Most agency sales do not die on price. They die in the tenth week.
I have sat through a fair number of these, and the ones that failed almost never failed on the number.
They came apart about ten weeks in, and almost always for something that was true on the day the process started.
The timing is not a coincidence. The early weeks of a sale are pleasant. There is a teaser document, a management presentation, a number that sounds like a change of life, and a room full of people being complimentary about something you built.
Then confirmatory diligence begins. On a normal timetable that lands six to eight weeks after the first approach, once heads of terms are signed and the buyer starts spending real money on its lawyers and its accountants. The questions change character. They stop being about what the business does and start being about what happens to it when you are not in it. That is roughly week ten, and it is where processes die.
I have sat on both sides of that room. I was a co-owner of Charles Barker when it sold, so I have been the person answering the questions. Later I was chairman of Weber Shandwick's business across EMEA and Asia Pacific, which meant being on the other side while somebody else's life's work was folded into a group. Throughout I am describing owner managed agencies in the UK and Europe, where the founder is still the largest shareholder, selling to trade buyers and to sponsors, rather than listed groups or businesses the founder left years ago. The two experiences taught me opposite halves of the same lesson, and the second half is the uncomfortable one.
Here are the four ways I have watched it happen. Opportunistic buyers exist and I have met them, but that is not what these four are. They are a buyer discovering something late that the seller could have known early.
One. The business turns out to be the founder
Diligence eventually asks three questions in a row. Who owns the client relationship? Who wins the new work? Who do the good people actually follow?
If the answer to all three is you, the buyer has just learned that it is not buying a business. It is buying a person, with a payment plan attached.
You can watch this land in real time. The earn-out gets longer. The lock-in extends. Someone raises the idea of a larger deferred element. Some of that is a buyer repricing a risk it has only just seen, in the only currency available, which is your time. Some of it is opportunism. From where you are sitting the two are almost impossible to tell apart, and the remedy is the same either way.
There is a related question that lands harder than people expect. Who is your successor? Not in principle, by name. If the honest answer is that there is not one, or that there are two people who each privately believe it is them and have never been told otherwise, the buyer now has a second problem stacked on the first.
The tell is usually in how the question gets answered. Almost every founder says the team is very strong, and almost every one of them means it. But when the buyer starts taking references, every road leads back to the same office.
This is not a criticism of anybody. Being the reason a business works is not a flaw. Some of the best agencies I have known were exactly that, and they were better for it. It is simply not the same thing as owning something transferable, and the gap between those two only becomes a number when a buyer is putting a price on it.
The fix exists, and it is genuinely unpleasant, which is why so few people do it in time. It means handing your strongest relationships to somebody else while you are still in the building to watch it go imperfectly. It means accepting that a client will occasionally be a little less delighted than they would have been with you. It costs something in the short term and it feels like a demotion inside your own company. It is also one of the very few changes an owner can make that reliably moves a valuation, and it has to happen two or three years before a process starts, not in week ten.
Two. The price was agreed. The structure never was.
A headline number tends to arrive early, and it feels like the hard part is done. It is not. The hard part is how the number gets paid.
How much is on completion, how much is deferred, over how long, and measured against what. Those four questions are the deal. The headline is only the advertisement for it.
Who is buying changes this more than anything else. Ciesco counted the 2025 media and marketing market splitting 69 per cent strategic buyers to 31 per cent financial sponsors, and the two write earn-outs very differently. A strategic buyer has a group to fold you into and central costs to allocate. A sponsor has to sell you again, and needs the place to work without you.
An earn-out is not a price. It is a job, with conditions, and you will not control all of the conditions. This is where I have seen the most avoidable damage done, and three structures cause most of it.
An earn-out measured on profit, inside a group that will allocate central costs to you. Your revenue can rise while the number that actually pays you falls, and both of those things can be entirely legitimate at the same time.
An earn-out measured on the revenue of named clients, inside a group that has every right to move a client to a different unit for reasons that have nothing to do with you and which nobody could argue with.
And an earn-out running three years, agreed with a buyer whose own strategy will very likely change inside two. You are being paid against a plan that the person who wrote it may not be there to defend.
Most of that is not bad faith. It is what happens when a mechanism gets negotiated late, by people who are tired and want to finish. Earn-out disputes that do turn on bad faith usually turn on exactly these clauses, which is the same argument for settling them early.
What you want, if you can get it, is a measure you can genuinely influence, over the shortest period the buyer will accept, with the treatment of central costs written down rather than assumed. Assumed is the word that does the damage.
Week ten is when the mechanism gets written down properly, and both sides find out they had been using the same words to mean different things.
There is one number worth working out before any of this starts, and most people never do. Not the headline. The amount you would be content with if the earn-out paid you nothing at all. If you could not live with that figure, you are not selling a business. You are being hired, with a deposit.
Three. The client relationships are personal, and the contracts say so
At some point the diligence list asks for the client contracts. This is where a certain kind of agency discovers what it actually has.
Rolling agreements terminable at thirty days. Change of control clauses that let a client walk, or renegotiate, the moment ownership shifts. A top five that represents an uncomfortable share of revenue. Sometimes no signed contract at all with a client of nine years, because the relationship never needed one.
Concentration is the figure most owners underestimate. If one client is a large share of revenue, the buyer is not really valuing your agency. It is valuing that relationship, and taking a private view on how long it survives you. A buyer of any size has a view on it, and on a deal large enough to carry a quality of earnings review it will appear there with a number attached. The asymmetry is not that the buyer hides it. It is that the seller has usually never run the same calculation, so the first time the figure is spoken aloud it is at someone else's estimate.
Then the buyer asks to speak to some of those clients. That request is not a formality and it is not rudeness. It is the buyer pricing what happens the morning after completion.
When the call happens, the buyer is not checking whether the client is happy. It assumes they are, or the process would not have reached this point. It is listening for one thing, which is whether the client talks about the agency or about a person. Clients answer that question completely honestly, and almost none of them realise they have been asked it.
Again, I would defend the way agencies do this. Short contracts and personal trust are not bad practice, they are how the work actually gets done, and a client who stays for nine years without a contract is telling you something a procurement document never could. It is just not the thing a buyer is paying for. A buyer is paying for revenue it can still see in three years. References and pipeline speak to that too, but the contracts are the only part of it a lawyer can hold up.
Four. Nobody agreed the story
I should declare an interest, because this is the section where the problem happens to be the thing I am paid to fix. So take the mechanism rather than my word for it.
There is a fair objection to make first, which is that this is transaction communications, and that is Brunswick and Teneo territory. That is true of a listed company deal. It is not true here. An owner managed agency selling for single digit millions has no financial PR adviser, no investor relations function and nobody whose job this is. That is precisely why the story goes unhandled, and why this failure is the most common of the four.
A process leaks. Not always, but far more often than anyone plans for, and the more people who are involved the sooner it happens. At that point there are three audiences who all need an answer on the same day, and they need different ones. Your staff. Your clients. Your market, which includes every competitor who would happily call both.
The question none of them asks directly, and all of them are actually asking, is why. Why now. And the honest answer is frequently something a founder cannot say out loud: I am tired, or I have taken this as far as I can, or I want the money while the market still looks like this.
The answer to why now has to satisfy three conditions at once. It has to be true, it has to be sayable, and it has to be the same in all three rooms. Those are much harder together than they look. A reason that is true but not sayable gets quietly replaced, under pressure, by one that is sayable but not true, and people can always tell. Not immediately, and not consciously, but they can tell.
What happens next is mechanical, and the mechanism is the only part of this section worth your time. Staff hear it secondhand, which reads as being kept in the dark, because it is. One or two of the good people quietly start taking recruiter calls they would have ignored a month earlier. A client asks their account lead a direct question and gets an answer that has not been agreed with anyone. And then the buyer's diligence finds precisely the attrition risk it was trying to price, except now it is real and it is documented, and the number moves.
The failure is not that the story got out. The failure is that there was not one.
I have been brought in at week ten to write a story that should have been agreed in week one. It is the most expensive briefing anybody ever asks me for, and the honest thing to say in the first meeting is that a good deal of the value it might have protected has already gone. I am supposed to tell you that communications solves this. Often it does. But the problem in front of me on that day was usually created by a decision taken months earlier, by people who were not thinking about communications at all, and my job is to say so rather than to sell a fix that arrives too late to work.
What any of this is worth to you
All four are fixable. By week ten what is left is mitigation and repricing, which is a different and much more expensive exercise.
The useful part is that none of them requires a process, an adviser or a conversation with a buyer to test. You can sit down on any ordinary Tuesday and ask yourself four questions.
Which of my clients would follow a person out of the door, and do I know which person?
What is the number I would be content with if every deferred element paid nothing?
Which of my client relationships exists on paper, and which exists only because of a person?
If this leaked on Friday, what exactly would I say to my staff on Monday morning, and have I ever written it down?
The point of the tenth week is not that it is dangerous. It is that it is late. Everything it exposes was already true, and almost all of it was knowable by anybody willing to look before there was a buyer in the room making it urgent.
So: of those four, which is the one you would rather not answer today?
If you want the detail on the mechanism that does most of the damage, that is the subject of the earnout is where a PR agency deal is won or lost. On what happens once the signature is on the page, see the soft issues decide whether the deal works. And on the work that happens years before any of this, see how I work with agency owners.
or email tim@timsuttonpr.com
Common questions
Why do PR agency sales fall apart in the tenth week?
Because that is roughly when confirmatory diligence begins, once heads of terms are signed and the buyer starts spending real money on lawyers and accountants. Tim Sutton, an M&A adviser to PR and communications agency owners, says the questions change character at that point: they stop being about what the business does and start being about what happens to it when the founder is not in it. Almost everything that kills a process then was already true on the day the process started.
What is confirmatory diligence in an agency sale, and why does it matter so much?
Confirmatory diligence is the detailed verification a buyer runs after heads of terms, typically six to eight weeks after the first approach. It is where client contracts, revenue concentration, management succession and the real ownership of client relationships get tested rather than described. Tim Sutton advises owners to run those tests on themselves years earlier, because by the time a buyer runs them the only options left are mitigation and repricing.
How far ahead should an agency owner prepare for a sale?
Two to three years, not two to three months. The single change that most reliably moves a valuation is making the firm work without the founder, which means handing your strongest client relationships to somebody else while you are still there to watch it go imperfectly. Tim Sutton describes this as genuinely unpleasant and rarely done in time, because it costs something in the short term and feels like a demotion inside your own company.
What should an agency founder say to staff and clients if a sale process leaks?
The same thing, on the same day, in all three rooms: staff, clients and the market. Tim Sutton advises settling the answer to why now before a process starts, and holding it to three conditions at once, that it is true, that it is sayable, and that it does not change depending on who is asking. A reason that is true but not sayable gets quietly replaced under pressure by one that is sayable but not true, and people can always tell.
Tim Sutton is a senior reputational adviser to boards in their hardest moments and counsel to PR-agency principals on both sides of a transaction. timsuttonpr.com · LinkedIn · Privacy · Legal