A four-stage path from diagnosis and baseline to a controlled pilot and an evidence-based scaling decision.
Days 1–15: diagnosis
Choose a process with an available owner, sufficient volume and a measurable problem. Map the actual flow, sources, exceptions and baseline. Define the pilot boundary and success criterion before selecting technology.
The deliverable is a scope decision: what is included, what remains outside and why.
Days 16–35: foundations
Fix data issues blocking the use case, define controls and design fallback. Build a prototype with representative data and test normal cases, exceptions and missing inputs.
Involve users in the design of human review. Adoption is built before release.
Days 36–65: pilot
Run the new flow within a limited scope while retaining comparison with the previous method. Measure outcomes, time, overrides, errors and use. Record every configuration change.
A weekly review should distinguish process, data, model and interface problems. Without this classification, teams fix the wrong component.
Days 66–90: scaling decision
Compare results with the baseline, validate the business case and document residual risks. Decide whether to expand, iterate or stop. Scaling requires permanent support, monitoring and ownership, not only technical capability.
Finish with a roadmap for later dependencies. The result of 90 days is an informed decision, not necessarily a full rollout.
Next step
Before choosing a tool, assess the process with the Process Readiness assessment. For complex initiatives, explore the AI Process Intelligence method and ControlRoom use cases.
Frequently asked questions
Can automation be completed in 90 days?
A narrow case can be validated and a scaling decision made; an entire organization cannot be transformed in that time.
What is the main pilot selection criterion?
A measurable problem with an available owner, accessible data and controllable risk.