The add-back calculation determines the earnings figure a buyer applies a multiple to. Get it right and the multiple lands on defensible adjusted EBITDA. Get it wrong and diligence strikes the inflated portion and reprices the deal.
Across operators with $20M to $100M in revenue, 30 to 50 percent of claimed add-backs fail institutional review. The result is an 8 to 18 percent EBITDA defense gap that translates directly into multiple compression. This article walks the eight accepted categories, the six routinely rejected, and the discipline that defends EBITDA before a buyer is in the room.
An EBITDA add-back is an adjustment that restates reported EBITDA to reflect normalized, sustainable, transferable earnings. It removes non-recurring items, owner-specific costs, and one-time events that a buyer will not inherit. Accepted add-backs increase the earnings a multiple is applied to and therefore increase enterprise value. Rejected add-backs do not, regardless of how the seller has presented them.
Eight categories are accepted when each one is documented and passes the test of non-recurring, non-operational, and transferable. Select a category to read the calculation and what defends it in diligence.
Restate owner pay to the market rate for the role a buyer would have to hire, in either direction.
The owner draws $850,000. Market compensation for the equivalent operating role is $400,000. Add back the $450,000 of above-market pay.
This is the single largest add-back in most lower-middle-market deals. It is accepted only when the market rate is benchmarked to a role, not asserted.
Six categories are rejected often enough that they account for most of the 30 to 50 percent of claimed add-backs that fail review. Each one fails the same test: it is not non-recurring, not non-operational, or not transferable to the buyer.
An event claimed as one-time that appears in two or more years fails the non-recurring test. A buyer reads a pattern as a cost of the operating model, not an adjustment.
The argument that the owner should have paid themselves more, so EBITDA is understated, is never accepted. Normalization corrects above-market pay, not hypothetical below-market pay the owner chose.
Annualizing a partial-year revenue gain is rejected unless signed contracts support it. A buyer will not pay a multiple on revenue that exists only in a projection.
Ordinary marketing reframed as one-time growth investment is rejected. If the spend is required to hold the current revenue, it is an operating cost, not an add-back.
Ongoing development reframed as non-recurring is struck. If the business needs the spend to keep its product current, it recurs and stays in EBITDA.
Adding back lost-customer revenue as if it would have continued is rejected outright. The buyer underwrites the business that exists, not the one that left.
Owner compensation normalization is the largest add-back in most lower-middle-market transactions, and it is the one most often calculated loosely. The discipline is to benchmark actual pay against a defensible market rate, then to normalize family payroll and benefits with the same evidence standard the rest of the bridge holds. The terms used here are defined in the glossary.
Identify total owner compensation in the P&L, then benchmark it to the market rate a buyer would pay to hire the equivalent operating role. The add-back is the above-market difference, documented to a salary survey or comparable role, not asserted.
Family members on payroll at above-market rates, or for roles the buyer would not fill, are normalized to the cost of replacement or removed entirely. Each name is tied to a function and a defensible market rate.
Owner-specific benefits, vehicles, insurance, and discretionary perquisites are separated from the benefits a buyer would actually provide. The normalization keeps what transfers and removes what does not.
The bridge moves from reported EBITDA to adjusted EBITDA one documented line at a time, so a buyer can trace every dollar of normalization. The worked example below starts from a $4.20M reported base and reaches a defensible $5.98M, an adjustment of $1.78M, with each line evidenced.
| Line Item | Adjustment | Running EBITDA |
|---|---|---|
| Reported EBITDA | $4.20M | $4.20M |
| Owner compensation normalization | +$0.45M | $4.65M |
| Personal expenses | +$0.18M | $4.83M |
| One-time legal and litigation | +$0.30M | $5.13M |
| Non-recurring restructuring | +$0.22M | $5.35M |
| One-time losses | +$0.15M | $5.50M |
| Acquisition transaction costs | +$0.12M | $5.62M |
| Pre-launch product investment | +$0.20M | $5.82M |
| Discontinued operations | +$0.16M | $5.98M |
| Adjusted EBITDA (defensible) | $5.98M | $5.98M |
On a 7.0x multiple, the $1.78M of defensible add-backs is worth roughly $12.5M of enterprise value. If 30 to 50 percent of that adjustment fails review, the operator surrenders $3.7M to $6.2M of value that disciplined preparation would have protected. The bridge is the document that decides which side of that range the deal lands on, which is why it is built before the LOI, not during diligence.
The calculation discipline that defends EBITDA in diligence, from reported earnings through categorization, evidence, the institutional test, and a defended EBITDA bridge.
Start from audited or reviewed earnings before interest, taxes, depreciation, and amortization as the baseline every adjustment builds on.
Separate each proposed add-back into accepted categories such as owner compensation and one-time costs, and flag the categories institutional capital routinely rejects.
Attach support for every add-back, including invoices, board minutes, or contracts, because an undocumented add-back is an unrealized one.
Test each item against the standard of non-recurring, non-operational, and transferable, and remove anything that fails.
Construct the bridge from reported to adjusted EBITDA so the buyer can trace every dollar of normalization.
The EBITDA Quality Calculator scores the bridge across eight dimensions, the same dimensions a buyer and a quality of earnings provider apply. It models the reported-to-defensible gap before the buyer does, so the operator can close it on their own terms. The methodology behind the read sits in the Financial Truth Ladder framework.
Whether each add-back is genuinely non-recurring or repeats across periods.
Whether invoices, contracts, and board minutes support every adjustment.
Whether the normalized earnings transfer to a buyer who runs the business.
How much of the adjustment rests on the owner remaining in the business.
Whether one-time gains are subtracted with the same discipline losses are added.
Whether a small number of large add-backs carry the entire bridge.
Whether any adjustment annualizes revenue without contract support.
Whether the bridge survives the standard a quality of earnings provider applies.
The 8 to 18 percent EBITDA defense gap is closed in preparation, not in diligence. Test the bridge with the QofE Pre-Read, climb the Financial Truth Ladder, and read further across the Operating Library.