A digital service review can produce a useful idea without showing whether it will work in everyday use. An Albany business might identify a better enquiry form, clearer customer instructions or a simpler internal approval process. A small pilot gives the team a way to test that idea before making a broader commitment.
Use this checklist to prepare a pilot with a defined question, limited scope and clear decision. It is intended for a manageable service improvement, with appropriate specialist advice where the proposal affects sensitive information or high-consequence decisions.
Check that the idea answers a specific problem
- Describe the problem in ordinary language.
- Identify who experiences it and at which stage.
- State what should become easier or more dependable.
A proposal to add another form is incomplete until you know why the current process fails. Perhaps customers do not know which details to provide, or staff cannot tell who should reply. These problems may need different changes.
Keep the service goal separate from the chosen product. A useful pilot can show that the original idea needs adjustment, including a simpler change than the one initially proposed.
Check the question the pilot must answer
- Choose one main uncertainty.
- Define the evidence that would support the next step.
- Record limitations that the pilot will not resolve.
For an enquiry improvement, the question might be whether a customer can provide the information needed for assessment without extra explanation. That is more useful than asking whether participants like the new design.
The Digital Transformation Agency’s research guidance recommends testing tasks and considering the whole user experience. These design principles can inform a small business pilot without making the government’s delivery framework a requirement for private Albany businesses.
Check the boundary, people and information
- List the tasks and users included.
- Use safe examples or only the authorised information needed.
- Explain how participants can ask questions or stop.
Keep the first pilot narrow enough to observe and support. A limited trial should not quietly connect every account or replace a process used by all customers. Identify who can approve any expansion.
Include people who reflect the relevant task and access needs. Testing only with the person who designed the process can hide unclear instructions. Take care with personal information and obtain appropriate permission for observation or recording.
Check the complete journey and an exception
- Follow the task from the user’s starting point to the intended result.
- Observe the staff action receiving the output.
- Include a correction, missing detail or another realistic exception.
A form submission may succeed technically while creating an unusable request. Watch what the next person needs to do, including checks and follow-up. Keep an acknowledgement distinct from a confirmed order or appointment.
Use a controlled demonstration for situations that could create real commitments. Do not make unnecessary purchases, lodge false complaints or send unapproved messages to customers simply to generate a test event.
Check time, cost and responsibility
- Name the pilot owner and support contact.
- Set an agreed spending and effort boundary.
- Record the start, review and commitment dates.
Include preparation, staff practice, supplier assistance and corrections alongside any subscription. Confirm what happens when a free period ends and who can cancel or approve ongoing use.
The Government’s business goal guidance discusses clear, measurable goals and review. Apply that discipline by specifying the useful result and how the pilot will be assessed, rather than inventing a target unsupported by existing records.
If an outside provider helps, confirm its WA contact times and the assistance included during the pilot. Identify who will respond if the test stops working, and give participants a clear route to that person. A pilot needs support proportionate to the task, rather than a promise that everyone will work it out.
Check how findings will be recorded
- Separate direct observations from assumptions.
- Record unresolved tasks as well as successful ones.
- Note additional work created for another person.
Use concise notes tied to the task. If somebody needs prompting, record where and why. If the pilot is too small or unrepresentative to support a strong conclusion, state that limitation.
A faster screen action does not necessarily mean a faster complete process. Include checking, correction and follow-up, and avoid turning estimated minutes into a claimed cash saving without a reasonable basis.
Check the decision and return arrangements
- Choose who decides to continue, change, test further or stop.
- Identify records and access to retain if the pilot ends.
- Confirm the process people will use after the decision.
Keep a return route for essential work so a failed pilot does not leave staff without instructions. Remove unnecessary access through an authorised process and reconcile any temporary records.
Keep the pilot notes and approved instructions together so another authorised person can understand what was tried. This is particularly useful if the original organiser is unavailable when the review occurs.
If continuing, document the next limited scope and review point. The pilot succeeds when it gives the business usable evidence and a clear next decision, even if the original idea is changed or set aside.