Every stage has a deliverable.
The plan below is an example, not a quote. What is fixed is the shape: the order of the work, one thing finished at each stage, and the scope agreed in writing before any work starts.
The shape of an engagement
- Free 60-minute diagnostic
- Fixed scope, fixed price
- A deliverable at each stage
- Then the care plan
The first hour costs nothing, and you keep what comes out of it.
On a call or on site, and it works far better with the people who actually do the work in the room rather than only the person who signs. There is no deck. We ask what the week looks like, then we time the parts of it that happen the same way every time.
You leave with a written, ranked account of where your hours go — what repeats, roughly what it costs you in hours a week, and which of it we think is worth automating. That is yours whether or not you ever hire us, and it is useful on its own.
The hour tells us where to look; it is not a measurement. If you go ahead, the first two stages go and measure it properly, because a fixed price has to be built on observed work rather than on a conversation.
If the honest answer is that nothing in your week is worth automating yet, that is what the hour is for too.
Book a diagnosticAn example six-week plan, stage by stage.
An illustration, not a quote — the scope is made up. Yours gets its own dates, in writing, before anything is signed.
- 01
Measuring
We sit with the people doing the work — on site or over a screen share — and time what repeats. Nothing gets proposed yet. A system designed off a description of the work is a system that meets the description instead of the work.
From you Half a day each from two or three people, and read access to the systems the work lives in.
You get Every repeating task we found, with an hours-per-week figure against it.
- 02
The map, and the build order
What we measured gets written up and ranked: what we would automate, what we would leave alone, and what is a broken process rather than a slow one. If the measurement does not match what we were told in the first hour, the plan gets corrected here.
From you An hour to argue with the ranking. You know the business; we have only been watching it.
You get A written, ranked map of where the hours go — yours to keep whether or not you continue — and an agreed order to build in.
- 03
The first system
The biggest item on the list gets built, into the systems you already run. It handles real work on a narrow slice — one job type, one branch, one queue — so that if it is wrong, it is wrong somewhere small.
From you The person who owns that process, for a couple of hours.
You get One system running on live work, with a person checking every output.
- 04
The rest of the shortlist
The same pattern, faster, now that we know your data, your naming and your exceptions. The slice on the first system widens as it earns it rather than on a date.
From you Exceptions, as they surface. They will surface, and that is the useful part of this stage.
You get The shortlist built. None of it trusted yet.
- 05
Your team tries to break it
Everything runs on live work while we sit next to the people using it. This is the stage that finds the customer whose name has a slash in it, the job type nobody mentioned, and the Friday afternoon when three things happen at once.
From you Your team using it for real, and telling us plainly what annoyed them.
You get A list of what broke, with every item fixed or explicitly deferred in writing.
- 06
Live, and handed over
Two training sessions and a written runbook: what each system does, what it will not do, where it hands to a person, and what to do on the day it is wrong. Monitoring and alerting go on, pointed at us.
From you An hour each for the people who will use it, and a decision on who gets called first.
You get It is live, the build is closed, and running it becomes our job.
The build ends. Then the care plan runs.
The build is fixed scope at a fixed price. When it closes you have the systems running on live work, a written runbook, training done, alerting pointed at a named person on our side, and a list of what we deliberately did not automate and why.
What you do not have is something that keeps working on its own. That is not a flaw in what we built; it is what automation is. Integrations break, models change, and your business does not hold still. So the monthly figure is not an upsell you can decline — it is quoted with the build, so the whole cost of year one is on the table before you commit to any of it.
If we stopped, it would stop. Not that afternoon, but within weeks.
Monthly required, quoted with the build
What the month actually covers
- Being the one who notices. An API version gets retired, a field gets renamed in your CRM, a model changes how it reads a message. Someone has to catch that before your customer does, and it is not going to be you.
- Fixes, with no ticket price. Breakage is not a change request. Repairing what we built is the thing you are already paying for, not a line item that appears afterwards.
- Small changes as the business moves. A new job type, a new form, a new branch, a new person who needs different access. Businesses move; a system that cannot move with them stops being used within a quarter.
- The next item on the map. Once what we built is stable, we go back to the ranked map and take the next thing on it. The map does not stop being useful the day the build closes.
- The tooling, held and operated. AI model usage and any tooling the build needs run on accounts we hold and operate, and are billed through to you at cost — quoted before we start, never marked up. Your own systems and data stay yours.
A 60-minute diagnostic, then a fixed scope. You keep the map either way.
Free, no pitch deck. We look at what repeats and tell you which of it AI can actually do well — including when the answer is none of it.