Consultation
Most AI budgets are spent answering the wrong question well. A short engagement to find the right question is cheaper than the build that follows it.
What it is
We look at where work actually happens in your business and tell you, in writing, which parts are worth automating, which are not, and what it would cost to be wrong about each.
You get a ranked map with the reasoning attached, including the things we would refuse to build. We are not paid to find work, so finding none is an acceptable outcome.
How we build it
Inventory the work
A week of watching what your team does, measured in volume and in cost of error. Not a survey: observation.
Rank by cost of being wrong
A cheap mistake made ten thousand times a day is a better candidate than an expensive one made twice a year. Frequency and reversibility decide the order.
Write down what not to build
The most valuable page in the document. Every line on it is money you did not spend on a demo that looked impressive in a board meeting.
Who it is for
A good fit if
- Leaders who have been told to "do something with AI" and have no idea where to start.
- Teams with a stalled pilot that nobody can explain the value of.
- Anyone about to sign a large contract and wanting a second read first.
Not a fit if
- Engagements where the conclusion is already decided and you need someone to agree.
- Strategy documents that exist to be filed. We write things meant to be argued with.
See it
A reference build
The recommendation not to build
The most useful thing a consultation produces is usually a no. A process is too unstable, the volume is too low, the data does not exist, or a rule engine would do it for a fraction of the cost. Saying that early is the whole value of the engagement.
- A written map of candidate work, ranked with the reasoning
- An explicit list of what not to build and why
- A rough order of cost for the things that survive
This is a system we run, not a client case study. We do not publish client numbers we cannot show you the workings for.
Talk to someone who builds this
No qualification call, no discovery deck. Describe the work and we will tell you whether this is the right line for it.