Instagram LinkedIn X Youtube
Back to Blog

Two Agencies With the Same Revenue and Very Different Margins

Two $3-5M agencies, same revenue, same services. One is a lot more profitable than the other. The difference is how client work gets delivered.

From the outside you can’t tell them apart. Same kind of clients, same tools, similar teams, similar websites. The difference only shows up when you look at how a single deliverable gets made, step by step, and who’s doing each step.

The agency that never designed its delivery

This is most agencies, and it’s not because anyone did something wrong. It’s how agencies grow.

A client asks for a service, you build it on the fly, and you price it off the last proposal. The process gets passed from the founder to a first hire to the next hire, and nobody zooms out to look at the whole thing. Each person adds their own steps and workarounds, and none of them get written down.

Senior people absorb the hard parts because it’s faster than explaining them. AI gets bolted on where it’s easy, and the hours it saves go to the client as lower prices or extra deliverables nobody bills.

Margin slips a little every year, and the owner stays stuck in delivery because nothing runs without them. No single year looks like a crisis. It looks like a business that’s always a little busier and a little less profitable than it should be.

The agency that rebuilt one service on purpose

The second agency did one thing differently. It picked a service and designed it.

You sit with the delivery team and map how each deliverable gets made, step by step. Each step gets one owner, an hour estimate, and a yes or no on AI. The mechanical steps go to AI, so the team spends their hours on the strategy and craft only they can do.

Then you reset the price to match what the service costs to make and what it’s worth. Margin goes up, the team runs delivery without you, and you’ve got numbers a buyer would pay for.

The second agency decided to design the thing instead of letting it happen to them.

What the mapping session looks like

This part sounds abstract until you’re in the room, so here’s what it looks like.

You get the people who deliver the service around a table, or on a call with a whiteboard. That means the writer, the strategist, the account manager, and whoever else touches the deliverable, so you get how it runs instead of how the owner thinks it runs. You pick one deliverable, say a blog post, and you walk it from the moment the client asks for it to the moment it’s published.

Somebody says “then I write the brief.” Who writes it, how long does it take, and what’s in it? Somebody else says “then it goes to review.” Whose review, and what are they checking for? Every step gets written down with a name next to it.

The surprises come fast. It might be a step two people both think the other one owns, a review step left over from one difficult client years ago, or a senior person doing something a coordinator could do with a checklist.

Once it’s all on the wall, the AI question gets a lot easier. Some steps are mechanical, like pulling data, formatting, a first draft from a clear brief, or checking against a style guide. Some need judgment. You mark each one, and you decide on purpose what happens with the hours you get back.

“Our work is too custom to standardize”

I hear this from almost every agency, and there’s something to it. No two clients are the same, and the best agencies win on the parts that are custom.

But when you map a service, most of it turns out to be the same every time. The research steps, the handoffs, the formatting, the QA, the report. The custom part is usually the strategy and the creative, and that’s exactly what you want your senior people spending their hours on.

Designing the repeatable steps is what frees up the time for the custom part. An agency that treats every step as custom ends up with senior people doing data pulls, and less of their time goes to the thinking clients hired them for.

Why one service and not all of them

Most owners have never done this for a single service. That’s usually where the fastest money is.

Trying to redesign the whole agency at once is how this never gets done. Everyone is still delivering for clients while you do it, and a project that touches every team stalls the first time a big account needs attention.

One service is small enough to finish. You map it, cost it, fix the leaks, reset the price, and the numbers show you whether it worked. The team that went through it learns how, and the next service goes faster because they’ve seen it done.

The gap compounds

Run those two agencies forward a few years and the gap gets wider. The first one keeps adding clients to a process nobody designed, so every new account adds a little more drag. The second one adds clients to a process with known costs and a price that covers them, so every new account adds profit.

By the time either owner wants to sell, a buyer can see the difference in an afternoon. One has margin, documented processes, and a team that runs without the founder. The other has the same revenue and a lot of questions.

Both agencies started in the same place. One of them sat down and designed how a single service gets made.