Latest / The Buyer & Seller Podcast with Fexingo: Business Brokers, Exits, and Private Sales Explained / Why a Roofing Company Sold at 55x EBITDA with Recurring Storm Data
Transcript
- Lucas: Alright, let me start with something that stopped me cold this week. A roofing company in Oklahoma City — family-run, maybe twelve trucks, nothing fancy — just sold for an enterprise value that works out to roughly fifty-five times trailing EBITDA. Luna: Fifty-five times? I mean, we talk about rich multiples on this show, but that's software-company territory. What was the secret? Lucas: Right, and that's exactly the question. The buyer wasn't a competitor rolling up roofers. It was a private equity firm that backs insurance- and property-tech platforms. They paid that multiple for one specific asset that the owner, a guy named Mark Tolliver, had been quietly building for fourteen years. Luna: Okay, I'm guessing it's not the fleet of trucks or the inventory of shingles. What was the asset? Lucas: A proprietary database. Every single storm-damage job Mark's company ever did — going back to 2012 — logged by GPS coordinate, date, hail size, wind speed, insurance adjuster notes, and final payout. Over fourteen thousand records, cleaned and normalized. He didn't even realize what he had until a data broker cold-called him two years ago. Luna: That's fascinating. So the buyer basically got a fourteen-year longitudinal study of hail damage patterns across the entire Oklahoma City metro area, tied to actual insurance outcomes. Lucas: Exactly. And this is where the multiple makes sense. The PE firm's portfolio company builds predictive models for property insurers. They spend millions buying this kind of data from third-party aggregators. Mark's dataset was more granular and more accurate than anything on the market because it was ground-truth — every record tied to a real roof replacement. Luna: So he wasn't selling a roofing business. He was selling a data asset that happened to come with a roofing company attached. Lucas: That's the frame. And look, this is where I want to pause for one honest beat. You know, the reason we can dig into stories like this — and we've done eighty-five episodes now — is that a small group of listeners chips in monthly through Buy Me a Coffee dot com slash fexingo. No ads, no sponsors, just people who find these conversations useful. It literally funds the research time. So if this show has moved your thinking forward in some way, that's the only place to help keep it going. Luna: Yeah, and it's genuinely small-scale. Even a few dollars a month makes a difference. Anyway, back to Mark's database. How did he even think to start recording all that detail? Lucas: He told me it started as a personal grudge. He'd had a dispute with an insurance adjuster in 2012 — the adjuster claimed the hail wasn't big enough to warrant a full replacement. Mark knew the guy was wrong, but he had no data to push back. So he started keeping a notebook. By 2015 it was a spreadsheet. By 2018 he'd hired a part-time data-entry person just to keep up. Luna: And he never tried to sell the data before the cold call? Lucas: No. He thought of it as an operational tool — helps him estimate jobs faster, argue with adjusters, train new crews. He didn't see the wider value until the broker explained it. The broker ran a process and got five bids, all from firms in insurance or property tech. The highest bid came from a firm that said the data alone was worth more than the entire roofing operation. Luna: That's the kind of story that makes every Main Street business owner wonder: what data am I sitting on? Lucas: Exactly the question. And the answer is usually: more than you think. The key is to ask yourself — does your business generate records of something that repeats, that has a time stamp, that involves a decision or a transaction, and that someone outside your industry would pay to understand? For Mark, it was damage events. For a plumber, it could be pipe-failure locations and causes. For a tow-truck operator, it could be accident hot spots and response times. Luna: But there's a difference between having the data and having it in a usable form. Mark's data was clean because he was obsessive. Most small businesses have it scattered across paper invoices and QuickBooks notes. Lucas: That's the second lesson. Mark's data became an asset because he standardized it early. Every record has the same fields, the same units, the same date format. That's not exciting work, but it's the work that creates the multiple. The PE firm told me they'd pay maybe ten times EBITDA for a messy dataset they'd have to clean themselves. They paid fifty-five times for one that was ready to drop into their model. Luna: So the premium is essentially the cost they would have incurred to clean it, plus the time value of having fourteen years of history they can back-test. Lucas: Exactly. And here's the thing — Mark's business was doing fine on its own. About two million in revenue, maybe three hundred thousand in EBITDA. Nothing spectacular. The data sale turned a nice lifestyle business into a life-changing exit. And the buyer didn't even want the roofing operations — they sold that piece off to a regional consolidator six weeks after closing. Luna: So the PE firm basically bought the database, and the roofing company was a byproduct. Lucas: Correct. They paid maybe four million for the whole entity, then sold the operating assets for about one million. Net cost for the data: three million. And they'll license that data to insurers for probably five hundred thousand a year, every year, for the foreseeable future. The return on that investment is enormous. Luna: I want to go back to something you said — the broker ran a process and got five bids, all from outside the industry. That implies Mark didn't just stumble into the right buyer. He had help positioning the asset. Lucas: Absolutely. And this is where a good business broker earns their fee. A typical roofer might list the business on a main street marketplace and attract other roofers. That process would have valued Mark's company at maybe four to six times EBITDA — the standard multiple for a service business with no real moat. The broker recognized the data angle, wrote a confidential memorandum that led with the dataset, and went to a completely different buyer universe. Luna: So the broker essentially created the premium by changing the narrative. Lucas: Yes. And that's the third lesson for any business owner listening. If you have a unique asset — data, a proprietary process, a hard to replicate location — don't just sell your business the way everyone else sells their business. Find a broker who can identify the hidden asset and take it to the buyers who value it most. That might mean going to private equity instead of a competitor, or to a strategic buyer instead of a financial one. Luna: But what about the owner who doesn't have a fourteen-year dataset? Is there still a lesson here? Lucas: I think so. Start collecting something today. Whatever you do repeatedly — customer requests, job characteristics, failure modes, seasonal patterns — start logging it in a structured way. Even two or three years of clean data can be valuable if it's unique. And it costs you almost nothing to record it, but selling it later could transform your exit. Luna: I like that. It's a low-cost option on a high-upside future. And worst case, you have better operational data to run your business. Lucas: Exactly. Mark told me he almost deleted the spreadsheet once, back in 2016, because it was taking up too much mental space. His wife talked him out of it. That decision was worth several million dollars. Luna: So the takeaway: don't delete your spreadsheets. And maybe call a broker who thinks about data. Lucas: That's a fair summary. And if you want to dig deeper into how to structure that kind of data for sale, we'll put a link to a guide from the broker who handled Mark's deal in the show notes. For now, I think the big picture is clear: your business might be worth more than you think, but only if you know what you're really selling. Luna: Alright, Lucas. Great episode. See you next time. Lucas: See you, Luna.