Practice 02 · AI & Automation Adoption
Everyone sells you AI. Nobody owns whether it works.
Readiness, selection, governance, rollout, and benefits — vendor-neutral, and run with the discipline of a provincial-scale deployment: four health authorities, roughly 4,000 clinicians.
The problem
The technology layer keeps getting bought. The operating-model layer keeps getting skipped.
Healthcare organizations are under real pressure to adopt AI — from boards, from funders, from clinicians who are already using consumer tools quietly and riskily. The vendors are ready. The budgets increasingly exist. And yet the pattern repeats: a pilot that impressed everyone stalls after procurement, governance questions surface after the contract is signed, adoption becomes nobody's actual job, and benefits get declared in a slide instead of measured in the operation.
None of that is a technology failure. It is an operating-model failure — and it is entirely predictable, because the organization bought a tool when what it needed was a change.
What the practice covers.
Readiness & opportunity assessment
Where AI and automation would actually pay off in your operation, what has to be true first, and what to leave alone — grounded in your workflows and your privacy obligations, not a vendor's demo.
Vendor-neutral selection & procurement support
Requirements that reflect the real workflow, structured trials on real work, privacy terms your counsel can sign, and a recommendation from someone with nothing to sell.
Privacy & governance
PIPA and PIPEDA alignment, privacy impact assessments, consent and transparency practice, and the governance frame that made a provincial rollout defensible — sized to your organization.
Deployment, change & benefits realization
Templates and workflows rebuilt around the tool, teams trained, adoption owned, and a baseline-to-90-day benefits report that says what actually changed — in your numbers.
How it lands.
AI readiness assessment
Fixed fee · 3–5 weeksThe honest map: opportunity, risk, sequence, and cost. Ends with the Decision Memo your board can act on — including “not yet” where that is the truth.
Selection & deployment engagement
Milestone-billed · 6–16 weeksFrom shortlist to adopted-and-measured: trials, governance, rollout, training, and the benefits-realization report at the end.
Standing AI advisory
Retainer · by agreementA senior, independent voice at the table as the portfolio grows — for organizations making AI decisions quarterly, not once.
Every engagement runs the Operating Review — Diagnose · Design · Deliver · Transfer — with milestone-gated fees and measured results. How we work
The fair question
“The vendor says they handle implementation. Why add a consultant?”
The vendor implements their product. They do not redesign your workflows around it, weigh it against the competitor they lose to, tell you when the honest answer is to buy nothing, carry your governance through a privacy review, or stay accountable for whether the benefit materialized after their invoice cleared. Vendor implementation is real and useful — and it is structurally incapable of independence. That gap is this practice.
Bring us the AI decision that's been stalled the longest.
Reading from this practice: Choosing an AI scribe in BC · Why local AI belongs in a clinic