A pilot should answer a specific operating question. Before proposing one, establish what the business is trying to improve and whether the current workflow can support a fair test.
Discovery questions
- What problem occurs, how often and with what consequence?
- Where does the work start and finish?
- Who owns the workflow and exceptions?
- Which systems and permissions are involved?
- What baseline data is available?
- What must remain under human control?
Observe a few real cases. A manager's description may miss informal steps employees rely on. Check for duplicate entry, hidden spreadsheets and approvals conducted through personal messages.
Define the trial before choosing the tool
Set the scope, duration, eligible cases and success measures. Include failure measures such as incorrect routing, duplicated records, customer complaints and manual correction time. Identify conditions that would pause the trial.
Confirm that staff can keep operating if the tool is unavailable. A manual fallback should be practical and tested, not a note in a proposal.
Explain what the pilot cannot establish. A short trial may demonstrate technical feasibility without proving long-term return or performance during seasonal peaks. Record changes in workload and staffing so the result has context.
A good proposal gives the business a decision to make at the end: continue, revise or stop. It should not presume expansion merely because installation succeeded. The existing multi-location pilot guide can help turn discovery into a bounded trial with a named owner.
Existing guides: Clinic enquiry routing · Reception model · Pilot design.