Why the numbers on the page are only half the story… and what happens when no one questions the other half.

Professionally, I live in the world of M&A and nothing frustrates me more than watching deals that should have made it across the line collapse. Why they collapse is often glaringly apparent but equally nuanced, and I find myself coming back to this topic again and again because I think that for buyers, sellers and their advisors we need to talk about it more openly. I spend a lot of time thinking about the deals that don’t close. Not the ones that fall apart over relatively straightforward items like price incongruence…those are clean, almost respectful in their simplicity. I’m talking about the ones that die in due diligence. Quietly and expensively. After months of momentum, optimism, and hundreds of hours poured in by people, on all side of the table, who genuinely believed this was going to work. These are the ones where everyone was sure the deal was happening… right up until the moment it wasn’t.

And if I’m being candid, these are the ones that frustrate me the most because so often, they’re preventable.

One of the most glaring issues that surfaces during due diligence and causes deal collapse is the quality of the seller’s adjusted EBITDA.

Here’s what we observe. A business comes to market with reported EBITDA of, say, $3 million. But the seller presents an adjusted EBITDA number, let’s say $6 million, maybe $7 million, once you layer in the normalizations. Owner compensation. One-time legal costs. A legal expense from two years ago. An inventory write-off they insist was non-recurring. A one-time team bonus. Multiple personal trips. A consulting project that’s “completed.” The list grows. And on paper, each one might even be defensible in isolation.

But here’s the thing most people fail to say out loud in those early meetings: when your normalizations represent fifty percent or more of your adjusted EBITDA, you’re not presenting a business with strong earnings… you’re presenting a thesis. You’re asking a buyer to believe that the real business is the one that exists after a dozen hypothetical adjustments… and I want to sit with that for a moment, because that asks for an extraordinary amount of trust. Trust that has to be earned, not assumed.

I’ve sat across the table from sellers…and advised sellers…who are genuinely offended when diligence challenges their add-backs. I understand that reaction, and I have compassion for it, because they’ve been living inside the adjusted number for months or even years. They believe it to be true. Their transaction expectations… on price, on structure, on what life looks like post-close… are all anchored to this adjusted EBITDA and what it implies based on the multiple they expect for their business. So when a Quality of Earnings (QofE) provider comes back and eliminates $1.5 million off the adjusted figure, it doesn’t just change the math. It changes the entire emotional architecture of the deal. There’s a saying in M&A that you want the seller to "buy the boat in their mind"…but in these situations, they’ve already pictured the beautiful 70’ yacht and now they can afford 50’ at best. From what I know of yachts, that’s a fairly significant difference.

And that’s usually where things start to unravel.

Now, I want to be really clear about something, because I think nuance matters here. I’m not saying large normalizations automatically disqualify a transaction. They don’t. We are literally closing a substantial deal as we speak with significant EBITDA add-backs, because the deal makes strategic sense for our client and the add-backs have been verified by the QofE provider. We’ve completed deals where the majority of the EBITDA story only emerged through substantial adjustments. It happens, especially in founder-led businesses where the P&L has been run as a lifestyle vehicle for years. The owner’s salary is $800K when the market rate for a replacement CEO is $250K. There’s a boat buried in the auto expenses. The spouse is on payroll for a role that doesn’t exist. These are real, verifiable, defensible normalizations and they can double the earnings of a business overnight. It happens. But it makes closing the deal that much more challenging, and everyone involved needs to walk into the process with their eyes open about that.

The problem isn’t size. The problem is verifiability. And the problem compounds when sellers or their advisors treat the adjusted EBITDA like a marketing number rather than an auditable one.

Take inventory write-offs as an example because this is one of the most commonly mishandled add-backs in the mid-market and it deserves its own conversation. A seller writes off $1.2 million in obsolete inventory and presents it as a non-recurring normalization. On the surface, it looks clean. But best practice demands that you trend that charge over three to five years. If the business has been writing off $200K to $400K annually as a matter of course, then only the amount above that baseline is legitimately non-recurring. The rest is a real cost of doing business… and it belongs in EBITDA.

When I see a seller add back one hundred percent of an inventory write-off without any historical context… that tells me something. Not necessarily that they’re being dishonest but that they haven’t been properly advised, or that their advisor is more focused on winning the engagement or swinging for the fences than on building a defensible story that will actually close the transaction. Either way, it’s a diligence problem that’s going to surface. And when it surfaces late in the process, the cost is measured in more than dollars. It’s measured in relationships, in trust, in the emotional toll on a seller who genuinely thought they were weeks away from closing.

Because here’s what actually happens when an add-back gets challenged in diligence and I think it’s worth walking through this, because the human side of it rarely gets discussed. The buyer’s QofE team flags the issue. The buyer’s deal team calls the seller’s advisor. The seller’s advisor calls the seller. The seller gets defensive likely because they’ve already mentally spent the proceeds. Now you’ve got a principal who feels like they’re being called a liar, an advisor who’s trying to hold the deal together, and a buyer who’s wondering what else they haven’t been told. Trust erodes. Timelines stretch. Legal bills climb. And in the background, the buyer’s investment committee is starting to ask whether this deal is worth the trouble.

I’ve seen deals survive this although it’s usually through creative structuring or because the deal itself is highly strategic and both parties are motivated to find a path forward. Earnouts, escrows, share rollovers, holdbacks, reps tied to specific financial metrics. These are tools that exist precisely for this type of situation: to bridge the gap between a seller’s view of value and a buyer’s tolerance for risk. But they’re imperfect instruments, and they work best when both sides enter them with realistic expectations and a genuine willingness to be creative together. When trust has already been damaged by aggressive add-backs that didn’t hold up…the appetite for that kind of collaborative compromise tends to disappear.

So what’s the takeaway here? It’s not “don’t normalize.” Normalizations are essential to presenting the true earnings power of a business, and they’re present on nearly every mid-market M&A deal. The takeaway is that every single add-back needs to be treated as though it will be verified under audit conditions. Because functionally, it will be. The QofE provider is going to pull the thread. The buyer’s CFO is going to ask the uncomfortable question. And if the answer is “well, management told us it was non-recurring”… without supporting data, without historical trend analysis, without third-party corroboration… that add-back is going to get killed.

And sometimes, it takes the deal with it.

From my experience there are two key ways to get ahead of this…and I share them because I genuinely want to see more deals close well, for everyone involved. First, the seller should work to reduce normalizations before going to market. Get rid of all non-core business expenses. Stop with the one-time projects. Put yourself in the eyes of a buyer and ask what they would want to see in this business post-close. Operate the business the way a buyer would want to take it over. And second, ensure every normalization that does remain is highly defensible and can be easily supported by the seller’s external accountant and during a QofE. If you can’t hand it to a third party and have them validate it independently, it’s not ready or suitable for diligence.

The best-run M&A processes I’ve been part of, the ones I’m proudest of, are the ones where the hard work is done before going to market. Pressure-test every normalization. Trend every line item. Anticipate the diligence questions and have the answers ready before they’re asked. It’s less exciting than building a flashy CIM with an aspirational EBITDA number… but it’s the difference between a deal that closes and one that dies on the table.

Because in this business of M&A, adjusted EBITDA isn’t just a number. It’s a promise. And promises that don’t hold up do kill deals.

‍