How Marketing Agencies Get Valued: What Buyers Pay For

A buyer is not pricing your revenue. A buyer is pricing how much of that revenue still shows up the month after you step back. That is the whole answer to "how to value a marketing agency," and everything below is the list of ways a buyer tests it.

I've reviewed 50+ deals as a buyer, bought into Reach Social Commerce in October 2025, and sold AMZ Advisers to Chief Media this August. I'm not going to hand you a multiple. Ranges for agency deals vary widely with size, mix and structure, and I'd rather show you the levers than quote a number I can't source for your business.

Why "a multiple of revenue" is the wrong question

The default advice for "how much is my agency worth" is a multiple of revenue. It is easy to say and easy to repeat. It is also how founders walk into a first call with a number in their head that no buyer will underwrite.

Revenue is what you billed. Earnings are what is left after real delivery cost. Value is what a buyer believes will keep being earned after the person who built the relationships changes seats. Those are three different numbers, and the gap between the first and the third is where valuations are made or lost.

If you want the definitions, I covered them in SDE vs EBITDA. Short version: larger deals price on EBITDA, which means your owner pay has to be replaced with a market salary for a real replacement before the number is calculated. Founders who skip that step find out in diligence.

Here is the frame I use on every agency package that crosses my desk. Six factors. Each one asks the same thing in a different way: how much of this still shows up when the founder steps back?

The six factors a buyer underwrites

Red flag	What good looks like	Risk
Project and one-off revenue	Retainers with real term and notice	High
One client above a third of revenue	No client above a set share	High
Founder owns every key relationship	Account leads who own their clients	High
Margin hidden behind blended numbers	Margin visible by service line	Medium
Delivery depends on one or two people	Documented process and a bench	Medium
Headline price mostly earnout	Majority of value in cash at close	Medium

The table is the cheat sheet. Here is what sits behind each row.

1. Revenue quality: retainer, project, or hope

A buyer sorts your revenue into buckets. Contracted recurring retainers with a notice period are the best bucket. Month-to-month retainers are next. Project work is the weakest, because it has to be sold again to exist next year. The same $1M in revenue is a different asset depending on the mix.

What a buyer asks for: contract terms, notice periods, and churn history by client for the last 24 months. Churn is the number most founders cannot produce on demand. If you cannot say how many clients left last year and why, the buyer will assume the answer is worse than it is.

2. Client concentration: what happens if the top client leaves

This is the cleanest stress test in the deal. Take your top client out of the revenue line and look at what remains.

Illustration, not a benchmark: if one client is 35% of revenue, a buyer has to price a business that might be 65% of its current size by next quarter. They will either discount the whole business or tie part of the price to that client staying. I wrote about the same pattern on the brand side in how to increase SDE before selling, and agencies feel it more, because a client can leave on short notice.

The fix is slow. Concentration only improves by adding other clients, not by shrinking the big one. Start 12 months out.

3. Founder dependence: who owns the relationships and who runs delivery

Ask two separate questions, because buyers do. Who do clients call when something is wrong? Who actually does the work, reviews it and signs off on it?

If the answer to either is "me," the buyer is partly buying a job. They will price in a transition period, ask you to stay, and tie part of the money to you staying. That is not a punishment. It is a buyer pricing a real risk.

I built AMZ Advisers on purpose to pass the diligence I run on other people's businesses, so account managers carry the relationships and documented process runs delivery. That is the version of founder independence a buyer can verify. I covered what changes after you sign in staying on as CEO, so I won't repeat it here.

4. Margin and the real cost of delivery

Agency margin looks fine until a buyer rebuilds it. They take your P&L and ask what it costs to deliver the work if the founder is replaced by a market-rate hire, if contractors are brought onto proper terms, and if the unpaid hours you put in are paid for.

The delivery mix matters here, especially the offshore and onshore split. Cheap delivery lifts margin, but a buyer also has to underwrite the team, the turnover and the management time that sit behind it. I wrote about both sides in the hidden cost of scaling an agency offshore. The short version: a margin that depends on a delivery structure a buyer cannot verify gets discounted.

Also show margin by service line. A blended number hides the line that loses money, and a buyer will find it.

5. Team depth and key-person risk

Count how many people the agency could lose before a client noticed. If the answer is zero, you have key-person risk, and it is not always the founder. It can be the one strategist who owns three big accounts or the one developer nobody else understands.

A buyer asks who is on the org chart, how long they have been there, what they are paid against market, and which two names would hurt most to lose. Retention agreements for those two names are sometimes part of the deal. The team shape I look for when I buy is in the five patterns I see across the agencies I review.

6. Deal structure: cash at close, earnout, rollover

Two offers with the same headline number can be very different. The question is how much of it is certain, and when.

Component	What it is	What to ask
Cash at close	Paid at signing	Is anything held back or adjusted?
Earnout	Paid if targets are hit later	Who controls the levers that hit the target?
Rollover equity	You keep a stake in the buyer or the combined company	What is it worth, and when can you sell it?
Holdback or escrow	Held against claims for a period	What triggers a claim and how long is the hold?

Earnouts are the one to model carefully. SRS Acquiom's 2026 M&A Earnouts and Claims Study reports that earnouts pay out about 21 cents on the dollar (excluding life sciences). Treat the earnout as a maybe. Price your decision on the certain cash first, then see what the upside is worth to you.

That is why the headline number and the number you keep are different. Rollover is not worthless, but it is a bet on someone else's business. If you want to understand who runs the process and how that changes your bargaining position, read broker vs M&A advisor.

Why two agencies with the same revenue price differently

Put two agencies side by side at the same revenue. One has retainers with notice periods, no client above a modest share, account leads who own the relationships, margin tracked by service line, and a founder who has taken a two-week vacation without a call. The other has project work, one dominant client and a founder on every call.

A buyer pays more for the first one and structures the second one with more conditions. The difference is not size. It is how much of the revenue a buyer believes will still be there in 12 months. That is the reason I say a buyer prices persistence, not revenue.

What a buyer will actually ask for

Expect a request list. The items that slow founders down are always the same:

• Revenue by client by month for 24 months
• Contract copies and notice terms
• Churn history with a reason for each loss
• P&L by service line, with owner pay normalized
• An org chart with tenure and pay
• Documented processes for onboarding, delivery and reporting

If any of those would take you more than a week to produce, that is the work to do first. My due diligence checklist is written from the brand side, but the discipline is the same: clean records, reconciled numbers, nothing the buyer has to discover.

The 8-question self-score

Answer yes or no, honestly, today. Each "no" is a place a buyer will discount you.

1. Is more than half of your revenue on recurring contracts with a written notice period?
2. Could you lose your largest client and still have more than 65% of your revenue?
3. Can you produce 24 months of churn by client, with reasons, in a day?
4. Does someone other than you own each of your top clients' day-to-day relationship?
5. Could the agency deliver for a month without you touching client work?
6. Do you know margin by service line after real delivery cost, and after paying yourself a market salary?
7. Could you lose your two most important employees without a client noticing?
8. Would you take the offer if the earnout paid nothing?

Count your yeses. Seven or eight: you are in a strong position, and the next step is a real conversation. Four to six: you have a clear 12-month plan. Under four: do not take a buyer call yet. Do the earnings-quality work first, and the number you get later will be better for it.

Not ready to sell? This is also your operating scorecard

Those eight questions are not a sales prep list. They are what a well-run agency looks like. Recurring revenue, spread-out clients, a team that delivers without you, visible margin and a bench. You do not need a buyer to benefit from any of them.

If the founder dependence answers are the ones that stung, what it means when a 25-year agency buys your agency covers the readiness signals. If you want the brand-side version of valuation, how to value an e-commerce business is the counterpart to this post.

If you are further along and want a direct buyer with no broker and no committee, here is my process. And if you are a brand, not an agency, and you want the channels run by operators, that is Amazon management and TikTok Shop. I send the deal patterns I see each week in the newsletter.

Frequently asked questions