JSM buyer checklist
How to Choose a JSM Customer Voting App
Evaluate a Jira Service Management customer voting app with a practical matrix covering access, organization demand, scoring, exports and trial checks.
In short
Choose a voting app using observable acceptance criteria. Start with customer access and the demand signal your team needs, then test scoring, exports and administration.
In this guide
Choose a Jira Service Management customer voting app by testing who can vote, what they can access, how demand is counted and how your team will use the result. A visible vote button is only the beginning. A useful app must fit your project type, customer model and prioritization process.
Disclosure: We publish Wezy and use its documented behavior as one concrete example. This guide is not an independent ranking. The acceptance criteria are intended to help you evaluate any candidate honestly, including deciding that native Jira or your existing product-discovery workflow is sufficient.
Write the buying requirement before the feature list
A useful requirement names an actor, an object and a decision. For example: “Signed-in customers must support selected requests they can already access, and agents must distinguish several votes from one customer organization from demand across several organizations.”
Compare that with “We need a voting platform.” The second statement leaves important questions unanswered: public or private? Customers or Jira users? Existing requests or new ideas? Raw popularity or organization demand?
If the desired activity is voting inside Jira, first review native work-item voting. If the decision belongs in a product roadmap, review JPD’s JSM feedback options. Avoid buying an app to solve a requirement that belongs elsewhere.
Use this evaluation matrix
| Area | Ask the supplier | Evidence to collect |
|---|---|---|
| Compatibility | Which hosting, project and account types are supported? | Your actual project opens the settings and customer flow |
| Request access | Does voting change sharing or show otherwise private requests? | An authorized customer succeeds; an unauthorized account cannot view the request |
| Vote lifecycle | Can customers add, inspect and withdraw their own support? | Saved states and corresponding agent metrics |
| Organizations | Is organization attribution optional, and who is eligible? | Multiple customers, shared organization and unattributed examples |
| Scoring | What enters the score, and who changes it? | An understandable calculation and activation process |
| Reporting | Can agents compare the dimensions they need? | A review using accessible, known requests |
| History | What coverage exists now, and what is retained? | Visible date coverage and a clear missing-history state |
| Exports | Which data is delivered under which permissions? | A small export that matches the selected scope |
| Administration | Who controls availability, recovery and erasure? | Documented roles and tested reversible controls |
| Commercial fit | Which edition and subscription cover this site? | Current Atlassian terms reviewed by the purchasing owner |
Treat a missing answer as an unresolved requirement, rather than giving the candidate an assumed pass. Do not combine every criterion into an arbitrary numerical score: customer data exposure cannot be offset by attractive dashboard styling.
Test customer access with real roles
Administrator access is not evidence that portal customers can use a feature. Prepare a customer who can access the request and a separate account that should not access it. Check the request type, participants, organization sharing and any issue-security restrictions before evaluating the app.
Atlassian’s portal-access guidance explains the distinction between open and restricted access. Your app trial should preserve the access design you intend to operate, rather than making everything visible just to obtain a successful demonstration.
For Wezy, enabling project rules and enrolling a request do not grant customer access. Customers must sign in, and the supported scope is company-managed JSM Cloud. See the documented roles and setup.
Separate people, organizations and scores
Use a hypothetical example: ten customers from one organization support Request A; six customers from four organizations support Request B. Raw votes and organizational breadth tell different stories. Neither automatically identifies the right investment.
Ask whether several customers from one organization create several votes but one represented organization. Ask what happens when a customer does not select an organization, changes attribution or withdraws support. Do not assume an organization selector means verified membership or request-sharing permission.
In Wezy, attribution is optional and eligibility depends on customer membership, service-desk association and project allowance. Standard includes customer votes, organization demand and a fixed demand score. Advanced adds weights and configurable scoring. Saving an Advanced configuration is separate from explicitly activating it.
The buying question is whether your team can explain the output. Ask a reviewer to describe why two requests have different positions. If they cannot, clarify the formula and inputs before using the ranking in a planning meeting.
Test reporting without mistaking missing history for zero
New installations cannot demonstrate months of history they have not collected. Wezy retains recorded analytical snapshots for 90 days in Standard and up to 365 days in Advanced, but available trends depend on actual coverage. Retention is not historical backfill.
During a short trial, check current demand and the history state you actually have. Confirm that exports match filters and access. Keep a note of the active scoring configuration so later comparisons remain interpretable. Also check whether Jira fields and search results have caught up before treating a temporary indexing difference as a lost vote.
Agree a small pilot and an exit decision
Choose a limited set of suitable requests and a named agent, administrator and customer tester. Use the pilot checklist to record pass/fail results, observed messages and follow-up work. Include withdrawal, not only a successful first vote.
Before approving a subscription, answer three questions: did the app collect the intended signal, did it preserve the access model, and will someone use the result in a real review? If the answer is yes, document ownership and proceed. If not, decide whether configuration, a different product or a simpler native workflow is the better fit.
For request-level voting in a compatible project, review Wezy and the available trial on Marketplace.
Frequently asked questions
Should we choose the app with the most features?
Choose the smallest workable solution for the defined need. Extra features only help if your team will use and administer them.
Is a successful agent demonstration enough?
No. Test the intended customer role and an account without access. The actual portal journey and request permissions are part of the requirement.
Should a demand score decide what we build?
No. Use it alongside impact, effort, risks and obligations. A request with fewer votes can still deserve priority, and a vote does not promise delivery.
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.