The five stages
Discover
Understand the challenge, the people closest to it, the evidence already available and the constraints that shape what is possible.
Design
Translate the opportunity into a clear intervention, program, solution, learning journey, operating model or experiment.
Build
Create the assets, experiences, prototypes, tools and infrastructure needed to make the design real.
Activate
Deliver, facilitate, launch or pilot alongside the people who will use it, not around them.
Improve
Use evidence, feedback and outcomes to refine what has been built and decide what should happen next.
Agree what success looks like
before the work starts.
Before delivery, we turn the brief into a shared roadmap: what is changing, what we are building, how success will be measured and who owns what.
Tailored blueprint
Define the custom elements, participant journey, outputs and dependencies required for the engagement.
Outcome and KPI alignment
Agree the outcomes and measures that matter before choosing activity metrics.
Engagement plan
Confirm stakeholders, decision points, participation requirements and sign-off.
Feedback loops
Build structured ways to capture feedback and adjust the work during the engagement.
Accountability
Assign an owner to each milestone, decision and deliverable.
Responsible use
is part of the workflow.
Where AI is part of the engagement, governance is designed into the way people learn, build and make decisions, aligned to applicable requirements and the client's own policies.
Governance and policy
- Authorized tools and data
- The AI and data requirements that apply to the client
- The client's own AI usage policy, acknowledged and embedded
- Role accountability
Practical safeguards
- When not to use AI
- Verification of outputs
- Bias and hallucination checks
- Human approval gates
Delivery
- Responsible-use content in relevant sessions
- Safe-use settings shown on real tools
- Policy acknowledgment and escalation routes where required
Measure the change at the level
it is meant to create value.
We build measurement around the intervention. A learning program should not be measured like a software deployment, and a pilot should not be measured like a culture initiative. The framework below is useful for AI implementation where these layers apply.