JSM pilot checklist
JSM Voting App Pilot: A Seven-Day Evaluation Checklist
Evaluate a JSM customer voting app with a seven-day pilot covering access, saved votes, organization demand, ranking, exports and clear rollout decisions.
In short
A seven-day evaluation schedule with observable acceptance criteria, an evidence sheet and explicit continue, revise or stop decisions.
In this guide
A useful JSM voting app pilot proves that the intended customers can save support on appropriate requests, unauthorized customers cannot see them, and agents can interpret the resulting demand. Use the seven-day schedule below as an evaluation plan, not an installation deadline, a trial entitlement or a promise of business results.
This checklist uses Wezy as the concrete example. We develop Wezy, and its supported scope is JSM Cloud, company-managed projects. Follow the current product guide alongside the tests; check Atlassian’s installation flow for the editions and subscription terms actually available to your site.
Define the pilot before installing anything
Choose reviewed feature requests, an agent owner, a project administrator and customer testers. Include authorized and unauthorized request access, plus organization membership cases if attribution matters.
Write the decision you want the pilot to support: “Can we collect demand on existing portal requests without changing their intended audience?” is testable. “Will this transform our customer experience?” is too vague for a bounded evaluation.
Use synthetic or carefully reviewed request content. Atlassian’s JSM access guidance distinguishes customers, agents and other project roles. Your test accounts should reflect the roles your rollout actually needs.
Seven-day pilot schedule
| Day | Main question | Evidence to collect |
|---|---|---|
| 1 | Can we configure the intended scope? | Saved rules, project type and test request list |
| 2 | Does customer access behave correctly? | Permitted and denied account results |
| 3 | Can customers save and withdraw support? | Portal states and current demand checks |
| 4 | Does organization demand mean what we expect? | Attribution cases and distinct organization totals |
| 5 | Can agents use the ranking responsibly? | A documented triage decision on known requests |
| 6 | Can we operate and troubleshoot the workflow? | Export scope and a known failure/recovery observation |
| 7 | Should we expand, revise or stop? | Acceptance sheet, unresolved issues and named owner |
Move a session if testers are unavailable; cover the questions in sequence.
Day 1: install and save a narrow configuration
Open Wezy Customer Voting on Atlassian Marketplace with an authorized administrator. Review permissions and complete the intended installation and licensing flow.
In the company-managed project, select the test request types, open stages and exact allowed statuses, then save. An agent reviews content and JSM sharing before explicitly enrolling each request.
Acceptance criterion: the administrator can identify the saved rules and the agent can distinguish enrolled requests from other requests. Installation alone does not satisfy this criterion. Use the first customer vote guide if this is your team’s first setup.
Day 2: prove both permitted and denied access
Have Customer A open an accessible enrolled request. Have Customer B, without request access, check the portal list and the direct request URL. Keep the browser sessions separate from your agent account.
Acceptance criterion: A sees the intended request and B does not obtain its content through the voting journey. If the outcome is wrong, inspect JSM sharing and configuration before expanding the audience. The permissions checklist gives the detailed test sequence.
Request visibility is part of the requirement; do not broaden sharing to increase participation.
Day 3: verify saved votes, withdrawal and uncertainty
Have A save one vote, read the confirmation and revisit the request. An agent checks Wezy Customer Demand. Then A withdraws and the agent checks the changed metrics.
Acceptance criterion: the team can show the saved customer state and corresponding current demand. Count confirmations. After a connection error, read the state and use the offered confirmation action before deciding what happened.
Do not require immediate JQL changes: portal confirmation, app-managed Jira field publication and indexing are separate stages. Record the observed result and investigate an unexplained discrepancy without inventing a refresh guarantee.
Day 4: test organization attribution with known examples
If useful to your workflow, allow a test organization associated with the service desk and use customers whose membership can be verified. Compare an attributed vote with an unattributed vote.
For a transparent example, two customers representing the same organization should mean two active votes and one distinct represented organization. A third unattributed vote increases active votes without adding an organization.
Acceptance criterion: the agent can explain V, O and optional attribution without treating an organization weight as permission. If Advanced scoring is being evaluated, save and preview it before explicit activation. Verify the active formula afterward; a license or saved draft is not activation.
Day 5: make one real prioritization discussion better
Ask agents to compare two known requests using vote count, represented organizations, recent demand and request context. Add effort, customer impact and any operational obligations outside the voting score.
Acceptance criterion: the team can explain a decision, including why it might choose the lower-ranked request. Voting informs triage; it does not promise delivery. The JSM prioritization framework can structure that discussion.
Do not declare trend success from missing history. A new installation accumulates snapshots; an uncovered period is not zero demand. Check complete covered UTC dates before using historical comparisons.
Day 6: check operations without destructive experiments
Prepare an analytical CSV with known dashboard filters. Check that the received requests match the scope the exporting account can access. Document where the file should be stored and who may use it.
Try a reversible condition such as removing enrollment from a test request, then inspect the displayed state. Keep this separate from erasure. Removing enrollment closes the current voting journey without deleting the Jira request or erasing Wezy data.
Acceptance criterion: the owner can explain how to stop new participation, interpret an unavailable state and request help with a sanitized description. Do not test irreversible erasure or uninstall solely to create a troubleshooting exercise.
Day 7: decide using an evidence sheet
Record each critical test as passed, failed or not tested, with account role, request key, observed state and owner. Avoid including private request content in a shared pilot summary.
Expand only when the essential access and saved-vote checks pass. Revise when the workflow fits but configuration or instructions need work. Stop when a core requirement is outside the documented scope, such as anonymous public voting. A quiet pilot with few participants can validate operation; it cannot establish broad customer adoption.
Frequently asked questions
Does seven days mean Wezy’s trial lasts seven days?
No. This is a suggested evaluation schedule. Check the terms offered by Atlassian during installation for your site and edition.
How many customer votes do we need for a successful pilot?
There is no universal total. Functional acceptance concerns correct behavior. Business validation needs a representative audience and an observation period appropriate to your customer base.
Should every passing pilot lead to a full rollout?
No. Compare the evidence with your original requirement, cost and team responsibilities. A technically successful test may still reveal that another workflow fits better.
Ready to run the evaluation? Open Wezy on Atlassian Marketplace and keep the official setup documentation available for each test.
Published by the team behind Wezy Customer Voting. Product capabilities are linked to their official documentation; worked examples are illustrative.
Let customers vote on your JSM portal.
Install Wezy from Atlassian Marketplace, open one request for voting and see the first vote in Jira.