A CEO I worked with had been with the same agency for almost two years. The reports came in every month. Charts trending in the right direction. Traffic up. Cost per click down. He would review the deck on a call, nod, and feel like things were moving.
Then someone on his team asked a simple question: how much of this is turning into customers?
Nobody had a clean answer. Not the marketing manager. Not the agency. Not the CEO.
When we pulled the data, close rate by source traced all the way through to signed contracts, the picture was uncomfortable.
The channel absorbing the most budget was the worst-performing one in the mix. The channel producing the most revenue had no deliberate marketing behind it at all. No referral program. No systematic outreach to the partners sending business their way. Nothing.
This is not an unusual story. I have walked into this situation more times than I can count, across industries, company sizes, and agency relationships of every kind. The data almost always tells a version of the same story. What differs is how uncomfortable the CEO is willing to get when it does.
I want to be clear about something before going further. The instinct when you recognize this pattern is to blame the agency. That is usually the wrong place to land.
Most agencies are talented and genuinely want to do good work. They also arrive with a playbook. A set of channels they know how to run, metrics they know how to report, and a dashboard template that gets applied to most clients with modest customization. That is not cynicism on their part. It is efficiency. The problem is that it treats your business as roughly similar to the last one, which is only true up to a point.
How agencies define success when you don't
Agencies measure what they can access and report what makes the engagement look productive. When a client hasn't defined what success actually means, in terms of the right customers acquired at a defensible cost, the agency fills that gap with the metrics they can control: impressions, clicks, traffic, cost per lead.
Those numbers can trend in the right direction for months while the underlying business problem gets worse. It is not usually intentional. It is what happens when activity becomes the proxy for outcomes, and nobody with enough authority challenges that substitution.
Clients often encourage it without realizing it. They want to see something for their money. The agency provides something visible. The question of whether that visible activity is pointing at the right buyers, the right margins, or the right definition of a good customer tends not to come up until the frustration is significant enough that someone finally asks.
Here is a more specific version of what this looks like in practice.
I worked with a roofing contractor that had been running Google Ads for about a year before I came in. They were paying $8,000 a month to an agency primarily to manage and monitor campaigns that had been set up months earlier and weren't being materially changed. On top of that, they were spending $15,000 to $20,000 a month on the ads themselves. The ads were generic. The kind of copy that could belong to any roofing company in any market: fast estimates, licensed and insured, financing available. Nothing wrong with any of it. Nothing specific to them, either.
The firm had a real specialty. Metal roofing, architectural products, historical preservation work. The kind of projects that require a different buyer, a longer conversation, and a significantly higher price point. A custom metal roof on a commercial or historic building runs $100,000 or more. An asphalt shingle replacement on an average house runs around $12,000. Metal is three to four times the price, and for crews that know what they are doing, the margins are significantly better.
The generic ads were pulling the wrong buyers entirely. About three quarters of the leads who submitted a form ghosted the company when they called back. Of those who did respond and get an on-site estimate, 10 to 15 percent closed. Nearly all of them were price-shopping for commodity work. The cost to acquire each of those customers was around $3,000. On a $12,000 job with real labor and material costs, that math is punishing. On a $100,000 specialty project, the same acquisition cost is trivial.
The monthly report had showed cost per lead trending down. Which was accurate. It just was not the question that mattered.
When I laid this out, both the CEO and the sales director understood it immediately. They also did not want to cut the Google Ads budget. The fear was that the pipeline would dry up entirely if they stopped, even a pipeline full of the wrong work. That is a pattern I have seen more than once: a company that knows a channel is not producing the right results keeps running it because stopping feels more dangerous than continuing. Activity, even inefficient activity, feels like control. Turning it off feels like a leap.
This firm eventually parted ways with the agency and brought the ad management in house. Other companies I have worked with in similar situations kept their agency and saw a real turnaround once the targeting and brief were reset. The relationship was not the problem. The strategy behind it was.
Either way, my push was the same: stop treating paid search as the whole answer. The CEO had a name for it. He called it "Lord Google," meaning the firm's fortunes rose and fell based on whether the algorithm felt generous that week. That is a fragile place to run a business from. Paid search is reactive by nature. It reaches buyers who are already looking, often already comparing prices, and it commoditizes quickly when every competitor is bidding on the same terms. To reach the higher-value buyer, the one considering a $100,000 specialty project rather than a price comparison on asphalt shingles, you need channels that build credibility before the search happens. A portfolio that tells the right story. Case studies written for the buyer you actually want. Conversations that start before a project is even formally scoped. The companies that got out of the Lord Google trap were the ones that stopped waiting for the right lead to show up and started building the conditions that make the right buyer think of them first.
The question that cuts through most of these situations is simpler than it sounds: what does a good customer actually cost you, all the way through, and what are they worth over the life of the relationship?
Two customers who look identical in a leads report can have completely different economics. One closes quickly, pays on time, refers two more, and asks for you by name on the next project. The other negotiates the price down, pays in 90 days, and you never hear from them again. A marketing program that generates more of the second type while suppressing the first is making the business worse, not better, regardless of what any activity metric says.
Most growth-stage companies have never traced that picture. They know their revenue. They may know their margins. They do not know which marketing channels are producing customers who actually build the business versus customers who consume it.
Getting that picture is the starting point for everything else. What channels to invest in. What the agency should actually be optimizing for. What a good lead looks like before it reaches the sales team. What success means in a monthly report.
When I rewrote the brief for that first CEO, based on what the data was showing, the agency stepped up. Same people, same fee, different output. Because they finally knew what they were being asked to produce. The brief is a leadership decision, and it requires someone with enough standing and enough context to write it honestly. That is rarely the agency. It is rarely the junior marketer in the seat. And it is usually not something a CEO has time to build from scratch while running the business.
The reports had looked good for two years because nobody had told the agency what good actually meant. Once someone did, the picture changed quickly.
Rob Higley runs Reveal, a fractional CMO practice based in Indianapolis. He works with B2B companies where marketing is producing activity but not the right results. rob@revealcmo.com · 317-430-3769