How’s your Ops?
Could be better.
Every agency’s operations are slightly (or very) behind where they want them to be.
It’s called having a finite amount of time and you don’t like thinking about it because mortality probably. But do spend some more seconds on this article. No one on their deathbed is wishing they had read less Dear Agency Founder.
(Or written less of it! AHAHAHAHAHAAHA)
Ops is very important.
But there’s too much focus on breadth rather than depth.
The breadth of Ops
This is how many of the things you do have a documented process that gets followed. Or, because it’s 2026, how much stuff is automated and run with tokens?
Never enough, and there’s a reason for that. Because if you had the time to build too much then you weren’t doing the real work of sales and delivery. AI can build your systems quickly now for sure, but it’s a matter of focus and energy.
You’re not going to be one of those agencies that brag about their incredible systems. You’ve got people to meet!
For me breadth is for show, it’s depth for the pro.
The depth of Ops
This is the important bit.
Ops is not a set of systems, in fact it can be no systems. Ops is a set of responsibilities and the depth of your Ops is how many of those responsibilities you’ve allowed to be handled by the business rather than just you.
My definition of Ops is:
Deliver the work we’ve won, on time, to the right quality, to a target profit and get paid for it.
It’s not a “how” because the how changes. As you grow you will achieve (or not achieve) this outcome in various different ways, starting with you making it all up and ending with whole departments (or at least individuals) dedicated to achieving each part.
You can’t not do Ops, you’re always doing Ops, you can only do it badly.
And so the depth of your Ops is how much you let go of these things and hand them over.
This is how Ops gets you out of delivery.
High-stakes Ops
You get what you incentivise.
Give your team a process and they will follow that process. This is low-stakes Ops. You feel like you’ve handed over responsibility but you haven’t handed over accountability.
If things happen that take the team outside that process then this falls down. If there’s opportunity to improve the process, it won’t be taken.
Low-stakes Ops also maps quite nicely to the things you can give AI to do.
Tell the team to make a profit and they will find a way (or you’ll find out you need a better team, get one, and they’ll find a way).
This is high-stakes Ops because… what if they don’t? Well that would be bad, but now you’ve got an interesting founder problem to solve.
And it’s not a problem you can solve with AI, AI has no skin in this game.
Solving this will have more impact than connecting your systems with Zaps or writing down processes. That’s not a founder approach, however satisfying it may be.
The founder approach is “how can I point this team I’ve invested in at a target and keep on top of whether they are hitting it?”.
The higher the stakes on that target, the deeper the Ops.
The stakes matter.
It’s the difference between completing work and making sure the client is happy.
It’s the difference between raising an invoice and making money.
It’s the difference between you spending less time in delivery, and actually stepping away.