We Fired 70% of Our Clients: Here’s the Math

Next Wednesday, Sept 30, 1PM ET. Jake and I are going live for an hour on the Trust Gap and the system we built to close it. Free, no recording, and everyone who stays to the end gets a copy of the book shipped before it’s on sale. Save your seat.

Most agencies won’t tell you who they turn away, because most agencies don’t turn anyone away.

We moved on from 450 clients.

At the time, we had about 650. Those 450 were roughly 70% of the client count and 30% of the revenue, and they generated something close to 90% of the problems. The remaining 200 produced 70% of the revenue and almost none of the headaches.

At a high level, these were startups that needed miracles rather than marketing. Enterprise accounts that treated us like a vendor. Companies with budget and a pulse, so we said yes.

We were subsidizing our own frustration.

Jake will tell you cutting them was a purpose decision, and it was. It was also the best financial decision he’s ever made, and those two things being the same thing is the entire point of this article.

THE PROBLEM: VOLUME LOOKS LIKE WINNING

Here’s how this happens to a marketing department, and it happened to mine.

Volume goes up. Cost per lead goes down. Every chart points the right direction, and you have a green dashboard that says you’re doing the job well.

Then sales tells you they’re grateful for the leads, but they’re spending twice the time to close the same number of deals. Fulfillment corners you and says something critical about the new clients you’ve been signing. Support tickets double. Returns climb 40%.

I hit every target I set that year and watched the business still miss its goals.

And I want to be honest about my first reaction, because it’s the one most marketing leaders have:

Sales can’t close, that’s a sales problem! Fulfillment doesn’t like these clients, that’s an attitude problem! Refunds are up, that’s a product problem!

My metrics were green. So it wasn’t on me… right?

That’s the volume trap, and the reason it’s so hard to escape is that you’re succeeding at what you’re measured on the entire time you’re doing damage. Nobody’s going to stop you. Your numbers are good. 

The better question, and the one that took me too long to ask, is how might I be contributing to the problem?

THE FRAMEWORK: THE $10,000 QUESTION

Jake sat down with me and the numbers. Everything looked reasonable. Then he asked something that broke the ideal customer profile framework:

If we can afford to spend ten thousand dollars to acquire one great customer, and we don’t need high volume to hit our goals, how would you think about this differently?

For a performance marketer, that’s a disorienting question. Driving acquisition cost down is the job. It’s what the platforms optimize for, and it’s what every dashboard is built to celebrate.

But a customer who stays ten years, expands, and refers other businesses to us is worth a multiple of what you’d ever pay to acquire them. A $50 lead that churns in 90 days costs you the acquisition, the onboarding, the support load, and the sales time, and returns almost nothing.

One of those is expensive and it isn’t the $10,000 one. 😅

THE 5 LAYERS OF AN IDEAL CUSTOMER PROFILE

Once fit is the target instead of cost, you need an ideal customer profile framework specific enough to actually pursue. These five layers make it real:

• Demographics are the skeleton. Service-based owner, established, product-market fit, ready to scale. Necessary yet nearly useless on its own.
• Psychographics are the heart. Ours value partnership over vendor relationships and care about their teams.
• Beliefs are the soul. They believe in building something substantial rather than extracting from it. This determines which message lands and which one bounces.
• The transformation is the desire. From managing every marketing detail themselves to having a steward they trust with it.
• The obstacles are the friction. Externally, finding a partner after being burned. Internally, the fear of being burned again. Know these, and you kill the objections before anyone raises them.

Write it as a story you could read aloud to your sales team and have them say “Yes! I know exactly who that is.” Then give that person a name, so anyone can ask “is this a Carrie?” and everyone in the room knows the answer immediately.

Then validate it with two lenses, because your gut can lie. Interview your best customers and use their language rather than yours. Then pull the data: lifetime value against average, retention, referral rate, cost to serve. If the customers you claim to love don’t show up as your highest value and lowest cost to serve, either your definition is wrong or you’re romanticizing people who feel good and don’t perform.

THE PROOF

We made the shift. Turned off the volume campaigns. And for three weeks the dashboard went red every single morning.

On day nine, one of my campaign managers asked whether she could turn the old campaigns back on. She was right that it would have worked. We’d have been green again by Friday.

I didn’t cave, and here’s why:

Before we flipped the switch, Jake and I had agreed on the timeline, the metrics we’d watch, and exactly how long we’d run it before reassessing. That agreement was the only thing standing between me and a panic revert. If I’d made this shift without his buy-in, I’d have turned the campaigns back on by day four and told myself I was being responsible.

The math matters. The strategy matters. But the conversation before you flip the switch is what holds you together when the numbers get ugly—it’s what turns a one-time cut into an actual client retention strategy instead of a mood.

The other thing I’d do differently is who I brought in. I went to sales early, which was right. I told them they’d see fewer leads and those leads should close faster and stay longer, and I asked for 60 days. That bought me time and gave them something to measure.

I did not do that with fulfillment, and that was a mistake. They had pattern recognition I didn’t have. They could spot a bad-fit client on day one from signals that never appeared anywhere in my marketing data.

I nearly forgot customer service entirely, and they turned out to have the clearest picture in the company of who never should have bought. Their data became my strongest proof point when the volume question came.

Every department that touches the customer after you acquire them has information you need and frustration you can relieve. Go to them before the shift, not after.

TAKE THIS AND USE IT MONDAY

🎯 For Marketing Leaders: Pull lifetime value, retention, and cost to serve by customer segment, then compare that list to the customers you’d have named from memory. Where the two disagree is your actual strategy problem. Then audit every campaign, audience, and keyword against one question: was this built to attract the right customer, or built for volume? The split is usually worse than you expect. Before you turn anything off, get agreement with your CEO on the timeline and the metrics you’ll both watch, and tell sales, fulfillment, and support what’s coming and what they should expect.

📊 For CEOs: When lead volume drops 40% on purpose, your reaction in that first month decides whether your marketing leader ever makes a hard call again. Agree on the measurement window before the change, then hold to it. And if you’ve never asked what you could afford to spend to acquire your single best customer, ask. The answer usually reveals that your team has been optimizing against a number nobody ever chose deliberately.

THE TRUTH

Strategy isn’t deciding what to chase. It’s deciding what to walk away from.

That’s the part that sucks, because walking away means saying no to revenue that’s sitting right there and telling your dashboard to go red for a while.

We turn away more than 80% of the companies that approach us. Not as a posture. Because we did this math once and we’re never going back.

We put both sides of this on the Indispensable Marketer show this week. Jake’s solo episode is the 450 clients and the five-layer profile. The conversation episode is the three weeks I watched everything go red and didn’t flip it back.

Both are on YouTube now.

— Chad de Lisle
Head of Marketing, Disruptive Advertising

P.S. On September 30th at 1PM ET, Jake and I are going live for an hour on this exact problem. Free. We’re walking through the Trust Gap and the system we built to close it, and there’s no recording, which is on purpose. Everyone who stays to the end gets a physical copy of the book shipped before it’s on sale anywhere.