EBITDA vs SDE: The Valuation Metric That Decides What You Actually Pay

Deal Alert AI is reader-supported. We earn commissions from affiliate links at no cost to you.

This post is based on a video from our Deal Alert AI YouTube channel. Watch the original or read the full breakdown below.

I have watched buyers lose deals they wanted because they showed up with the wrong valuation metric. Not because their analysis was sloppy — because they were measuring the wrong thing entirely and then wondering why the seller stopped answering emails.

Here is the scenario I see over and over. A buyer comes from a corporate finance background, or an MBA program, or an investment banking analyst seat. They have spent years living inside EBITDA. Earnings before interest, taxes, depreciation, and amortization. It is the language of leveraged buyouts, of comparable company analysis, of every deal model they have ever built. So when they look at a $400,000 content site listed on Empire Flippers, they open the P&L, calculate EBITDA, apply a multiple, and produce a number that is 30 to 40 percent below what every other buyer in the market is offering.

Then they conclude the market is overheated. It is not. They are just using a metric designed for a completely different type of business.

Why EBITDA Systematically Undervalues Small Online Businesses

EBITDA was built to compare companies that have professional management in place. When you are evaluating a manufacturing business with a CEO, a CFO, a VP of operations, and forty employees, the salaries of those people are genuine operating expenses. They do not go away when the business changes hands. The new owner inherits the payroll. EBITDA correctly treats those costs as real, ongoing, unavoidable.

Online businesses under roughly $5 million in earnings do not work that way. In most cases there is one person — sometimes two — running the entire thing. That person may pay themselves a salary through the business for tax reasons. They may run their phone bill, a portion of their car, their home office, their software subscriptions, and their annual trip to a conference through the company. All of that lands in the expense column. All of that reduces reported net income and EBITDA.

But here is the part that matters: when you buy the business, that owner leaves. Their salary stops. Their personal expenses stop. Their car payment is no longer a line item. The cash that was flowing to them now flows to you, and it does not matter whether it arrived labeled as "salary" or "distributions" or "profit." It is all the same money. EBITDA does not capture that. SDE does.

The Arithmetic That Costs Buyers Deals

Let me put real numbers on this because abstractions do not close deals.

You are looking at an ecommerce brand doing $1.1 million in revenue. Net income on the P&L is $80,000. The owner pays themselves a $60,000 salary that runs through payroll. They also expense a home office at $4,800 a year, a vehicle at $7,200, and roughly $3,000 in conference travel. There is $2,000 of depreciation on some equipment purchased three years ago and $1,500 of interest on a small line of credit.

If you value that business on net income, you are working with $80,000. At a 3x multiple you offer $240,000. The seller laughs and moves on. If you value it on EBITDA, you add back the depreciation and interest and get $83,500. At 3x you offer $250,500. Same outcome. The seller is not laughing anymore, they are annoyed.

Now calculate SDE. Net income of $80,000, plus $60,000 owner salary, plus $4,800 home office, plus $7,200 vehicle, plus $3,000 travel, plus $2,000 depreciation, plus $1,500 interest. SDE is $158,500. At the same 3x multiple, the business is worth $475,500. That is nearly double your EBITDA-based offer, and it is the number the seller has in their head, because it is the number their broker put in the listing.

You were not being disciplined. You were being wrong. The $60,000 salary is money you get to keep the day after closing. Leaving it off your valuation does not protect you — it just removes you from the conversation.

The Complete SDE Formula, Line by Line

Every valuation I run at Deal Alert AI starts with the same reconstruction. You take the P&L the seller provides and you rebuild it from the bottom up. The formula is not complicated, but each component requires judgment.

Start with net income — the actual bottom line after everything. This is your foundation and it should tie exactly to the tax return or the accounting export. If the seller's stated net income does not match their filings, stop and ask why before going further.

Then add back, in order: owner salary or draws for one working owner; owner benefits including health insurance, retirement contributions, and payroll taxes on that salary; personal expenses run through the business such as vehicle, home office, personal phone, personal travel, and family members on payroll who do not actually work; depreciation and amortization as non-cash charges; interest expense since the new owner's capital structure will differ; and one-time non-recurring expenses such as a website rebuild, a legal settlement, or a failed product launch that will not repeat.

The sum is SDE. Note one important constraint: SDE adds back the compensation of one owner. If a husband and wife both work 40 hours a week in the business, only one salary comes back. The second person represents a real labor cost you will have to replace, and it belongs in expenses.

When EBITDA Is Actually the Right Call

I am not arguing that EBITDA is useless. I am arguing it is misapplied at the small end of the market. There are three situations where EBITDA becomes the correct metric, and recognizing them matters as much as knowing the SDE formula.

The first is deal size above roughly $3 to $5 million in earnings. At that scale, businesses almost always have professional management in place. There is an operations lead, a marketing manager, possibly a general manager running day-to-day. The owner may be genuinely passive. When that is true, the management salaries are real recurring costs, and adding them back produces a fantasy number. Value it on EBITDA.

The second is when institutional buyers or lenders are in the room. Private equity firms, family offices, search funds, and SBA lenders all speak EBITDA. If you are competing against those buyers or borrowing from those institutions, you need to be able to present the deal in their language even if you personally model it on SDE. Being able to translate between the two is a competitive advantage, not an inconvenience.

The third is roll-up acquisitions. If you are buying five Amazon FBA brands and consolidating them under one operating team, the individual owner salaries do not disappear — they get replaced by a shared management overhead line. In that structure, SDE overstates what you will actually see. Model the combined entity on EBITDA with a realistic management cost baked in.

How to Stress-Test Every Add-Back Before You Believe It

Sellers and brokers have every incentive to be generous with add-backs. A $10,000 add-back at a 4x multiple is $40,000 of purchase price. That math is not lost on anyone preparing a listing, which is why aggressive add-back schedules are the single most common source of valuation disputes I see.

The test I use is simple and I apply it to every single line: would a new owner who is not this specific person still have to incur this expense to run the business at the same level of performance? If yes, it stays in expenses. If no, it is a legitimate add-back. There is no third category.

Apply it to travel. A seller expensed $18,000 in travel last year. If $12,000 of that was flying to Shenzhen twice to inspect manufacturing and negotiate with suppliers, a new owner will absolutely still need to do that — or pay a sourcing agent to. That stays in expenses. If $6,000 was a family trip to Portugal where the owner answered emails from a hotel lobby, that is an add-back. You need the breakdown, not the total.

Apply it to conferences. An SEO conference where the owner built three of the site's five largest backlink relationships is arguably operational. A general industry event attended for enjoyment and networking that produced no measurable business outcome is personal. Ask for the outcome, not the receipt.

Apply it to home office and vehicle. If the business is fully remote and always was, the home office deduction is a tax convenience, not an operating cost — legitimate add-back. If the business ships products from a garage and that vehicle makes weekly runs to the fulfillment center, the vehicle cost is real and it stays.

Your Add-Back Verification Checklist

Run this every time, in this order. It takes about 45 minutes on a clean deal and it has saved me more money than any negotiation tactic I know. Do not skip steps because the seller seems trustworthy — good people prepare optimistic numbers all the time without any intent to mislead.

1. Tie net income to the tax return. Request the last two years of filings. If the P&L bottom line and the tax return disagree by more than a few percent, get the reconciliation in writing before proceeding.
2. Get the add-back schedule as a separate itemized document. Not a summary line. Every individual add-back, with a dollar amount and a one-sentence justification.
3. Confirm only one owner salary is added back. Ask directly how many people work in the business and how many hours each contributes weekly. Second operators are a real cost.
4. Apply the replacement test to every line. Would a different owner incur this? Write yes or no next to each item. No exceptions, no "probably."
5. Demand documentation for anything over $2,000. Bank statements, credit card exports, or payroll records. Verbal explanations do not count at that size.
6. Check whether "one-time" expenses appear in prior years. Pull three years if available. Recurring one-timers are the most common inflation tactic in the market.
7. Separate travel into business-required and personal. Get the trip-by-trip breakdown with purpose. Do not accept a lump sum.
8. Calculate your own replacement labor cost. If you will not personally do 40 hours a week of the owner's work, subtract the market rate for whoever will. This is the step buyers skip most often.
9. Model EBITDA alongside SDE if the deal is over $1M. Know both numbers and both implied multiples before you send an offer.
10. Recalculate the multiple on your adjusted SDE. The asking price divided by your verified SDE is the real multiple you are paying. Compare that to comparable listings, not the seller's stated multiple.

How This Changes What You See in the Market

Once you internalize the SDE framework, listings start to look different. A business advertised at a "3.2x multiple" might actually be a 4.1x multiple after you strip out unsupportable add-backs. Another advertised at 4.5x might be a genuine 3.8x once you account for a legitimate owner salary the seller conservatively left out.

This is exactly why Deal Alert AI uses SDE as the primary scoring metric across every listing we track on Empire Flippers, Flippa, and Quiet Light. Different brokerages present earnings differently. Some lead with net profit, some with SDE, some with a vaguely defined "owner earnings." Comparing those numbers directly is meaningless. Normalizing everything to a consistent SDE definition is the only way to say whether a $340,000 newsletter is actually a better buy than a $290,000 Shopify store.

The practical effect is that you stop shopping on price and start shopping on multiple. Two businesses at identical asking prices can represent wildly different value depending on how their earnings were constructed. The buyer who does the reconstruction work sees that gap. The buyer who trusts the listing headline does not, and ends up paying a 5x multiple while believing they paid 3.5x.

Bringing the Number to the Negotiation Table

Knowing the right SDE is only half the job. The other half is presenting it in a way that moves a deal forward rather than blowing it up. I have seen buyers send a 400-word email listing every add-back they rejected, and watched sellers disengage entirely within hours.

The approach that works: acknowledge the seller's number, then present your reconstruction as a working document rather than a verdict. "I have the business at $142,000 SDE rather than the $168,000 in the listing. The delta is mostly the second operator salary and the travel breakdown. Can we walk through those two items?" That opens a conversation. Sellers who have real justification will provide it, and sometimes they are right and you adjust upward. Sellers who cannot justify a number will usually concede quietly rather than argue.

Price your offer off your verified SDE, not theirs, and be explicit about the multiple you are applying. When a seller sees "3.5x on $142,000" they can evaluate the logic even if they disagree with the inputs. When they see a raw dollar figure with no reasoning attached, they assume you are lowballing and stop engaging. The math is your credibility.

Every deal I have closed at a good price came from doing this reconstruction before the first offer, not after. The buyers who win in this market are not the ones with the most capital — they are the ones who understand exactly what they are buying and can prove it in a spreadsheet. If you want the SDE-normalized version of every listing hitting the market this week, that is what we built Deal Alert AI to deliver.

By Sophal Lanh, Founder of Deal Alert AI