What AI Scoping Costs: a Fixed Price You Know Up Front
One stage, at a fixed price quoted after a short call and known before it starts, ending in a written scope and a binding quote for the next stage.
How the price is set
Stage one, characterization, is a fixed price for one operational problem. We quote it after a short call, once it is clear which problem is being characterized and which data needs examining, and you know it before the stage starts: no hourly billing, no additions along the way, no surprise at the end. The stage usually runs a few weeks.
The price is not listed here because characterizing the same problem does not cost the same at every business. A plant whose data sits in an orderly ERP is a different job from a business whose data is scattered across spreadsheets. What is the same for everyone: the price is set in advance and does not move.
One operational problem means, for example, how much to order of each item, which customers are starting to churn, or which documents the team keeps looking for and cannot find. If three such problems sit on your desk, that is either three characterizations or one that ranks them, and we decide which on the call beforehand.
What the price covers
- Mapping the business problem and the workflows around it, with the people who actually run them rather than from a diagram.
- An assessment of the data that genuinely exists, not the data everyone assumes exists: what is there, at what quality, and what is missing.
- Success metrics (KPIs) defined in business terms, for example reducing purchasing error by at least 10% from the current state. The metric is fixed before anything is built, so nobody moves the goalposts afterwards, us included.
- An explicit integration plan: which systems, where the data flows, and who approves access.
- A binding quote for stage two.
The person sitting with you in the characterization sessions is the person who writes the code and implements the system. There is no sales layer and no separate delivery team, so nothing is lost in an internal handover.
What you hold at the end
A characterization document with the problem definition, the state of the data, the proposed approach and the agreed success metrics, alongside a binding quote for stage two. The document is yours whether you stop or continue with someone else. It is worth real money the next time somebody tries to sell you an AI system, because it lets you compare proposals against a written scope instead of against promises.
Why this stage in particular is priced up front
Because it is the only stage whose scope is known before we have seen any data. The price of stage two and stage three is set by data readiness, how many decisions the system has to serve, how deeply it integrates with the systems you already run, and whether everything must run inside a closed network. Those four factors are examined during characterization itself, which is why a binding quote for what follows arrives at the end of it rather than before it. What drives the price sets out each one, and is your data AI-ready lets you size the largest of them yourself before we have even spoken.
The honest exit point
Sometimes characterization leads to the conclusion that AI is not the right tool for the problem, or that the data is not ready yet. We say so in writing, along with what is worth fixing first, and part on good terms. That is a legitimate outcome of the stage rather than a failure of it, and it is far cheaper than discovering the same thing after a year of building.
When you are better off not paying for scoping
If the problem is already characterized on your side, with a written scope and agreed metrics, what you need is a quote for the build and not another characterization. Send the document and we will answer it.
If you are looking for a website chat widget, off-the-shelf tools do that well and cheaply, and we will point you there.
If you need an organization-wide programme or broad staff augmentation, a large integrator is the right address, and we will tell you that on the first call rather than after you have paid.
When the work is office work: a retainer instead of another salary
When the project is an agent that takes on repetitive office work, such as drafting the same document again and again, finding documents, or flagging the exceptions someone needs to look at, it helps to think of the cost after delivery the way you think of a position in the office, not the way you think of a software license. Instead of another salary, a fixed monthly retainer that covers running the agent day to day, monitoring, calibration and maintenance.
That also sets the right question: not what the software costs, but what that slice of work costs you today, in hours and in salary. We work that out together during scoping, on your own numbers, so you know what you are comparing against before you decide to continue.
How to start
With a 20 to 30 minute call in which you bring one operational problem that hurts. You do not need to arrive knowing what AI is supposed to do about it: working that out is precisely what stage one is. After the call you will know whether there is a project here and, if so, exactly what goes into the scope of the characterization and what its fixed price is. Book a call.
What does the scoping stage cost?
A fixed price for one operational problem. We quote it after a short call, once it is clear which problem is being characterized and which data needs examining, and you know it before the stage starts. That is the full price of the stage: no hourly billing, no additions along the way, no surprise at the end. The stage usually runs a few weeks.
Why is the price not on this page?
Because characterizing the same problem does not cost the same at every business. A plant whose data sits in an orderly ERP is a different job from a business whose data is scattered across spreadsheets and email, and one price for everyone would mean guessing. What is the same for everyone: the price is set in advance and does not move.
What do you get at the end of scoping?
A characterization document with the problem definition, an assessment of the data that actually exists, the proposed approach and the success metrics (KPIs) agreed in business terms, alongside a binding quote for stage two. The document is yours either way, whether you stop or continue with someone else.
What if scoping concludes it is not worth continuing?
It stops there, and that is a legitimate outcome of the stage rather than a failure of it. You get the reasoning in writing, along with what is worth fixing first. The findings stay with you, and that conclusion cost you one scoping stage instead of a year of building.
Is the first call already stage one?
No. A 20 to 30 minute call exists to work out whether there is a project here at all and which problem is worth characterizing. After it you receive a fixed quote for the characterization, and stage one begins only if you accept it, with the scope settled on that call.
After delivery, what does an agent doing office work cost?
A fixed monthly retainer, instead of another salary. It covers running the agent day to day, monitoring, calibration and maintenance, and it too is set per client by the scope of the work. To judge it fairly, compare it with what that slice of work costs you today in hours and salary, not with the price of software.