using only publicly available data. The core thesis was that if you could find signal in publicly available data, you could enrich it with private data sets.

​​“We wanted to pressure test this thesis with one sector and see if there are data sources that exist in the world, publicly available to start, that could be informative and potentially predictive in terms of what type of asset and where we could invest,” said the PE investment partner leading the project.  

The investment team laid the groundwork, identifying the target industry and what kinds of questions the investment team needed the data to answer in order to be confident in making an investment. Then they reached out to Tribe.

“I was nervous about locking ourselves into the cost of the project,” said the investment partner. “You know, I see the bios and resumes of these technologists, but I’m out of my depth assessing those types of candidates. But right away, we got a lot of comfort from Tribe that there’s flexibility built into the relationship and off ramps along the way. So that gave me a lot of confidence to move forward.”

From there, Tribe assembled a team of three specialists and scoped a project for speed and maximum impact. The plan was three sprints – six weeks – to launch an MVP to test the team’s hypothesis about the power of data science as a true differentiator.

The Tribe team

Tribe put together a team of three specialists – a technical product manager, an ML engineer specializing in predictive models, and a data scientist with a background in working with public data sources. This team then worked directly with the investment team to understand their goals.

“We didn’t have to spend a lot of time doing back-to-basics stuff,” said the investment partner. “ I was really impressed by how much the Tribe team got exactly what we were trying to do without much teaching.”

“We were very impressed with all of it,” they added, “The work product but also the communication with non-technical people like myself.”

“Over the course of a week and after only a few conversations the Tribe team deeply understood what we were trying to do,” said the PE firm’s senior vice president (SVP) of data.  “And they had contributed not only in the execution of the plan but brought new ideas, additional data sets, and tested more than anyone thought possible over the period of time we’d set out. We got much more out of this project than we thought we would. And that’s in large part due to the quality of the people Tribe brought in.”

"We got much more out of this project than we thought we would. And that’s in large part due to the quality of the people Tribe brought in.”
Senior Vice President of Data, PE Firm

The approach


The investment team knew they wanted some way to score a specific geographical region within the US to determine how attractive it was for investment.

“The investment team knew at a high level what they wanted to accomplish,” said Nathan, the Tribe technical product manager on the project, “But it really was an experimental project. Every sprint we were trying new data, new approaches.They knew more about this core vertical than we ever will, so we paired their industry knowledge with our data expertise to help us find opportunities that would give them an edge.”

The Tribe team started with the hypothesis that they could use data to isolate the key predictors of adoption that would enable the investment team to better identify investment opportunities.  From there, they gathered publicly available data ranging from census data and social demographics metrics to consumption patterns and upcoming infrastructure changes and fed them into their custom ML model.

“Our analytical model ingested hundreds of predictive factors and aggregated those into themes,” said Winston, the lead data scientist on the project. “What we found is that the hypothesis was correct – there are specific past development and demographic variables that are highly predictive of activity in certain areas.”

Once the Tribe team had proven the ML model could pull signal out of public data in order to enable differentiated market insights, the next step was to package the technology in a way the investment team could easily use in their investment process.

“We wanted a visual for them to connect to the data in a way that would be easy for a non-technical audience to understand,” said Gabe, the ML engineer on the project. “So we created a heat map of the US, color-coded at the county level based on the score we’d developed to show internal analysts and external companies where to focus their resources.”

Six weeks into the project, the investment team had their MVP.

“We did in 6 weeks what I’ve seen teams take 6+ months to do,” said Winston. “Partly this is because the customer was so engaged with us in making the project a success. They knew what they wanted to get out of it and worked closely with us all the way through. And then Tribe knew each of our strengths and where we could add value. As a team it was like a mind meld. It was magic.”

"Tribe knew each of our strengths and where we could add value. As a team it was like a mind meld. It was magic.”
Winston, Tribe Lead Data Scientist

The results


With the toolkit in hand, the PE team saw multiple levels of value add. First, they were able to evaluate potential investments at increased speed and confidence. “We recently looked at a company operating in this vertical and were able to evaluate their growth plan,” said the investment partner. “It’s a much better way to go about evaluating how confident we should or shouldn’t be about a certain set of assumptions.”

Second, it helped the investment team showcase the value they could bring to prospective portfolio companies beyond capital. “We had conversations where they could see the value of the insights from a tool like this with public data,” they said. “And, from there, it’s easy to imagine what they could do with proprietary data. The kinds of insights they could get. Those are the conversations that start to get exciting. And, in some cases, it even led to more favorable terms.”


“Especially in this market, if you’re going to win, then you have to win on speed and price,” said the SVP of Data. “And the only way you’re going to get there is if you have the benefit of the conviction that a tool like this can give you. Then there’s that added bonus that the management team is going to have their pick of capital partners. So if you’re winning on speed, conviction, and then also direct benefit to their business – that’s the holy trinity of winning the best deals.”

“Especially in this market, if you’re going to win, then you have to win on speed and price. And the only way you’re going to get there is if you have the benefit of the conviction that a tool like this can give you."
Senior Vice President of Data, PE Firm

Building a roadmap


By leveraging public data, the project was successful in augmenting the deep expertise of the PE investment team and answering some of their most pressing business questions.

Now that phase one has proven its value, the investment team is planning to expand the scope of the project to incorporate data sets from portfolio companies, partners, and third party vendors. They are also excited about expanding the project across other verticals.

“The project we did here seems very repeatable in other sectors,” said the investment partner. “Seeing these results, we immediately think where else can we take these publicly available data sets to be predictive about the best type of investment for that market or asset?”

While the next phase is still in development, for the investment team, the overarching value has been clear. “A big focus for us in this project was finding out how and where we can make investments where data science is a true differentiator,” said the investment partner.

“The vast majority of our competitors are still fully discretionary using traditional approaches to sourcing, evaluating, and executing deals,” added the SVP of data. “But data science and technology are the last mile of innovation for almost every industry on the planet. And it’s a natural fit for private equity investing.”

"Data science and technology are the last mile of innovation for almost every industry on the planet. And it’s a natural fit for private equity investing.”
Senior Vice President of Data, PE Firm

Related Case Studies

Case Study

How Fantasmo is using machine learning to make GPS obsolete

Case Study

How Tribe AI Built a Model on GCP That Increased Security Questionnaire Auditor Efficiency by 55%

Case Study

How Togal AI Built the World's Fastest Estimation Software on AWS

Case Study

Togal.ai powers the construction industry into the age of machine learning

Case Study

Taking a Data-Driven Approach to Women's Fertility with Rita

Case Study

How Wingspan built a machine learning roadmap with Tribe AI

Case Study

How Tribe AI Shaped Truebit’s AI Strategy

Case Study

Insurance Company Uses ML to Optimize Pricing

Case Study

How Nota Built a Roadmap for AI-enabled Journalism with Help from Tribe

Case Study

GenAI Solutions: How Bright Transformed Workforce Training with Tribe AI

Case Study

Kettle uses machine learning to balance risk in a changing climate

Case Study

Tribe AI & Venture Labs: Accelerating Startups with Tailored AI Expertise

Case Study

How Tribe Helped Reservoir Bring Finance Infrastructure to NFT Trading

Get started with Tribe

Companies

Find the right AI experts for you

Talent

Join the top AI talent network

Close
Tribe