Agency owners believe they are good at this. Databox asked dozens of marketing agencies how often client expectations become an issue that has to be addressed, and offered “never” as one of the available answers. Not one agency picked it. In the same research, around 85% of those agencies said they do a good job of setting expectations during the sales process.
Read those two findings next to each other and you have the honest state of setting client expectations in this industry. Almost every agency in that survey believed it handled this well, and not one of them was willing to say the problems never arrive anyway.
I have been running agencies for more than twenty years. That gap between what we believe and what happens is not a training problem, and I know that because I have landed on the wrong side of it myself. Not often, and less often now than I used to. But the times it happened, it was never because I did not know what setting client expectations required. I knew. I just did not say it.
Why does setting client expectations fail even when everyone knows how?
The standard advice is not wrong. Define the scope. Agree on what success looks like. Set a communication cadence. Report on business outcomes rather than vanity metrics. The Forbes Agency Council has seventeen agency executives saying versions of this, and I would sign my name to most of it.
We have systems built around all of it, and setting client expectations sits inside our sales process rather than bolted on afterward. We walk the SOW with everybody who touches the agency-client relationship so that everyone understands what is in scope and what is not. When a key stakeholder changes six months in, we make time to repeat that whole process with the new person. We overwhelm clients with communication in the early days on purpose, because we want them seeing action and traction the moment the contract is signed.
So the process advice is covered. And yet the survey says the problems arrive anyway, at agencies that would describe setting client expectations as a solved part of their operation.
Here is what I think is going on. Setting client expectations is not one skill. It is two. The first is knowing what needs to be said. The second is saying it to someone’s face, early, when saying it might cost you the deal. Most of the published advice covers the first skill. The failures I can actually account for in my own business all happened at the second.
What I actually know on day one
When a relationship does fizzle out on a single misaligned expectation, I can usually trace it back and find that I knew in my gut from day one it was going to be an issue.
Not suspected. Knew.
So the interesting question is not what I missed. Nothing was missed. The question is what I did with the information, and there are only two answers I have ever been able to give myself honestly. Either I was desperate to add another account because of whatever was happening operationally at the time, or I was busy and distracted and did not want to give it my full attention.
Underneath both of those is something simpler that took me a long time to admit. I avoid conflict. I do not like having hard conversations if I do not have to, and that is a personal trait I have been working on for years. It is a strange thing to say out loud as an agency owner, because the job rewards people who are comfortable being direct. I am comfortable being direct once the relationship exists. It is the conversation that might end the relationship before it starts that I have historically flinched at.
Two engagements will show you what that flinch costs.
The account I should have declined
A former client referred a lead to us. Both were motivational speakers in the B2B space. The lead had heard glowing things about the campaigns we were running and wanted us to replicate them for him. He was urgent, he wanted it done as fast as possible, and he was on a shoestring budget.
He wanted to manage it himself, so we quoted a flat rate to build everything out and hand it over. Reasonable enough on paper.
Within a few conversations it was clear he did not know what he was doing and was going to make a mess of the campaigns. His landing pages were not finished either. His website was fine as a website, but it was not usable for advertising. We built good campaigns anyway. The ad copy aligned with the search terms, the audiences, and the landing page copy he gave us. Then the finished pages arrived and we knew immediately they were not going to convert.
We gave him a landing page analysis even though it was out of scope, with everything he could do to fix it. Rebuilding took him so long that he lost patience and said let’s just launch. We should have pushed back. We did not. We launched, handed it over, and I gave him a training session on what to do daily, weekly and monthly.
A month later he came back saying he had no real leads. We got on a screen share and I walked him through the account. The front-end metrics were solid. High-intent search terms matching closely to his keywords. CPCs slightly high but within range, and probably fixable with page speed work. Then I showed him the analytics: users landing, not engaging, bouncing. The same recommendations we had made a month earlier, now with data attached.
He eventually redesigned the pages, relaunched, and the campaigns performed better. He still was not capable of managing them. He would come back with things like the ads all disappearing, which turned out to be a filtered view in the interface. I showed him the ad editor so he could make bulk changes. I kept telling him it is cheaper to have us do this properly the first time than for him to do it wrong and pay me to fix it, that he is paying twice, and that his time as a founder and front man is not well spent inside an ad account.
I lost a lot of money on that account. He still texts me asking for fixes. He pays late, even though I charge him a fair hourly rate and work quickly to keep the cost down.
That account had red flags coming into it. I should have declined it gracefully and introduced him to someone lower priced who could genuinely help him. Or, better, I should have told him plainly that he was not capable of managing this himself and required a retainer.
Either sentence would have been a short conversation. Setting client expectations with him was never the difficult part. Saying the sentence was.
The discount I should not have given
The second one is subtler, because it happened with a client we already had.
They started as paid media only. Once we found a pile of issues in their CRM and sales process, we upsold them into a RevOps engagement. They beat me up on pricing and I reduced my rate.
I knew better. Reducing the rate means saying out loud what the lower number buys and what it stops buying, and I said none of it. I already knew exactly how they operated and how demanding they were, because we had been working with them. I wanted the additional revenue and I wanted it settled quickly because I had other things to get to, so I agreed.
That account became my single biggest time suck. Not my team’s, which is worth noticing. Mine personally.
What eventually fixed it was not a framework. In meetings with the CEO I started dropping hints about margins going sideways and about losing money on the work. He valued what we were contributing enough, and had come to treat me enough like an internal member of his team, that he opened the door to renegotiation ahead of renewal. We right-sized the engagement.
That worked. It also took time and depended on him valuing the work enough to open the door. Handling the pricing conversation properly would have taken one uncomfortable sentence at the moment he pushed back.
He is now launching a third brand inside the parent organization and pulling me into strategy work with other executives. I have given him a few things. I can see history repeating itself, and this time I am going to nip it in the bud and have the conversation about scope before we repaper the engagement rather than after.
Why is the tough conversation cheaper now than later?
Twenty-plus years in, this is the thing I am still learning. The tough conversation in the moment is easier than the much tougher conversation later, because the problem accelerates and exaggerates while you avoid it.
Setting client expectations early is cheap in a way that is very hard to feel at the time. Look at what each of those two conversations actually was.
With the speaker, the early version was one sentence during the sales conversation: you are not set up to run this yourself, so either take a referral to someone cheaper or sign a retainer. The late version was a long tail of late-paid hourly fixes, screen shares explaining interface filters, and a relationship that still generates texts I do not want.
With the RevOps client, the early version was one sentence about pricing. The late version was a stretch of margin bleeding, followed by a renegotiation that only worked because the client was willing to reopen the terms.
The early conversation costs you the risk of losing a deal you are not sure you want. The late one costs you money, attention, and the option to walk away from the engagement entirely without it being messy. The trade looks obvious written down. It does not feel obvious at the moment you have to speak, which is exactly why so much advice about setting client expectations does not stick.
What does setting client expectations look like when I get it right?
One of my principles in life is radical candor, and the strongest and longest relationships I have are built on it. I am unapologetically honest with prospects from day one when their goals are unreasonable, and I push my team to set realistic expectations in everything they do.
Everyone wants results yesterday. That is human nature and I do not hold it against anybody. But if you stretch the truth or exaggerate to keep a prospect warm, it comes back on you later with interest. For the ones who genuinely demand overnight results, we pass and introduce them to another agency. That is not a loss. Telling someone what they want to hear when you know it is not realistic is how you buy yourself an eventual churn.
The technical version of this matters just as much. Performance takes time to materialize, and the most important early work is conversion alignment. We have to be in lock step with the client on which metrics they actually care about and what the source of truth for that data is. What ad pixels measure and optimize toward is often very different from what a client sees in their CRM. Setting client expectations around performance means defining that early, with a tracking specialist making sure what we report matches what the client expects to see. Skip it and you will have a client who believes you failed while your dashboard says you succeeded.
None of that is a secret. Most agencies could write the same paragraph. Setting client expectations well is less about the paragraph and more about whether you will say the uncomfortable half of it while the prospect still has the option to walk.
The part nobody puts in the case study
Our relationships rarely die from oversight or negligence, because that is not how we operate. We do what we say we are going to do. When things do fizzle, it is misalignment, which is why we are so repetitive about communication. People do not always hear what you are saying. They hear what they want to hear, so you repeat the important parts until they are understood and acknowledged.
This does not happen to me often anymore. It happens a lot to more junior founders and agency executives, and I suspect it happens for the same reason it used to happen to me. Not because they do not know what setting client expectations requires. Because the sentence is uncomfortable, the revenue is right there, and it is very easy to tell yourself you will address it once the relationship is established.
You will not. Setting client expectations retroactively is a different and far more expensive exercise.
If you are reading this and there is an account you already have doubts about, you almost certainly know which one it is. You knew on day one. The only thing left to decide is whether you have that conversation this week or next quarter, and how much the difference is worth to you.