Latest / The Buyout Show with Fexingo: Private Equity, Roll-Ups, and Business Acquisitions / How Private Equity Is Buying Up Data Broker Firms
Transcript
- Lucas: Luna, I want to talk about one of the quietest roll-up plays in private equity right now — and I bet most of our listeners have never heard of the companies involved. Luna: Okay, you've got my attention. Which industry? Lucas: Data brokers. The companies you've never heard of that know everything about you — where you shop, who you vote for, what you stream, probably your blood type if they can find a way. Luna: And PE is consolidating them? I know a few big names like Acxiom and LiveRamp, but I thought that market was already pretty concentrated. Lucas: Not even close. There are hundreds of small to mid-sized data brokers in the US alone, each with a niche — one might specialize in automotive purchase data, another in credit card transaction data, another in health-related browsing behavior. And private equity is stitching them together. Luna: What's the economic logic? These are companies that sell data to marketers, right? Lucas: Right. And the economics are beautiful from a PE perspective. The typical data broker has gross margins north of 70 percent, sometimes 80 percent. They sign multi-year contracts with big clients — ad agencies, insurers, political campaigns — that renew automatically. High switching costs because the data sets are proprietary and custom-integrated. Luna: So predictable cash flow, sticky revenue, high margins. Textbook private equity target. Lucas: Exactly. And there's a scale advantage that's unique to data. If you own three different data sets on the same consumer, you can cross-reference them and build a richer profile. That's worth more per record than any single data set alone. PE firms call it a 'data moat'. Luna: Can you give me a concrete example of a roll-up in this space? Lucas: Sure. In 2021, Golden Gate Capital took Neustar private in a deal valued at about $2.9 billion. Neustar was mostly known for identity resolution — matching online and offline identities. Since then, Golden Gate has been tucking in smaller data companies under that umbrella. I don't think most people realize how much of the identity graph market is now owned by pe backed platforms. Luna: And what about the Acxiom spin-off? That was part of a bigger PE play too, right? Lucas: Yes. Acxiom was owned by Interpublic Group, but in 2018 they spun off their data business into a separate company called LiveRamp. LiveRamp then went on an acquisition spree — buying Data Plus Math, a TV attribution company, for about $150 million in 2021, and then later merging with Acxiom's marketing solutions division in 2023. The combined entity is a data powerhouse. Luna: And who's behind that? Is it PE money? Lucas: LiveRamp itself is publicly traded, but several PE firms have been major shareholders. Also, a number of the smaller deals in the space are funded by PE. For instance, Alpine Investors has been rolling up data companies in the business to business data space. And there's a firm called Serent Capital that's built a platform around data enrichment services. Luna: So the playbook is: buy a platform company, then bolt on niche data sets. What kind of multiples are they paying? Lucas: For the smaller tuck-ins, typically 10 to 15 times EBITDA. For the bigger platform deals, you're looking at 12 to 18 times, depending on the growth rate and the stickiness of the contracts. The beauty is that once you combine data sets, you can often increase the price per record and improve retention, so the combined entity trades at a higher multiple than the sum of its parts. Luna: But there's a regulatory risk, right? Data privacy laws are changing fast. Lucas: That's the wild card. The FTC has been cracking down on data brokers that collect sensitive location or health data without consent. In 2024, they banned a company called X-Mode from selling location data after it was caught tracking people visiting reproductive health clinics. And there are now state laws — California, Virginia, Colorado, Connecticut — that give consumers the right to opt out of data sales. Luna: Does that kill the economics? If everyone opts out, what's left to sell? Lucas: The thing is, most people don't opt out. The opt-out rates are something like 2 to 5 percent. And the data brokers are getting smarter — they're moving from selling raw data to selling 'insights' and 'analytics,' which are harder to regulate. The data itself isn't sold; it's used to power a software platform. That changes the legal classification. Luna: So PE is essentially betting that regulation will be slow and that consumers won't change their behavior. Seems like a safe bet so far. Lucas: Exactly. And if a federal privacy law ever passes — which has been debated for years — it could actually help the large platforms because compliance costs would squeeze out smaller players. That would accelerate the roll-up. Luna: Okay, so give me a sense of the size of this market. How much data broker revenue is there to consolidate? Lucas: The data broker industry in the US is estimated at over $200 billion in annual revenue. That includes everything from credit reporting to marketing data to fraud detection. A significant chunk of that is addressable by PE. The top five firms — Acxiom, Epsilon, Experian, Oracle's data cloud, and TransUnion — probably control less than 30 percent of the market. There's huge fragmentation. Luna: And those big players are also acquisitive. Oracle bought BlueKai back in 2014 for about $400 million. Salesforce bought Krux for a similar amount. So the tech giants are consolidating too. Lucas: Right. But PE has an advantage: they can be patient. They're not trying to integrate data into a broader cloud platform. They're just trying to maximize cash flow and sell the company in five to seven years to a strategic buyer or take it public. That focus on cash generation is a different playbook. Luna: It also means they're less concerned about the ethical implications, I'd argue. They just want the numbers. Lucas: That's a fair point. And it's one reason why data brokers are a controversial asset class. But from a pure investment perspective, the returns have been strong. A 2025 study by a group at Duke looked at pe backed data companies and found they outperformed their standalone peers by about 400 basis points in EBITDA margin expansion over three years. Luna: Four hundred basis points is huge. How do they achieve that? Lucas: Mostly through cost synergies — combining technology stacks, eliminating duplicate vendor contracts, and cross-selling data sets to each other's clients. Also, they often replace legacy sales teams with more aggressive compensation structures. It's the same playbook as rolling up dental practices or vet clinics, just applied to data. Luna: So data is the new dentistry. Consolidate, optimize, flip. Lucas: Pretty much. And the interesting thing is that data brokers don't have the same labor constraints. You don't need to keep a dentist happy after you buy their practice. You just need to keep the data flowing and the servers running. Luna: That feels more scalable. But also creepier, honestly. Lucas: It is. And I think that's part of the reason why these deals happen under the radar. You don't see press releases about 'Private Equity Buys Your Browsing History.' It's all very quiet. Luna: Let me ask about a specific recent deal. Did you see the one where Alpine Investors bought a company called InfoGroup in 2024? Lucas: Yeah. InfoGroup is a classic example. They aggregate business data — names, titles, emails, phone numbers — and sell it to sales teams. Alpine paid about 14 times EBITDA, which is on the higher end, but InfoGroup had been growing at 12 percent annually with gross margins above 75 percent. It's a cash machine. Luna: And then Alpine will likely combine it with other portfolio companies in the sales intelligence space. They also own a company called ZoomInfo? No, ZoomInfo is public. But they own similar assets. Lucas: Right. ZoomInfo went public in 2020 and is now valued at around $5 billion. That's the exit PE firms dream of — take a collection of data assets, put them in a growth wrapper, and go public. The multiples for data platform companies have been 8 to 12 times revenue at IPO. Luna: How does that compare to the multiples they pay on the acquisition side? You said 10 to 15 times EBITDA, which is maybe 2 to 4 times revenue. So there's a big markup. Lucas: That's the arbitrage. Buy at 3 times revenue, take public at 8 times revenue, and in between you've grown revenue by cross-selling and adding new data sets. It's a proven formula. The question is how long it lasts before regulators or consumers shut it down. Luna: Well, and before I forget — if these conversations have moved your work forward in some small way, a couple of dollars a month is genuinely what keeps them going. Buy me a coffee dot com slash fexingo, if you've gotten something out of them. Lucas: Yeah, it really does make a difference. Every bit helps us stay independent and dive into these niche corners. Luna: So back to the regulatory risk. Do you see any catalyst that could pop this bubble? Lucas: The biggest would be a federal privacy law with a private right of action — meaning consumers could sue data brokers for mishandling their data. That would upend the cost structure. But the current Congress seems gridlocked on that. The state-level patchwork actually favors large players because they can afford compliance teams. Luna: So PE is betting on continued gridlock and consumer apathy. That's a bet I'd take. Lucas: It has paid off so far. The data broker roll-up is probably one of the least noticed but most profitable themes in private equity over the last five years. And I think we'll see more of it, especially as AI models need more training data — where do you think they get it? Luna: Right from these same data brokers. So the AI boom just makes this data even more valuable. Lucas: Exactly. Data brokers are the pick and shovel sellers in the AI gold rush. And private equity is busy buying up all the shovels.