I've now sat on both sides of the agency-client breakup. As founder and as buyer. As the firing party and as the fired one. The data set isn't huge — maybe 40 transitions across two decades — but a clear pattern emerges, and I want to put it on paper.

Here are the four reasons clients actually fire agencies, ranked from most common to least. The list might surprise you. Performance isn't number one.

4. Performance (yes, last)

Let's get this out of the way. Pure performance failures — "the campaigns aren't generating leads, full stop" — account for maybe 15-20% of the firings I've seen. They happen. They're brutal. They're often the agency's fault. But they're not the dominant reason agencies get fired.

And here's the wrinkle: clients who fire agencies for performance failure usually do it within the first 90 days. After that, even underperforming agencies get more grace than they deserve, because the client has invested too much in the relationship to start over. Past month four, performance becomes secondary to the issues higher on this list.

3. Communication decay

Around month four to six, communication subtly degrades. The senior strategist stops showing up to weekly calls. The Slack response time creeps from 30 minutes to half a day. The monthly report goes from 8 pages of insight to 4 pages of metrics. Nobody calls a meeting to say "things are getting worse" — but the buyer feels it.

This is the most underrated reason agencies lose accounts. The work might still be fine. The performance might still be holding. But the client is no longer feeling the agency's energy. The relationship is starving without anyone naming it.

The fix is dumb-simple: publish a Monday morning written update every week, even when there's nothing exciting to report. Five bullets. What shipped. What's testing. What we're worried about. The boring weekly update is the load-bearing pillar of a healthy agency relationship. Skip it for two weeks and the relationship starts cracking.

2. Mismatched expectations about the buyer's role

Here's the one nobody talks about. Most agency-buyer relationships die because the buyer thought they were hiring "a team that handles everything" and the agency thought they were hiring "a team that needs strategic input."

Reality: marketing requires both. The agency brings execution and tactical expertise. The buyer brings business context — what's a good lead vs. a great one, what the sales team can actually close, what's happening with churn that marketing should know about, why the CFO is suddenly asking different questions. Without the business context, the agency is flying blind. Without the agency's execution muscle, the buyer is doing two jobs.

This goes wrong most often when:

Both parties are right. Both are losing.

The fix isn't a better contract. It's a 30-minute weekly call where the agency asks "what's going on in the business that we don't know about?" — and protects that calendar slot like it's the entire engagement.

1. The senior person disappeared

This is the number one reason. By a wide margin. Probably 50% of agency-client breakups can be traced to it.

The buyer hired the agency because they were impressed by the senior strategist who pitched. The senior strategist runs the kickoff, leads the first month's work, sets the tone. Then around month three, they hand the day-to-day to a junior account manager. The buyer notices.

From the agency's side, this is rational. Senior people are expensive. They can't be billable on every account every week. Pods of senior + junior staff are how agencies stay profitable. Fine.

From the buyer's side, this is a betrayal. They paid for the senior person's brain. They got the junior person's calendar.

The mismatch festers for two to four months. The buyer asks themselves "wait, am I really getting senior strategy here?" The answer is no, not really. The fire happens at month seven or eight, framed as "we want to try a different approach" — but everyone in the room knows it's about the senior person who stopped showing up.

The structural fix

I'll tell you what we do at Purple Frog because it's the only durable answer I've found:

The result, anecdotally: our average engagement length is north of 24 months. The industry benchmark is 12-14. The math works because client lifetime value carries us, not new-business velocity.

The honest part

If you're a buyer evaluating an agency right now, here's the question to ask in the pitch meeting that nobody asks:

"Will the person presenting today be on the monthly review calls in month nine?"

Watch how the agency answers. The pause. The qualifications. The "well, depending on the engagement size, we have a really strong account team that..." That's the answer. That's the future you're buying.

If the answer is a clean yes, ask for it in writing. If it's anything else, you now know why most agency relationships die at month six. Here's how we structure ours so they don't.