Approach
Clarify to decide, deliver to act
I start from the business as it really runs, not from technology. My job is to make complexity readable, then build the simplest system that meets the whole need.
Vision
Augmented work, not replaced work
In most organisations, a large share of the working day is spent imitating a machine: looking for a piece of information, copying it, re-keying it, chasing it, consolidating it. None of it calls for judgement, yet it ties up skilled people.
Useful technology removes those mechanical gestures and gives teams back what requires discernment: understanding an unusual case, weighing options, advising, looking after a relationship.
That is the test I apply to every project. If a system does not free up time for decisions, expertise or relationships, it is not the right one yet.
The whole chain
From vision to measurement, without gaps
I hold the thread end to end. The intent does not get lost between whoever scopes, whoever builds and whoever trains.
Understand
The business, the friction, the goals.
Structure
Processes, data, priorities.
Design
Journeys, architecture, options.
Build
A solid, tested, documented system.
Secure
Access, data, compliance.
Deploy
Production release and monitoring.
Drive adoption
Training and support.
Measure
Actual usage and business impact.
Method
Ninety days, five decisions
Every stage ends with an explicit decision. You know what has been delivered, what remains uncertain and what you decide next.
Day 0 to 10
Understand what really happens
Interviews with the people doing the work, process observation, an inventory of tools, data and constraints.
- Deliverable
- Project summary and identified risks
- Decision
- Does the problem deserve a system?
Day 10 to 30
Scope and prioritise
Compared options, MVP scope, target architecture, budget and success criteria.
- Deliverable
- MVP scope and costed options
- Decision
- Which option do we launch?
Day 30 to 60
Prototype or pilot
A first usable system, tried with real users and representative data.
- Deliverable
- Pilot in real conditions
- Decision
- Does the pilot deliver?
Day 60 to 90
Deploy, train, measure
Production release, team training, first usage and impact indicators.
- Deliverable
- System in production and indicators
- Decision
- Extend, adjust or stop?
Afterwards
Improve and hand over
Stabilisation, documentation, skills transfer or maintenance.
- Deliverable
- Documentation and team autonomy
- Decision
- Who owns the system tomorrow?
Reading guide
How to read a case study
Every case study fits in five lines, from the stake to the limits; the narrative develops each one. The limits are part of the proof.
Stake
What was at stake for the client, in one sentence.
Mechanism
How the system produces that change.
Proof
What demonstrates it: measurements, usage, material available on request.
Impact
What it brings to the business, in time, risk or quality.
Limits
What is not demonstrated, or still needs work.
Guardrails
Governance, security, measurement
Human validation required
Commercial, legal, financial or sensitive decisions stay in human hands. Systems prepare, people decide.
Security by design
Access management, data isolation, reproducible environments and traceability belong to the initial scope, not to a later project.
Measure before extending
Every system ships with its own usage and impact indicators, defined before they are measured. We extend what works and stop what does not.
Let’s talk about what AI could change for you
A 30-minute discovery call, by video and with no commitment, to talk about your project, your constraints and the best next step.