Skip to content

Decide

Work out what is actually worth building

Readiness assessments, use-case discovery and build-versus-buy calls, ending in a roadmap with costs attached. Often the most valuable output is the list of things we talked you out of.

Sound familiar?

  • Your board has asked what your AI strategy is and the honest answer is that you do not have one.

  • Every vendor is selling you AI and you have no way to judge which claims are real.

  • You have budget and enthusiasm but no way to rank twelve ideas against each other.

What we do

Capabilities

Readiness assessment
Whether your data, systems and processes can actually support what you want to build. Usually the binding constraint, and rarely the model.
Use-case discovery
Workshops with the people doing the work, producing a ranked list scored on value, feasibility and risk rather than novelty.
Build versus buy
Most of what you want already exists in a product. We tell you where that is true, because custom-building a commodity is the most expensive mistake available.
Costed roadmap
A sequence with effort, cost and dependency attached — something you can take to a board, not a slide of themes.
Governance and policy
Acceptable use, data handling, approval thresholds and audit. What your enterprise customers will ask about in security review.
Team capability
What to hire, what to outsource and what to train for, so you are not permanently dependent on an agency.

Outcomes

What changes

  • Before

    Every function has an AI idea and there is no way to compare them.

    After

    Use cases are assessed on the same axes — value, data readiness, risk — and ranked against each other.

  • Before

    The roadmap is a list of technologies rather than a sequence of decisions.

    After

    It is a costed sequence, with what must be true before each step stated plainly.

  • Before

    Data readiness is discovered halfway through the build.

    After

    It is assessed first, because it decides which use cases are possible at all.

How it works

The engagement

  1. 01

    Assess

    1 week

    Systems, data quality and process maturity. We find out what is possible before discussing what is desirable.

  2. 02

    Discover

    1–2 weeks

    Workshops with the teams who would use it, surfacing the unglamorous high-volume workflows that usually pay best.

  3. 03

    Rank and cost

    1 week

    Scoring on value, feasibility and risk, with an estimate attached to each and an explicit list of what to drop.

  4. 04

    Decide

    ongoing

    A roadmap you own. If the right next step is an ERP project or a data cleanup rather than AI, that is what it will say.

Go deeper

The playbooks set out how these processes run, step by step — what is standard, what gets configured, and what changes.

Read the playbooks

Stack

  • Process mapping
  • Data quality profiling
  • Evaluation design
  • Vendor assessment

Questions

The things people actually ask

How is this different from a consultancy slide deck?
The output is a costed, sequenced roadmap written by the people who would build it, so the estimates carry accountability. And we are willing to conclude that you should not build anything yet.
What if the answer is that we are not ready?
Then you have saved a year and a budget. Most AI projects fail on data and process, not on models, and the readiness assessment exists to find that out in a fortnight rather than a year.
Do we have to use you for the build?
No. The roadmap is yours and is written to be executed by anyone, including your own team. We would rather be chosen on the build than locked into it.
How long does this take?
Three to four weeks end to end for a company of 20–300 people. Longer and you are producing a document rather than a decision.
Who should be in the room?
The people who do the work, not only the leadership team. The best use cases are almost always named by whoever is currently doing the task by hand.
What does it cost?
Fixed price, quoted up front and sized to the scope. It is deliberately small relative to a build, because its job is to stop you spending the larger number badly.

Tell us what’s breaking.

Thirty minutes, no deck. We’ll tell you whether this is the right place to start.

Book a 30-minute call