How to judge a business's numbers before you make an offer

Listings sell a story. Tax returns sell what the owner was willing to report to the IRS. Your job before an offer is to decide which number you are actually buying, and whether that number still exists after you remove the seller's lifestyle, one-time cleanups and optimistic add-backs.

This page is a field guide for that work: SDE versus EBITDA, what counts as an add-back, why tax returns beat a polished P&L, how customer concentration changes risk, and the red flags that should slow you down. Definitions lean on the International Business Brokers Association glossary and IBBA Business Brokerage Standards. Verification habits track what SBA lenders already do under SOP 50 10 when they finance a purchase.

What to request in the first document drop

You can learn a lot before a formal quality-of-earnings engagement. Ask, under NDA, for:

• Three years of business tax returns (or Schedule C) and matching year-end financials
• Monthly P&Ls and balance sheets for at least the last twelve months
• A written add-back schedule tied to ledger accounts
• Revenue by customer or customer tier, plus any major contracts
• Debt schedule, leases and a list of related-party transactions

If the seller will only send a one-page summary, treat that as a signal about how diligence will go. Serious sellers who want a clean close usually expect these asks.

SDE and EBITDA answer different questions

IBBA defines Discretionary Earnings (also called Seller's Discretionary Earnings, or SDE) as earnings of the business prior to income taxes; non-operating income and expenses; nonrecurring income and expenses; depreciation and amortization; interest; and one owner's entire compensation, including benefits and personal expenses paid by the business. In plain language, SDE asks: what total financial benefit did one working owner take out of this company?

EBITDA, in the same glossary, is earnings before interest, taxes, depreciation and amortization. Adjusted EBITDA goes further: it starts from that idea, strips non-operating and nonrecurring items, and then replaces the owner's pay with a market-rate salary for the work that still has to be done. EBITDA-style metrics fit businesses that already run (or will run) with hired management. SDE fits owner-operated shops where the buyer expects to step into the seller's seat.

If a broker quotes SDE and you mentally treat it like EBITDA, you will overstate what is left after you pay yourself or a manager. For price context after you trust the earnings, see what a business making about $100k really costs to buy. When the earnings still stand, small business due diligence is the fuller pass on tax, legal, customers, employees and the lease.

Add-backs: IBBA's definition, and the three tests that survive diligence

IBBA defines add-backs as all or a portion of expenses added back to net income to get closer to the economic earnings the business actually produced. That definition is wide on purpose. Sellers use it; buyers and lenders narrow it.

Run every line through three questions:

• Is it documented? Payroll registers, invoices, bank clears, or a clear ledger entry. Undocumented "owner perks" are wishes.
• Does it stop under new ownership? A family cell plan might. A van the routes still need will not.
• Is it truly nonrecurring? A settled lawsuit can be. "One-time" repairs that appear every spring are just maintenance with better branding.

SBA's own credit analysis language in SOP 50 10 tells lenders to justify additions and subtractions to cash flow, including nonrecurring items, owner's draw, rent and related adjustments, and to calculate EBITDA as part of repayment analysis. If a federal lender has to defend the bridge from reported profit to underwritable cash flow, you should too before you bid.

Tax returns versus the P&L the seller prefers

A P&L can be recast in an afternoon. A filed return is what went to the IRS. SOP 50 10 requires lenders, for change-of-ownership files, to verify seller financial data and, outside certain Express products, to obtain tax return transcripts and reconcile them to the financials. Valuation work for a purchase also has to be checked against the seller's IRS transcripts.

That is your cue as a buyer. Ask for multi-year business returns (or Schedule C for a sole prop), interim financials, and a written add-back schedule. Then tie revenue and major expenses to bank deposits where you can. IRS Publication 583 is written for owners who need to keep records that support what they report; the same documents are what let you reconstruct the story when you are on the other side of the table.

Large gaps between the marketing P&L and the return are not automatically fraud. They are a reason to slow down, ask for the bridge, and decide which figure you will pay a multiple on.

Customer concentration and other durability tests

Earnings quality is not only about add-backs. Ask who pays the bills:

• Customer concentration. If a handful of accounts are most of revenue, price the loss of the largest one. Read contracts, renewal dates and switching costs.
• Related-party costs. Rent paid to the seller's LLC, family payroll, or below-market (or above-market) vendor deals should be restated to terms you will actually face.
• Working capital shape. Rising receivables, stale inventory or a payables cliff can turn a pretty EBITDA into a cash trap the week after closing.
• Owner dependence. If the seller is the sales team, SDE includes labor you may not replace at the same cost. That is closer to buying a job than a transferable business; see buying a business vs. buying a job.

Red flags worth an early pause

• Add-backs that lack invoices, payroll support or a clear "why it ends."
• "One-time" expenses that repeat across years on the monthly P&Ls.
• Revenue spikes in the last quarter before the listing, without a matching backlog or contract explanation.
• Refusal to share tax returns, bank statements or customer lists under a normal NDA.
• Books that cannot be reconciled to deposits, or a cash business with no corroborating records.
• For larger deals, remember SOP 50 10's Quality of Earnings trigger: Initial Acquisition or Business Expansion purchases at or above a $3 million purchase price require an independent QoE for the lender, including a cash proof. Even below that size, acting like you will need a cash proof keeps you honest.

Build a buyer-case earnings figure

Start from the most recent full-year tax return profit. Add back only what clears the three tests above. Subtract a realistic market salary for the work you will not do yourself. Subtract any related-party rent or vendor gap you will have to normalize. What remains is closer to the cash flow you can underwrite than the seller's headline SDE.

Then look at two years, not one. A single strong year after a weak one needs an explanation you can believe. Weighted averages exist in broker practice for a reason: IBBA even defines weighted average discretionary earnings as a way to handle year-to-year swings. You do not need their formula to refuse to price a spike.

If financing is part of the plan, remember that SBA change-of-ownership underwriting for an Initial Acquisition looks for historical debt service coverage of 1.25:1 under SOP 50 10 Appendix 15. Your personal offer math and the lender's coverage test should not live in separate universes.

A practical order before you write a number

Collect returns and monthly financials. Rebuild earnings two ways: seller case and buyer case (with replacement labor and without soft add-backs). Check concentration and related-party items. Only then decide an offer price and how much of your cash is equity versus reserves. If you expect to finance with 7(a), the same documents feed the lender file described in our guide to SBA 7(a) requirements when you buy a business.

You do not need a full Quality of Earnings report to walk away from a bad CIM. You need the discipline to treat adjusted earnings as a claim until the source documents agree.

Common questions