An adjusted EBITDA schedule is an argument, not a calculation. Every line on it is defensible taken alone, which is exactly why the total so often is not.
There is a test that resolves most of them, and it is not whether the expense was genuinely unusual. It is whether the work behind the expense still has to happen after closing, and whether somebody will have to be paid to do it.
THE TEST
Ask of each add-back: after closing, under a new owner, does this cost come back?
If the answer is no, the adjustment is real. If the answer is yes but in a different form, the adjustment is fiction wearing the clothes of an adjustment. And if the answer is that it depends on whether the buyer chooses to keep doing something, the line belongs in a separate category that gets negotiated rather than added back.
Most schedules mix all three and present the total as one figure.
THREE THAT USUALLY SURVIVE
A genuine one-time legal matter that has concluded. The case is settled, the counterparty is gone, and no successor dispute is pending. This is the cleanest category and it is also the one buyers should confirm rather than accept, because "concluded" and "quiet" are different states.
A discontinued line of business with its costs cleanly separable. If the company exited a service, closed the location and released the staff, those costs do not return. The word doing the work is separable — shared overhead that was allocated to the discontinued line does not disappear with it.
A non-recurring capital or systems project that is finished. An implementation that is complete and will not need repeating within the hold period is a real adjustment, provided the maintenance and licensing that follow it are still in the run rate.
THREE THAT USUALLY DO NOT
Owner compensation above market. This is the biggest line on most schedules and the most misunderstood. Adding back the difference between what the founder paid themselves and market rate is legitimate only if a market-rate person can do the same job. Frequently the founder was doing three jobs, and the adjustment silently assumes one replacement.
Family members on payroll. Sometimes genuinely non-working, in which case the add-back is real. Often performing real functions at below-market rates because of who they are, in which case the buyer inherits both the vacancy and the wage gap.
Personal expenses that are actually relationship costs. Client entertainment, memberships, travel and vehicles get added back as owner perks. In a relationship-driven business some portion of that spending is how the accounts get retained, and cutting it is not a cost saving but a revenue decision.
THE OWNER'S COMPENSATION PROBLEM
The largest single add-back in most lower-middle-market deals is the founder, and it is worth stating plainly what that line assumes.
A founder working sixty hours a week at below-market compensation is a genuine economic distortion, and adjusting for it is correct in principle. What the adjustment usually gets wrong is scope. The replacement is not one salary. It is whatever combination of people is required to cover selling, the key relationships, the final approval on anything unusual and the institutional memory that was never written down.
The honest version of that line is not a number. It is a staffing plan with a cost attached, and building it is the single most valuable hour of diligence available on most deals.
None of this argues against add-backs. Adjusted figures exist because reported figures genuinely do not describe the business a buyer is acquiring. It argues for a schedule where every line has been asked the same question.
GEX Management works with buyers on the quality of the earnings and the quality of the business behind them. If an add-back schedule is doing more work than it can carry, that is usually where the value is decided. Start a conversation.