Workplace AI Enablement PlaybookInstitutionalize

Adoption & Outcome Dashboard

Training completion and license counts measure spending, not change. Track the full chain from access to outcome on one page, and pair every number with its baseline — the measurement question is always the same: compared to what?

14
Use whenPilots are running — fed by the pilot measures and reviewed at every council meeting on a standing agenda.
OutputA live adoption and outcome dashboard; a maintained risk and decision register.
A

Reporting context

Reporting period
Prepared by
Council review date
1

Six measurement layers

Layer
Measure
Baseline
Current
Trend
Notes
Access
Approved tools provisioned; employees with access.
Usage
Active use; repeat use; use embedded in named workflows.
Performance
Time and cycle-time changes against baseline; quality; error and rework rates.
Experience
Employee experience of the changed work; customer or constituent outcomes.
Risk
Risk events; near misses; human-review burden and whether it is sustainable.
Capacity
Internal ownership; reusable organizational knowledge; reduced dependence on outside help.
Report against the baselines documented in the workflow inventory and pilot charters — every number paired with its baseline.
2

Risk & decision register

Date
Event or near miss
Control that caught it
Decision
Q2 pilot review
Draft briefly included an insufficiently de-identified example.
Caught in the grants manager's review — which is what the review was for.
Logged as near miss; de-identification step reinforced.
Log every risk event and near miss, including the ones the controls caught — those are evidence the system works, not embarrassments to bury.
You should now have