01 / FEATURE
AI readiness assessment
Data, talent, infrastructure, and governance, we map what you have and what you need.
AI Strategy & Governance, from boardroom to production.
Most AI strategies die in the slide deck. We translate ambition into a governed, sequenced plan that your board can approve and your engineers can execute.
What it is
Strategy that survives contact with reality.
Most AI strategy work ends with a sixty-page deck and a steering committee meeting that gets cancelled. Ours ends with a written target architecture, a triaged backlog of use cases, and a governance operating model your General Counsel will sign. The deliverables are reusable artefacts your teams reference for the next eighteen months, not a one-time read. Every strategy we produce is delivered with the engineers who'll build it in the room. If a recommendation can't ship, it doesn't make the deck.
What you get
02 / FEATURE
Use-case portfolio
Prioritised roadmap of opportunities, sized by value, complexity, and risk.
03 / FEATURE
Governance framework
AI Verify, NIST, and OWASP controls translated into your operating model.
04 / FEATURE
Board-ready narrative
A story your CEO and CFO can tell, with metrics that survive diligence.
05 / FEATURE
Execution runway
Sequenced delivery plan with named owners, success metrics, and funding milestones.
06 / FEATURE
Co-funding guidance
We map your eligibility for IMDA schemes and handle the paperwork.
The Framework
How it works
Eight to twelve weeks. Three workstreams in parallel, discovery, architecture, governance. Most clients hit eight weeks; twelve is the ceiling.
PHASE 01
DISCOVER
Interviews with C-suite and the operating layer. Inventory of every AI initiative, sanctioned, shadow, and stalled. We surface the politics as well as the tech.
Deliverable
Stakeholder map + initiative register
PHASE 02
TRIAGE
Score every candidate against the rubric, value, feasibility, governance load. Build the value model on your own numbers. Cut the bottom half with reasons attached.
Deliverable
Scored backlog + kill list
PHASE 03
ARCHITECT
Target-state design for the top three to five use cases. Service boundaries, data lineage, model lifecycle, integration contracts. Reviewable by architects, defensible to procurement.
Deliverable
Target architecture + migration path
PHASE 04
GOVERN
Decision rights, review gates, and the policy that ships with the code. Aligned to AI Verify and NIST AI RMF. Reviewed with your General Counsel and Risk function.
Deliverable
Operating model + policy library v1
Real-world example, ASEAN universal bank
From 41 pilots to 6 production systems in nine months. An eight-week strategy engagement under MAS FEAT and the bank's own model risk policy.
Initiatives in flight
Forty-one active AI initiatives narrowed to eleven prioritised, six shipped in nine months.
41 → 6
Decision velocity
Six overlapping committees collapsed into a single AI Council with written decision rights.
6 → 1
Pilot-to-production
Average time from pilot to a production deployment, before and after the engagement.
14mo → 11wk
The deck was the smallest thing they delivered. The thing we still use is the operating model, it ended a two-year argument about who approves what.
Group Chief Data Officer · ASEAN universal bank
Where this connects
Reference architecture flows into FORGE, the build framework that picks it up
Governance operating model flows into Responsible AI controls
Prioritised backlog feeds the AI Sprint that proves the first use case in four weeks
Available to executives as an ongoing relationship via the AI Leadership Clinic
Book a strategy session
30-minute call. We'll tell you whether strategy is the right first step, or whether you should skip straight to a sprint.
Book strategy session
30 minutes · reply within 1 business day