▪ The Six Phases
From ambition to governed production.
01
Discovery
2–4 weeks
Understand the problem before proposing an answer. We identify the process that needs to improve, the people and systems involved, who must approve the work and any limits we need to respect. Both sides then agree on a clear one-page brief.
Stakeholder mapping and interview series
Problem framing and hypothesis set
Success criteria defined in business terms
Risk register first pass
02
Assessment
3–5 weeks
Check whether the idea can work in your organisation. We look at four things: whether the data is usable, whether the technology is ready, whether your people have the right skills and whether the necessary rules and oversight are in place.
Data: source, quality, access and how current it is
Technology: platforms, security and monitoring
People: understanding, practical skills and support
Oversight: policies, risk level and named owners
03
Strategy
Sequenced
Choose what to do first and commit to a realistic plan. We compare each idea by value, difficulty, speed and risk, then create a 12–24 month roadmap with named owners, costs, dependencies and approval points.
Choose and rank the strongest opportunities
Define ownership, budgets and ways of working
12–24 month roadmap with named owners
Board-ready investment narrative
04
Build
8–16 weeks
Build the system to be reliable in everyday use. We begin with a repeatable set of tests based on real tasks and real data. Clear limits, monitoring and a safe way to reverse changes are included from the start.
Evaluation harness and graded task set
Safe connections to your existing tools
Firm rules and limits on what AI may do
Monitoring, activity records and problem alerts
05
Deploy
4–8 weeks
Roll out in three safe stages. First, the system runs beside the existing process without taking action. Next, it works with human approval. It only handles approved low-risk tasks on its own after the tests, rules and monitoring show that it is reliable.
Clear checks before the system receives more responsibility
Operating guides and plans for handling problems
Training for everyone who will use or manage it
Monitoring dashboards signed off with the client
06
Govern
Ongoing
Keep the system safe, useful and compliant after launch. We monitor performance, respond to problems and maintain the evidence your organisation needs under requirements such as the EU AI Act, NIST AI RMF, ISO 42001 and NDPA 2023.
Monitor changes in quality and business results
Clear rules for when the system needs updating
Respond to problems and learn from what happened
Keep the records and assessments needed for audits
▪ Engagement Shapes
Sized to the problem, not the invoice.
8–14 wks
AI Strategy
A board-approved plan with clear priorities, owners, budgets and a path to everyday use.
10–16 wks
Production Agent
One clearly defined process taken from the first workshop to safe use under human supervision, with testing and safeguards included.
12–18 mo
Transformation
A longer programme delivered in manageable stages, building systems, skills and knowledge that remain with your organisation.
Pricing depends on the work required, how long it will take and the experience needed. Strategy work has a fixed fee. Builds and larger programmes are divided into fixed-fee stages, with payment linked to agreed milestones. We can sign a mutual confidentiality agreement before any detailed discussion.
