The Add-Back Schedule Buyers Actually Believe
A founder once handed me an adjusted EBITDA schedule with thirty-one line items on it. The reported number was a little under four million dollars. The adjusted number was six. Every one of those thirty-one add-backs had a story, and the founder could tell me each one without looking at the page. The problem was not that the stories were false. Most of them were true. The problem was that a buyer was going to read the same page and see a company that needed thirty-one explanations to be worth what we were asking, and that impression would cost more than any single add-back could add.
This is the part of a sell-side process that advisors get to influence most directly and manage worst. The add-back schedule is where an advisor either builds credibility or spends it. Build it well and the buyer accepts your bridge from reported earnings to normalized earnings and moves on to the next issue. Build it badly and the buyer discounts the whole number, because a schedule that reaches too far teaches a buyer to distrust everything on it, including the items that are unimpeachable.
Here is the mechanic that founders do not see and advisors sometimes forget. A buyer does not evaluate add-backs one at a time and keep a running total. A buyer forms a judgment about the character of the schedule in the first few minutes, and then reads the rest of it through that judgment. Three aggressive add-backs near the top of the page change how a buyer reads the twenty-eight legitimate ones below. The order and the tone of the schedule are doing work before the buyer has checked a single figure against the general ledger.
So the question I ask before any adjusted number leaves our hands is not whether an add-back is defensible in isolation. It is whether it belongs to the category of adjustments a buyer will believe. In my experience the add-backs that survive diligence share three properties, and the ones that get struck fail at least one of them.
The first property is that the item is genuinely non-recurring or genuinely personal, and it is one or the other cleanly. The owner's compensation above a market replacement salary is a real adjustment. The one-time cost of relocating a warehouse that will not move again is a real adjustment. The half of the marketing budget the founder now believes was wasted is not, because it recurred every year and the business as it actually operated included it. Buyers draw a hard line between costs that leave with the seller and costs the seller wishes were lower. The second kind is not an add-back. It is an opinion about how the business should be run, and the buyer has their own.
The second property is that the item is documented in a form a buyer can verify without taking the founder's word for it. An add-back that lives only in the founder's memory is worth nothing in diligence, because the quality of earnings provider cannot tie it to anything. The owner's country club membership running through the company is a clean adjustment if it is a line in the general ledger with the founder's name on it. It is a coin flip if it is buried inside a travel and entertainment account with forty other transactions and no support. The adjustment does not fail because it is illegitimate. It fails because it is unprovable, and an unprovable add-back is indistinguishable from an invented one by the time a buyer's analyst is looking at it.
The third property is proportion. A schedule where the adjustments total ten or fifteen percent of reported EBITDA reads as normalization. A schedule where the adjustments approach half of reported EBITDA reads as a different company being constructed on paper, and a buyer's instinct is to ask what the business actually earned when it was simply operating. Past a certain ratio, each additional add-back does not raise the number a buyer will underwrite. It lowers their confidence in the base. There is a point where adding a true add-back makes your case weaker, and most schedules I am handed have sailed past it.
That is the counterintuitive part, and it is the part an advisor exists to enforce. The advisor's job on the add-back schedule is often subtraction. I have spent more hours talking founders out of legitimate add-backs than into them, because a defensible number that a buyer accepts whole is worth more than an aggressive number the buyer picks apart. When a buyer strikes eight of your thirty add-backs, they have not just removed eight items. They have earned the right, in their own mind, to be skeptical of the twenty-two that remain, and they will use that skepticism everywhere else in the deal.
The cost of getting this wrong is measurable in the same place every other diligence failure shows up, which is the gap between the letter of intent and the wire. Across the transactions we have studied, post-LOI price adjustments occurred in roughly two-thirds of deals, and the median adjustment was close to ten percent downward. A padded add-back schedule is one of the surer ways to end up in that two-thirds, because the quality of earnings process is built precisely to test normalizations, and a schedule that overreaches gives the buyer both a lower number and a reason to trust nothing else. I have written before about what sell-side advisors get wrong about quality of earnings timing, and the add-back schedule is where that timing pays off or does not.
The practical discipline is to build the schedule the way a quality of earnings provider will test it, before they test it. For every add-back, write the one sentence of support a buyer's analyst will need, and attach the document that proves it. If you cannot write the sentence or find the document, the add-back does not go on the page. Sort the schedule so the strongest, most obviously personal items sit at the top, because the buyer's judgment of the whole schedule is forming there. And run the total against reported EBITDA and ask honestly whether the ratio reads as normalization or as invention. If it reads as invention, cut until it does not, starting with the items that are true but weak.
This also changes what an advisor should be doing in the year before a process, not just in the schedule itself. Many add-backs are unprovable at exit because nobody set them up to be provable while the business was running. The country club membership, the family member on payroll who does not work in the business, the personal vehicles, the one-time system implementation, all of these are far easier to defend if they were coded cleanly in the general ledger as they happened rather than reconstructed from memory under deal pressure. That is one more reason the credibility of the exit is built in the operating years, a theme that runs through the three patterns that predict whether a deal closes at the LOI price.
The founder with thirty-one add-backs was not trying to deceive anyone. He was proud of how lean the real business was, and the schedule was his way of showing it. We took the page down to nine items, each one documented and each one clearly the kind of cost that would leave with him. The adjusted number came down by about four hundred thousand dollars from where he started. It closed at that lower adjusted figure with no further erosion in diligence, which made it worth more than the higher number would have been after a buyer finished taking it apart. A number a buyer believes is worth more than a number you have to defend. That is the standard we hold ourselves to at Cordis Group.