Why Agency Processes
Break at 25 People

You hire your first account manager and think: finally, some breathing room.

Then you hit 10 people and everything still mostly works. Communication happens in one Slack channel. You know what everyone's working on. Scope docs get done because three people between them have the tribal knowledge to write them.

At 15 people, things get a little messier. You add a project manager. You create a Notion doc for how proposals should work. Someone ignores it. You let it slide.

At 20, you're moving fast. Revenue is up. You feel like you've figured something out.

Then you hit 25.

And suddenly nothing works the way it used to.

The 25-Person Ceiling Is Real — and It's Not a Coincidence

Agency founders and ops leaders talk about the 25-person wall like it's a curse. "We were flying and then everything fell apart." "We grew from 18 to 28 in eight months and we almost lost the company."

This isn't anecdote. It's organizational physics.

The reason 25 is the breaking point comes down to communication overhead. Dunbar's Number — the cognitive limit of stable social relationships — is often misquoted as 150, but Dunbar's original research identified a series of nested groupings. The first real inflection is around 15 (the "sympathy group") and the second is around 50 (the "band"). But 25 sits directly in the gap between them.

Below 15–20 people, informal coordination works because everyone has direct, repeated contact with everyone else. Information spreads through proximity and conversation. Accountability is social — you don't miss a scope doc because you'd have to look the client lead in the eye tomorrow.

Above 50 people, companies have typically formalized enough that systems carry the load. SOPs exist. Role boundaries are clear. People follow process because the alternative is chaos they can already feel.

At 25? You're too big for informal coordination to hold everything together. And you're too small — and often too fast-moving — to have built the formal systems that replace it.

The result is a no-man's-land where the old way fails and the new way doesn't exist yet.

What Actually Breaks (It's More Specific Than You Think)

When agency founders describe the 25-person wall, they use words like "chaotic," "misaligned," or "everyone's pulling in different directions." That's accurate but unhelpful — it doesn't tell you what to fix.

Here's what's actually breaking:

1. The Scope Doc Process

Below 20 people, scope documents get written by two or three people who carry everything in their heads. They know the client, the history, the margin requirements, the delivery format. They can turn a chaotic client briefing into a tight, billable scope in an hour.

At 25, that knowledge is no longer concentrated. You have account managers who weren't there for the original pitch. Project managers who joined six months ago and don't know why you do it that way. New hires who've never seen your scope template.

The scope doc process breaks in two directions at once: junior team members can't produce them without significant oversight, and senior team members can't afford to review everything. The result is slow, inconsistent, or wrong scopes — which means missed deliverables, scope creep, and margin erosion.

2. Briefing Translation

Every agency lives and dies by its ability to translate client briefs into actionable internal work.

When you're small, a senior person does this translation almost unconsciously. They read the brief, absorb what the client actually wants versus what they said, and convert it into tasks that the team can execute without a lengthy back-and-forth.

At 25 people, that brief lands on someone who's overloaded, and it gets handed off — usually partially interpreted — to someone else. By the time it reaches the person doing the work, it's been through three layers of imperfect translation. The client gets something that technically fulfills the brief but misses the point. Then you're doing revision rounds you didn't budget for.

This is the "translation tax" in its most expensive form: a 30–40% drag on team capacity caused by the friction between client inputs and internal execution.

3. Accountability Architecture

In a small agency, everyone knows who owns what. You can walk across the office (or scroll up in Slack) and find out in 30 seconds whether something's been done.

At 25, the org chart has enough layers that ownership gets murky. Project managers own timelines, but who owns the scope? Account managers own the client relationship, but who owns the deliverable quality? Strategists define the work, but who makes sure the brief gets from the strategy deck to the team?

The gaps between roles become hiding places for undone work. Nobody's malicious — there's just genuine confusion about who was supposed to do what. And by the time the confusion surfaces, you're already in a client escalation.

4. Onboarding New People Into "The Way We Do It"

Below 15 people, onboarding is osmosis. New hires shadow experienced team members, absorb the culture, pick up the informal norms. It works because the team is small enough that informal transmission is fast.

At 25, you're hiring faster than you can transmit institutional knowledge informally. New account managers start producing scope docs that don't match your standards. New strategists write briefs in a format nobody recognizes. New project managers invent their own systems because they can't see the one they're supposed to use.

This creates two costly problems: rework to fix outputs that don't meet standard, and inconsistency in client experience that damages your brand reputation.

The Root Cause: You're Running on Tribal Knowledge

The common thread across all four of these failure modes is the same thing.

You've been running on tribal knowledge — the accumulated, unwritten expertise of the people who built the agency. And tribal knowledge scales to exactly the size of the tribe.

Below 20 people, tribal knowledge is efficient. You don't need to document "how we write a scope doc" because the three people who write scope docs learned by watching each other, and they all know the same things.

At 25, the tribe has grown past the threshold where informal transmission works. You have a knowledge asset — the sum of everything your best people know about how to do great work — and no infrastructure to distribute it.

The answer most founders reach for is documentation. Write everything down. Create SOPs. Build the playbook.

This works, but it's slow and it's fragile. A doc in Notion doesn't enforce itself. A playbook doesn't tell you when it's being ignored. And documentation doesn't help you when a client sends a chaotic brief at 9pm before a 10am kickoff.

What the 25-Person Wall Is Actually Asking For

Agencies that successfully cross the 25-person threshold don't do it by writing more Notion pages. They do it by systematizing the specific moments where tribal knowledge used to do the work.

There are three moments that matter most:

The Brief-to-Scope Handoff. The point where a client's words become a billable deliverable. This is where the translation tax is highest and where the most expensive mistakes happen. Systematizing this means creating a structured process — and ideally tooling — that captures what your best senior people do intuitively and makes it repeatable for everyone.

The Kickoff-to-Execution Handoff. The point where a strategy gets converted into tasks. This is where ownership gets murky and where the accountability gaps live. Systematizing this means making roles explicit, deliverables concrete, and timelines tied to someone's name before work starts.

The QA Gate. The point where work moves from internal to client-ready. Agencies that cross 25 people successfully have a clear, fast quality check that doesn't require a senior person to read everything before it goes out.

Agencies that fail to cross 25 people either don't identify these handoffs as systematizable, or they try to systematize them with static documentation that no one follows.

Why This Gets Worse Before It Gets Better

Here's the brutal truth about the 25-person wall: the revenue pressure that got you here makes fixing the underlying problem harder.

You hit 25 people because you were winning business. You were winning business because your team was producing great work. Your team was producing great work because your best people were deeply involved in every piece of it.

At 25, you can't be deeply involved in every piece anymore. But the systems to replace your involvement don't exist yet. So you're choosing between quality and capacity at exactly the moment when you can afford to lose neither.

This is why so many agencies stall at 25–30 people for years. They're not failing — they're surviving. But they're not scaling, because scaling requires the infrastructure they haven't built.

The Agencies That Make It Through

The agencies that successfully cross 25 people and build toward 50 and beyond share a few observable patterns:

They identify the translation tax explicitly. They don't just say "our process is broken." They identify where exactly in the workflow time gets lost to rework, revision, and miscommunication — and they address those specific points.

They make scope and brief quality a first-class metric. Not just delivery speed or client satisfaction — the input quality that determines both. Agencies that make it past 25 people track how long scoping takes, how often briefs get revised, and how frequently deliverables miss on the first pass.

They use tooling that enforces structure without requiring discipline. The playbook problem is that it requires people to choose to use it. The best agencies at 25+ people are using tools that make the structured path the easiest path — where following the process takes less effort than improvising around it.

They protect senior time ruthlessly. At 20 people, senior involvement in every scope is sustainable. At 30+, it's not. Agencies that scale successfully find ways to distribute senior judgment — through systems, through training, through tooling — so senior people are exceptions handlers rather than default reviewers.

What To Do This Week

If you're at 20–30 people and feeling the wall:

Do an honest audit of your scoping process. How long does it take from client brief to signed scope? How many people touch it? How often does the final scope differ significantly from the first draft? The answers will tell you where your translation tax is highest.

Find the two or three people whose departure would break everything. Not because you're going to lose them, but because what they know is your single point of failure. Whatever they know that can't be written down is the exact thing that breaks when they're overloaded.

Map your handoffs. Brief to scope. Strategy to kickoff. Kickoff to execution. QA to delivery. At each handoff, ask: who owns this, what does a good handoff look like, and how would you know if it went wrong? The gaps are where your 25-person problems live.

The 25-person wall isn't a sign that something's broken at your agency. It's a sign that you've grown past the threshold where informal coordination can carry the load. The agencies that treat it as a systems problem — rather than a culture problem, a people problem, or a motivation problem — are the ones that make it through.

The work is specific and it's fixable. But you have to see the translation tax for what it is before you can start paying it down.