Add-backs: which ones are real and which ones are not

When a small business is sold, nobody prices it on the profit in the tax accounts. The vendor's broker prepares an adjusted or normalised earnings figure: the accounting profit plus a schedule of add-backs said to represent costs a new owner will not incur. The multiple gets applied to that adjusted number. Which means every dollar of add-back you accept without evidence is a dollar of profit that may not exist, multiplied.

At a three-times multiple, a $50,000 add-back accepted on the vendor's say-so moves the price by $150,000. Reviewing the schedule line by line is the single highest-yield hour in any acquisition, and almost nobody does it with the evidence in hand.

The test for a legitimate add-back

An add-back is legitimate when three things are all true. The expense genuinely sits in the accounts and can be traced to the ledger. It genuinely will not continue under a new arm's-length owner. And nothing replaces it. Most disputed add-backs fail on the second or third test: the cost is real and personal to the vendor, but a new owner picks up a replacement cost the schedule stays quiet about.

The classic example is the owner's wage. Adding back the vendor's $180,000 salary is only honest if the schedule also puts back the market cost of replacing what the vendor actually did. If the vendor ran operations sixty hours a week, the replacement cost is a full-time manager, not zero. The honest adjustment is the difference between the two, not the whole salary.

A worked schedule

Below is an add-back schedule from a recent engagement, anonymised and materially altered. The vendor claimed $214,000 of add-backs against an accounting profit of $310,000, giving an adjusted earnings figure of $524,000. Here is what survived review.

Of $214,000 claimed, $63,000 survived. And the related-party rent line went the other way once the actual lease on offer was read. Adjusted earnings fell from $524,000 to around $364,000. At the multiple being discussed, that review moved the supportable price by roughly half a million dollars. The engagement fee was a rounding error against it.

The patterns that repeat

• Owner wages added back with no replacement cost put in. The most common and most expensive omission.
• Family-member wages added back while the family member's work quietly stays necessary.
• Related-party rent adjusted to whichever number helps the sale, while the post-sale lease tells a different story.
• Recurring costs labelled one-off. Repairs, website rebuilds and restructures that somehow appear three years running.
• Depreciation added back as non-cash on an asset base that genuinely wears out, with no maintenance capex provisioned in its place.
• Personal expenses claimed generously in the tax accounts, then added back generously in the sale accounts. The same dollar cannot be both.

Depreciation and the maintenance capex trap

The depreciation add-back deserves its own section, because it looks the most technical and does the most damage. The pitch is familiar: depreciation is a non-cash accounting entry, so it gets added back to show cash earnings. That is arithmetically true and commercially misleading. If the business runs vehicles, machinery, fit-out or IT that wear out, cash will leave the business to replace them. Just lumpily, rather than smoothly. Adding back depreciation without provisioning a maintenance capex allowance treats replacing essential assets as free.

The honest treatment adds back the accounting depreciation and then deducts a sustainable maintenance capex estimate, built from the age and condition of the actual asset register and the replacement history in the ledger. In asset-heavy businesses the two numbers can be worlds apart. And a vendor whose accounts show nil depreciation in some years and a full add-back in others is presenting whichever version flatters the sale.

Lenders read the same schedule

If the purchase is bank-funded, the add-back schedule is not just a negotiation document. It is a credit document. The lender's assessor rebuilds serviceability from adjusted earnings, and every add-back the assessor rejects reduces the debt the deal can carry. Buyers who accept a soft schedule find this out at the worst possible moment: the price was set on $524,000 of earnings, the credit team recognises $364,000, and the funding gap lands on the buyer's equity three weeks before settlement.

Running the evidence first solves both problems at once. The same normalisation bridge that disciplines the price negotiation becomes the earnings case in the finance submission, with every adjustment already traced and documented the way a credit team expects to read it.

How to handle the schedule as a buyer

Ask for the add-back schedule in writing, with each item traced to the general ledger. Accept nothing described as an estimate. For every add-back tied to a person, ask what replaces their work and at what cost. For every item labelled one-off, look for the same category in each prior year. And test related-party arrangements against the documents that will actually govern them after completion. The lease being offered, not the rent being paid.

This review is included in every level of our due diligence, from the entry-level verification up. Every add-back claimed by the vendor gets evidenced or rejected, with the earnings bridge rebuilt to match. The vendor is entitled to present the business at its best. You are entitled to the arithmetic.