[ AI readiness · Full playbook ]
// 14 / 20The buyer's playbook.
The working reference behind AI Automation Readiness Assessment 2026: what to ask, what to watch and how to run the first 90 days.
← Back to the guide[ RFP questions ]// 01
Questions that reveal real capability.
Ask for a relevant example, the decision made, the result or learning, and the person who would own the work.
- 01What measurable problem does this process create?
- 02How frequent and costly are exceptions?
- 03Which decisions can be deterministic?
- 04What data is sensitive or unreliable?
- 05What error could harm a person, customer or record?
- 06Where must a human approve, override or investigate?
- 07How will quality, latency, cost and drift be monitored?
- 08What manual fallback operates during failure?
[ Risk ]// 02
Make failure visible before signature.
| Risk | Early warning | Control |
|---|---|---|
| Wrong process selected | Low volume or value | Portfolio scoring |
| Automating waste | Broken process preserved | Simplify first |
| Unsafe autonomy | No human gate | Risk-tiered approvals |
| Weak data | Manual correction dominates | Data-readiness work |
| No operating owner | Pilot orphaned | Named product owner |
[ Value case ]// 03
Measure outcomes—not activity.
01Annual addressable hoursBaseline ________
90-day target ________
90-day target ________
02Cycle-time reductionBaseline ________
90-day target ________
90-day target ________
03Straight-through rateBaseline ________
90-day target ________
90-day target ________
04Exception and error rateBaseline ________
90-day target ________
90-day target ________
05Risk-adjusted valueBaseline ________
90-day target ________
90-day target ________
06Run cost per caseBaseline ________
90-day target ________
90-day target ________
[ First 90 days ]// 04
Move from evidence to operating rhythm.
01
Days 1–30
Inventory and score candidate workflows
Exit evidenceBaseline, owners, approved requirements and risks
02
Days 31–60
Validate data, prototype bounded pilot and evaluation
Exit evidenceWorking outputs, evidence and decision record
03
Days 61–90
Approve scale, operating controls, monitoring and ownership
Exit evidenceMeasured result, operating owner and next backlog
[ Reference desk ]// 05
Speak the same language.
| Term | Plain-language meaning |
|---|---|
| Deterministic | Same valid input produces a defined result. |
| Probabilistic | Output varies and requires evaluation. |
| Human gate | Required approval before consequential action. |
| Fallback | Safe manual or alternate route. |
| Observability | Evidence showing workflow behaviour. |
| Drift | Performance change over time. |
[ Playbook in hand? ]