The HVAC roll-up that almost happened, and the four months the founder will not get back.

The kitchen was the one his wife had wanted for twelve years. They had remodeled it the spring before, paying for it out of the year the business had run hottest in two decades. Granite countertops, soft-close drawers, a six-burner stove she had picked out herself from a showroom in Columbus. They were sitting at the island, two coffees, when the LOI came in over email.

He read it twice. He read the EBITDA number, $4.8M, and the multiple, 7x, and then the working capital adjustment, the customary holdback, the earnout structure. He did the arithmetic in his head, the way he did the bid math on a large commercial install. He looked up at his wife. He said, “they are offering eleven million dollars after the earnout if we hit.” His wife said, “is that good.” He said, “I think so.”

It was September 2024. The buyer was a regional roll-up operator backed by a lower-middle-market PE fund with a fifteen-year horizon. They were rolling up residential and light-commercial HVAC across three Midwest states. Our founder, who we will call Dale, was the second largest independent operator in his county and the fourth largest in his metropolitan area. He had built the business across nineteen years, from a single van to thirty-six trucks and a $4.8M EBITDA base. His wife had carried him through two of those years. The eleven million dollar number was the number that would make the years worth it.

Four months later, the deal was dead.

What the diligence found

The buyer’s quality-of-earnings team arrived in week one, on schedule. They wanted three years of financials reconciled to tax returns, three years of bank statements, a working trial balance reconciliation, customer-level revenue detail, vendor concentration, and the standard package of legal, employment, and insurance documentation.

Dale’s accountant produced what he could. The tax returns reconciled to the management financials within $40,000 across three years, which would have been acceptable if the unreconciled portion had been documentable. It was not. Some of it was personal expense run through the business at the accountant’s discretion, the boat, the wife’s car, a sliver of vacation home maintenance. Some of it was inventory that had been carried at cost on the management financials but never adjusted for write-downs Dale had taken in his head but not on paper. Some of it was uncashed deposits from residential service customers that had been recognized as revenue but not yet earned.

Each of these was knowable. Each was, individually, small. Together they came to about $180,000 across three years. The buyer’s diligence team flagged each one. The accountant explained each one. The buyer did not adjust their EBITDA number for the items individually, but they did adjust their confidence in the management financials as a whole. That confidence adjustment is the one that costs the deal.

By week six, the buyer was asking new questions. Did Dale’s largest commercial customer, which represented 22% of revenue, have any contract beyond the verbal one? It did not. Did Dale have a documented succession plan for his senior tech, who was 61 and held the relationships with the four largest commercial accounts? He did not. Did Dale have an employee handbook current to within the last three years, with the wage-and-hour policies updated for the new state regulations? He did not.

Each missing document was reported to the buyer’s investment committee. The committee, which was looking at fourteen other roll-up candidates that quarter, did the comparison the founder did not realize was happening. Dale’s business was attractive in revenue, in geography, in fleet size. It was unattractive in documentation, in customer contract structure, in succession depth. The committee priced the risk.

The QofE landing

In week eleven, the buyer’s quality-of-earnings report landed. The headline number had come down. The buyer was no longer offering 7x of $4.8M. They were offering 6.2x of $4.4M, with a larger earnout, a more aggressive working capital peg, and a longer indemnification survival period. The new headline number was $7.8M with the earnout fully achieved, $6.4M without.

Dale read the new offer. He read it a second time. He went and found his wife.

“They lowered the number,” he said. “By how much?” “About three million dollars.” “Why?” “They found things in the books.”

His wife is a careful person. She is also the person who has lived with him for thirty-one years. She asked him whether the things were real, and Dale said yes, they were real but small. She asked him whether the things had been there for years, and he said yes, they had. She asked him whether anyone had ever told him this would matter, and he said no, no one had.

She did not say the thing she was thinking. He did not say the thing he was thinking. They sat in the kitchen for forty minutes.

The unwind

Two weeks later, Dale called the buyer and asked whether he could withdraw from the process. The buyer was professional about it. They wished him well. They asked, with the matter-of-factness of people who had done this dozens of times, whether he might be interested in re-engaging in 18 to 24 months. He said he might be. He hung up the phone and sat at his desk for a while.

The four months had cost him roughly $80,000 in advisor fees and an estimated 25% of his operating attention. The business had taken a small but real hit in the months he had been heads-down in diligence; the senior tech had noticed that Dale was distracted, and one of the largest commercial relationships had drifted slightly, costing them about $200,000 in run-rate revenue by year end.

The biggest cost was psychological. Dale had told his wife it was happening. He had told his senior tech, in confidence, that he might be retiring within the year. He had run the math, in his head, on what he would do with the eleven million dollars. He had imagined the kitchen with the windows open in a place that did not have winters. He had not done any of these things performatively. He had done them because the deal was real to him.

Then it was not.

What he is doing now

We caught up with Dale eight months after the deal died. He had hired a fractional CFO at $4,500 a month and given them a single mandate: clean up the books to the point that another QofE would not find anything. The CFO worked through the personal-expense reconciliation, the inventory write-downs, the deposit timing. By month six, the management financials reconciled to tax returns within $8,000. The work cost him $35,000 plus his own attention.

He had also started a conversation with his largest commercial customer about a three-year master service agreement. They had been doing business on a handshake for eleven years. The customer was surprised Dale wanted paper, but they signed paper, with reasonable terms and an annual price adjustment mechanism. The MSA changed the way the customer concentration reads to a buyer; the concentration is the same, but the durability of the relationship is now documented.

He had not yet made the senior tech succession hire. He said he was working on it. We did not press.

Dale’s plan, as he laid it out, was to spend 2025 and the first half of 2026 doing the operational work the diligence had surfaced. He plans to re-engage with the same buyer, or with a different buyer, in the fall of 2026. He thinks the new offer will be at the original 7x of $4.8M or better, with a smaller earnout and a cleaner structure. He thinks he is two years away from the same eleven million dollars he was offered in September 2024.

He may be right. The work he is doing is the right work. The documented contract with the commercial customer, the cleaned-up books, the eventual succession hire, all of these will move the buyer’s number when he comes back to market. The two-year delay is not free, but it is recoverable.

What was not recoverable

The four months are not recoverable. The conversation with his wife. The conversation with the senior tech. The kitchen in September, when he sat at the island and did the arithmetic and looked up at her and said the number out loud. None of that is recoverable. The deal did not just fail. The version of his life that included the deal closing also failed, and he had been living in that version, however quietly, for sixteen weeks.

When founders talk to us about the Misalignment Tax, they talk about it as a financial number. It is a financial number. But every founder we have spoken with who has been through a failed transaction also describes the second tax, which is the emotional tax of having imagined a future and then losing it. That tax is not in any model. It is also not in any of the documentation work Dale is doing now.

The single useful thing to say to a founder reading this two to five years out from their own transition is this: the documentation work is the cheapest of the five fixes, it can be done in ninety days, and the cost of skipping it is not just the financial discount. The cost is also the four months that Dale will not get back, sitting in the kitchen that his wife had wanted for twelve years.

The work is cheap. The not-doing is expensive. The scorecard takes twelve minutes.