AI strategy
Why AI pilots stall between demo and production
The gap is rarely model capability. It is unit cost, data readiness, evaluation, and an unanswered question about accountability.
7 min read
Practice 04
AI Adoption and Organisational Roadmap
Move from scattered pilots to an operating model that ships.
Engagement shapes
Adoption audit
3 weeks
Roadmap and operating model
8 weeks
Enablement delivery
Priced per cohort
Embedded adoption lead
Retainer
Who this is for
- Organisations where AI usage is real but ungoverned and uneven
- Leaders who cannot report on adoption because nobody agreed how to measure it
- Teams consolidating a sprawl of overlapping tools and licences
The problem
Adoption usually fails as an organisational problem rather than a technical one. The tools work. The operating model around them does not exist.
Nobody owns the decision rights, so every team re-litigates the same questions about data handling, review thresholds, and which tools are allowed.
Without agreed metrics, leadership cannot tell productive adoption from licence spend, and the programme loses its funding at the first budget review.
What we do
The workstreams in an engagement
01
Current-state audit
Tools in use, shadow usage, spend, and what has actually changed as a result.
02
Capability and skills
Assessed by role and function, so enablement targets the real gap.
03
Operating model design
Centre of excellence, federated, or hybrid, with decision rights and escalation written down rather than assumed.
04
Platform and tooling standards
A supported set, a review path for additions, and vendor rationalisation.
05
Enablement programme
Built per persona, from executive through to engineer, with facilitator material included.
06
Change and communication
Incentives, internal messaging, and the visible sponsorship the programme needs.
07
Metrics and reporting
Definitions first, then a dashboard specification leadership will read without translation.
08
Guardrails
Acceptable use, data handling, review thresholds, and escalation paths, written to be followed rather than filed.
Deliverables
What lands on your desk
You keep all of it, including the method behind it, so the work can be repeated without us.
- Adoption maturity baseline with the measurement method documented
- Four-quarter adoption roadmap with owners and gates
- Operating model, decision rights, and RACI
- Enablement curriculum and facilitator material
- Metric definitions and dashboard specification
- Policy pack covering acceptable use, data handling, disclosure, and review thresholds
Standards we work against
- NIST AI Risk Management Framework
- ISO/IEC 42001
What we do not do
- We do not run a training programme with no operating model behind it.
- We do not report adoption as licence count.
Engagement shapes
Three sizes, not one package
Durations are indicative and depend on estate size. Scope and price are fixed in writing before the engagement starts.
| Shape | Duration | What you get |
|---|---|---|
| Adoption audit | 3 weeks | Baseline, shadow-usage picture, and the top blockers |
| Roadmap and operating model | 8 weeks | Model, roadmap, policy pack, and metrics |
| Enablement delivery | Priced per cohort | Trained cohorts and internal facilitators |
| Embedded adoption lead | Retainer | Someone accountable for the roadmap landing |
Related insights
Our thinking on this
Next step
Start with a briefing, not a proposal.
Thirty minutes on ai adoption. We will tell you whether we are the right firm for the problem, and who to talk to if we are not.