How to Estimate Agency Capacity When Nobody Tracks Time

It’s a question I get from agency owners. Nobody on my team tracks time, so how do I know if we can take on five more clients?

When I ran my agency, our fractional CFO asked me to go through this exact exercise, so we’d know where we were over resourced and where we were under resourced before we hired or rightsized the team. I never did it. I was too busy in the business, and when your brain wants to be precise in the absence of precise data, you freeze.

Now I do it for the agencies I advise. It’s simple on the surface, and it’s really hard to do on your own while you’re inside the machine running it.

The time tracking trap holds people up

A lot of owners believe you can’t do any of this math without time tracking data, and without accurate time tracking data at that. So they wait. They tell themselves they’ll get the team tracking first and run the numbers once there are a few months of history.

Meanwhile the decisions don’t wait. The prospect wants an answer on the proposal. The team lead says they’re slammed. Somebody has to decide whether to hire, and that decision gets made on vibes anyway, with nothing written down to check it against later.

Time tracking data is great to have. It’s the best way to reconcile an estimate and make it more accurate. But you don’t need it for a first pass.

How the estimate gets built

Start with a gut check. Is the team busy? Are they at capacity? Does it seem like they could take on more? Your read is anecdotal, but it’s all you’ve got, so start there. It gives you a lay of the land and something to hold the math up against at the end.

Ballpark the hours in each client’s scope. Without time tracking, the only way to get estimates is to ask the people who have done the work a hundred times. On average, how long does it take you to do this piece? Then add it up across everyone who touches the deliverable, from the strategist to the writer to whoever does QA and sends it, and multiply that by how many of those deliverables are in the scope.

It doesn’t need to be precise. Some tasks get done in 20% of the time when your best person does them compared to someone else. You don’t need that level of detail at this stage. You need a ballpark of how many hours each client is paying for in a month.

Hold those hours up against what you’re paying your team and what you’ve sold. The difference is what’s available. I built a calculator that does this, and none of the numbers in it come from time tracking. They’re all estimates, chained together.

Run the same math on the scope you’re pitching. What does the new work take to deliver? Do you have room for it? If not, who would you need to hire?

The total hides which roles have room

Break it down by role, or by department, and look again. This step can change the answer completely.

An agency can look like it has plenty of room in total and still have no room at all in the one seat the new work depends on. Your writers might have hours to spare while your one developer is booked solid through the quarter. Five new clients that all need development work don’t fit, no matter what the total says.

The role view also tells you who to hire. “We need more people” is a feeling. “We’re short one designer’s worth of hours, and the account team has room” is a hiring plan.

“If the estimates are wrong, I’m making decisions on bad numbers”

This is the fair objection, and it’s the one that keeps people from starting. The estimates will be off. Some tasks will take longer than the team thinks, and some will take less.

The alternative to rough numbers is no numbers. Even if the model is only 60% accurate, it’s better than zero.

What the model gives you is guidance where you had none. It turns “I think we’re busy” into a number you can argue with. And the question you’re asking is whether the team has room for five more clients, so you don’t need the answer to the hour. You need to know whether you’re close to the line or already over it.

I also don’t treat the first version as the answer. It’s the beginning.

You tune it with time tracking

You run the math, you make the call, and then you watch what happens. If you oversold, you’ll feel it. The team gets stretched, deadlines start slipping, and you’re either hiring or figuring it out after the fact.

That’s when time tracking earns its place. Put it on the team for two weeks and compare what they spent against what the model said. Two weeks is enough to see which estimates were too light and which roles are carrying more than anyone thought. Then you update the model and run it again.

This is where the order matters. Time tracking with nothing to compare it to is a pile of hours. Time tracking against an estimate tells you exactly where the estimate was wrong, and that’s the thing you can fix.

Why it’s easier from the outside

Looking back, what froze me when our CFO asked was the precision. I knew the work too well to be comfortable with a ballpark, and I was too deep in the week to step back and build one.

From the outside, it’s a different job. I know enough to make ballpark estimates, I know what data to ask for, and I know how to chain the different pieces together into a realistic picture. That’s most of what an owner needs before they say yes to five more clients.

So start with the estimate you can make this week. It won’t be right. It’ll be close enough to decide, and it’ll get closer every time you tune it.