Your team is doing 500 hours a year by hand. We’ll show you which hours to take back first.
Two working sessions with a senior Cogent engineer. We take your business apart, find every workflow Claude can take, and hand you a ranked list with the hours and the money attached.
No slides. No hard sell. We’ll tell you in 15 minutes whether this is worth doing.
You use Claude every day. You know what it can do, because you have watched it do it.
Your team has the same licence, the same tool, the same access. And they use it to reword emails.
So the good work still comes to you. The proposal that needs judgment. The report that has to be right. The decision nobody else feels confident making. You have added an AI department to your job, and you are the only person in it.
Meanwhile the work carries on being done by hand. Not the interesting work. The recurring, weekly, someone-has-to-do-it work. In most businesses this size that is somewhere north of 500 hours a year, which is a quarter of one person’s entire year, spent on things a well-built system would handle.
Every month you wait, that gap gets a little wider and you get a little more of it on your desk.
It isn’t that your team won’t use it.
The reason Claude works for you and not for them has nothing to do with curiosity, or training, or how technical anyone is.
It works for you because you already know the answers. You know what a good proposal looks like here. You know which clients get the careful version. You know the three things that have to be checked before anything goes out. You supply all of that without noticing, in every prompt you write.
Your team does not have any of it. So they get generic output, decide the tool is overhyped, and go back to doing it the old way. Which is the correct decision, given what they were working with.
The fix is not more training. It is getting what is in your head into a place the whole team can reach.
Two sessions. One ranked list. Every hour accounted for.
Before we start
You send us what you have already tried and where you think the friction is. We read it before session one so we are not spending your time on background.
Session one
You and the one person who will run this afterwards. We go through the business function by function and surface every recurring workflow: who owns it, how often it runs, what it costs in time, and where it queues behind a single person.
Between sessions
We cost every workflow, size the effort to automate it, score it against what Claude actually does well, and rank the lot. This is where most of the work happens and you do not have to be in the room for it.
Session two
We put the ranked list in front of you and argue about it. You know things about your business we cannot see in one session, and the order changes when you push back. We leave with the top three agreed.
The readout
We walk you through the finished document, section by section, so you can use it without us.
What you walk away with
A written teardown of your business, covering:
- Every recurring workflow we found, named in your language
- Who the bottleneck is on each one, by name
- Hours per week and hours per year
- What each one costs you annually
- What automating it actually takes
- A priority score, and why it scored that way
- Your top three, in order, ready to start on Monday
Plus the total. Hours found, annual cost, and where to begin.
This is for you if
- You personally get real value from Claude and your team does not
- You have between roughly 5 and 60 people
- There is at least one person in the business everything queues behind
- You want to know where to start, not be told AI is important
This is not for you if
- Nobody in the business has used Claude seriously yet. Start there first, and we will happily point you at how.
- You want us to build it this week. The teardown tells you what to build. It does not build it.
- You are looking for a list of tools. We are not going to recommend software.
No teardown has run yet. Here is what we have built for teams like yours.
The teardown is new. Cogent is not. These are teams that stopped doing work by hand, permanently.
less time spent drafting content, and research that used to take years now takes hours.
reduction in manual data entry on order processing. A team stopped doing something by hand permanently.
increase in order handling capacity, without a proportional increase in headcount.
“Working with the Cogent team has been an exceptional experience from start to finish. If you’re looking for someone who can combine deep technical skill with strategic insight to build AI-powered systems that drive real business impact, they are your partner!”
Cogent Labs · 60+ engineers · Clients in 22 countries · Anthropic Claude Partner Network member
What people ask before they book.
“I already know where my bottlenecks are.”
Probably. Most founders do. What you likely do not know is which of them Claude handles well, and that is where people get it wrong. The obvious candidate is usually the thing that annoys you most, which is rarely the thing with the best return. We rank by what works, not by what irritates.
“Is this just a sales call for the buildout?”
No. It is a paid piece of work with its own deliverable, and plenty of people take the list and act on it themselves. If you want us to build it afterwards we will quote for that, and you are under no obligation.
“Who needs to be in the room?”
You, and the one person who would run this afterwards. Not your most senior person. The one who will actually use it. Having them hear it first-hand is half the point.
“We tried this already and it didn’t stick.”
That is the most common reason people book. Setup without context does not stick, because the tool has nothing to work from. The teardown is about finding what has to be captured, and in what order.
“What if you don’t find much?”
Then you do not pay. See below.
We find 500 hours, or you don’t pay.
Five hundred hours a year is twelve and a half working weeks. A quarter of one person’s entire year.
If we go through your business and cannot find at least that much automatable work, the teardown is free. We have not seen a business of this size where it was close.
You already know what this can do. Let’s get the rest of your team there.
Fifteen minutes. We will tell you whether the teardown is worth doing for a business like yours, and if it isn’t, we will say so.
Book a 15-minute call$497. Two 90-minute sessions and a 30-minute readout. 500 hours or you don’t pay.