When a grocery file should talk SDE versus EBITDA

Recast the P&L before anyone names a multiple. Mark the SDE valuation worksheet, put owner hours on a wage line, and book a confidential intro with three years of statements. Grocery Store Business Broker — HedgeStone Business Advisors, led by Jason Taken — will tell you whether this file should talk SDE or EBITDA before you argue a price. Jason also reads grocery paper through Jaken Finance Group, which is how a DSCR conversation stays honest when the “earnings” number is still a job.

This comparison is not a branding contest. It is a labor question. If the seller is the meat cutter, the closer, and the bookkeeper, you are pricing a living. If a hired general manager already sits on the payroll and the owner only visits, you are pricing an investment. Mixing those stories is how a neighborhood box gets an EBITDA speech and a two-store file gets an SDE add-back it has not earned.

Owner pay is the first line that decides the definition

SDE is seller’s discretionary earnings: the cash a working owner can take after you put the store on a buyer’s cost structure. You add back the owner’s compensation, a personal auto, and a one-time legal bill that will not repeat. You do not add back shrink that shows up every year, the Sunday overtime that keeps the night crew, or a “consulting” check to a relative who will still be needed on Tuesday.

Read how add-backs are treated before you build a stack of favors. Add-backs that repeat every year are the business. Recasting a grocery P&L is the worksheet. Recast financials are what a financed buyer will underwrite, not the tax return’s bottom line and not a verbal “I take about two hundred.”

A one-store owner-operator almost always belongs on SDE. The buyer is buying a job plus a going concern. If you price that file on EBITDA, you have left the owner’s market wage inside the earnings as if a stranger will work for free. The first lender who models a cutter and a closer will collapse the number. See anatomy of a grocery P&L at the department level, then evaluating a store’s P&L so meat and produce are not hiding inside a “healthy” storewide blend.

Family hours are the quiet version of the same mistake. A spouse who runs the front end without a W-2 is not free labor. A son who closes six nights is not an add-back. Put a wage on those hours. If the recast only works because those people disappear, you do not have SDE. You have a story.

Hired management is when EBITDA stops being a costume

EBITDA earns its keep when the store already pays for the labor a buyer will not personally supply. A hired general manager, a cut-room lead, and a bookkeeper on the payroll mean the owner’s “salary” is already closer to a return on capital. Adding that salary back a second time double-counts. The single-store versus small-chain comparison is the operating version of this test: one neighborhood box with the seller on the floor is SDE; two or more stores with a GM is usually EBITDA.

Do not promote a file to EBITDA because the multiple “sounds more institutional.” Institutional buyers still subtract the manager you have not hired. If the seller claims to be “absentee” but works sixty hours in produce, you are looking at SDE with a marketing caption. Walk the schedule. Ask who locks the door. Ask who calls the wholesaler when a load is short.

Preparing three years of financials matters more on an EBITDA file because a buyer is underwriting a management team, not a person. One clean year with a new GM and two prior years of owner chaos is not a three-year EBITDA story. It is a transition. Price the transition, or stay on SDE until the hired bench has a history.

SBA 7(a) can finance a change of ownership; the published maximum is $5 million. The program does not care which label you prefer. It cares whether cash after an honest labor recast covers debt service. Unreported cash does not survive that test. If the only way the EBITDA works is a drawer that never hits the bank, you do not have a loan file. You have a rumor. Read selling a store with unreported cash before you try to dress that drawer up as earnings.

National sold multiples attach after the definition, not instead of it

BizBuySell grocery and supermarket businesses sold in 2025 printed a 3.38 average earnings multiple, a 2.66 five-year average for 2021–2025, a 1.60–3.15 interquartile range on that sold sample, and a 0.43 average revenue multiple in 2025. Those are national listing-site sold figures. They are not a chain-M&A print and they are not this box. The multiples report keeps the sourced sample in one place. The valuation hub is the method. How to value a grocery store is the longer walk.

Apply that band to the earnings definition you actually recast. A 3.38× speech on a sloppy SDE is not diligence. A 3.38× speech on EBITDA that still needs a $90,000 manager is a different store than the one in the CIM. Revenue multiples hide rent and labor. Use 0.43 only as a cross-check, then return to the recast. The valuation calculator will not save a file that picked the wrong denominator.

FMI reports $1 trillion in 2025 supermarket sales and 45,575 U.S. supermarkets in 2024. FMI also publishes $668,377 average weekly sales per supermarket and $19.59 weekly sales per selling square foot in 2025. Those figures describe the Speaks supermarket channel. They are not a reason to write an offer, and they are not an earnings definition. FMI’s 2.1 percent average net profit for food retailers in 2025 is a food-retailer net sample. It is a reminder that grocery is thin. It is not your SDE and it is not your EBITDA. Do not quote 2.1 percent as “the multiple” or as “what owners take home.”

Teaching example, labeled illustrative: recast SDE of $180,000 on a working-owner box, times the 1.60×–3.15× interquartile band, is about $288,000 to $567,000 for the going concern before inventory at cost. The same store forced onto EBITDA by leaving the owner wage inside the number is not a better asset. It is a smaller cash-flow story wearing a larger label. Subtract holes — short lease, missing manager, department losses — before you sit near 3.38.

Inventory, rent, and labor still sit outside the label

Neither SDE nor EBITDA prices the stockroom. Inventory at cost is a separate line, counted at close. Retail tags are not cost. Write the unsellable rules before 2 a.m. See how inventory is counted at closing and closing-day count procedures. Turns are not one number. Center store and the perimeter do not share a clock. A pretty SDE that ignores a stale grocery aisle is a fight on inventory night.

Occupancy is the other number that survives any earnings caption. A below-market related-party rent juiced the seller’s earnings and will reverse on day one. Recast to market rent before you multiply. Run the rent-as-percent-of-sales checker on the actual lease. CAM true-ups can erase a year of “extra” SDE. Read lease diligence so assignment, remaining term, and exclusive use are not footnotes under the multiple.

Labor you will hire is the third survivor. State minimum wage is a model input. Owner “salary” that is really unpaid family hours reverses on Tuesday. If you are the buyer, model the cutter, the produce lead, and the closer. If you are the seller, show those costs now so the recast and the LOI match. How to read a CIM is the annotation method: circle the earnings definition, then circle the labor that is missing.

Scan data is how you prove the week that sits under either definition. A weekly-sales spreadsheet without department mix is a rumor. Read verifying sales with scan data and align sales-tax filings, EBT deposits, and wholesaler statements. A gap is mix, downtime, or fiction. Red flags that should kill a grocery deal exist so you stop multiplying a story.

A wholesaler over-advance is debt. It is not an add-back on SDE and it is not “other income” on EBITDA. If the buyer cannot assume the guarantee, price the payoff or walk. Banner program fees belong in whichever earnings definition you keep if the buyer will keep the fascia. Buying a franchise or banner store versus independent is the consent problem; the banner contract is the document. A CIM that captions “adjusted EBITDA” on a one-store owner-operator is a tell. Circle the caption, then ask who is standing in the cut room after close.

Put the earnings definition in the LOI, not in a handshake

The LOI terms that matter for sellers and the LOI terms that matter for buyers should name the earnings definition the price is standing on. “Three times earnings” is not a formula. Three times recast SDE, inventory at cost extra, with a working-capital peg, is a formula. Three times EBITDA on a two-store file that already pays a GM is a different formula. If the parties do not write the denominator, diligence becomes a relitigation of add-backs.

Comparing offers means cash at close, note terms, inventory treatment, and license risk — not which buyer used the more fashionable earnings word. A higher headline on sloppy EBITDA can be a worse check than a tighter SDE offer with inventory rules and a transition period the seller will actually work. Escrow, holdbacks, and earn-outs exist because grocery has shrink and a week the buyer can break. Earn-outs on grocery are hard because the buyer controls mix after Monday. Do not use an earn-out to paper over a fight about SDE versus EBITDA.

Working with attorneys and CPAs is how the recast survives contact with tax returns. Counsel who have never closed a grocery file will argue about labels. Bring them the worksheet and the department P&L. The deal-structure glossary is the shared vocabulary. The seller hub and the buyer hub are the two sides of the same recast.

If the file is a working-owner neighborhood box, stay on SDE and recast like a buyer who will stand in the cut room. If the file already pays for management and the owner is capital, stay on EBITDA and stop adding the owner’s wage back. If you cannot tell which story you are in, you are not ready to quote a multiple. You are ready to recast.

When the worksheet is marked and you can say the earnings definition out loud, schedule the confidential intro. Bring three years of P&Ls, the payroll register, and the lease. That is enough to decide whether this grocery file talks SDE or EBITDA — and to stop arguing a number that does not have a denominator.

How a buyer would price When a grocery file should talk SDE versus EBITDA

These figures are a teaching walk-through for When a grocery file should talk SDE versus EBITDA, not a quote and not a sold comp. Order of operations: weekly sales, recast SDE, inventory at cost, then a multiple band.

Input or output	Illustrative figure
Weekly sales (illustrative)	$92,000
Annual sales	$4,784,000
Recast SDE	$172,224 (3.6% of sales)
Inventory at cost	$341,714 (~14 turns)
Occupancy check	$148,304 / year (3.1% of sales)
Low going-concern	$686,162 at 2.00x SDE + inventory
High going-concern	$806,719 at 2.70x SDE + inventory
Sketch DSCR on a note	1.38

Inventory is cash at close, not a multiple. Confusing the two is how sellers over-ask and buyers under-cash the wire.

Book the intro call with your side of the table ready: seller files or buyer capital. The calendar is the only next step.