JSM jql implementation guide
JQL for JSM Customer Votes: Build a Demand Review Queue
Build a Jira Service Management demand queue from Wezy vote fields. Handle empty values, stable sorting, field IDs and delayed indexing without inventing votes.
In short
Use Wezy's published numeric fields to shortlist customer-supported requests in Jira. Resolve your site's field IDs, keep missing values separate from zero, and validate the query against known requests before saving a queue.
In this guide
A customer-demand queue needs two things: a clearly defined candidate set and reliable published vote fields. With Wezy, query the app-managed numeric fields in Jira rather than trying to count portal clicks. Use the queue to select requests for review, not to replace incident priority or to decide what customers may see.
This guide assumes Wezy is installed on a company-managed JSM Cloud project. It develops the field example in the official Wezy guide into a testable queue design. The queries below are templates; they have not been executed against your Jira site.
Map the fields before writing the query
An administrator should identify the actual custom-field IDs in the site’s field administration. Atlassian supports the cf[CustomFieldID] alias, which avoids ambiguity when two fields have similar display names.
| Wezy field | Use in the queue | Important boundary |
|---|---|---|
| Wezy Customer Votes | Minimum current support | Active votes, not organizations |
| Wezy Organizations Requesting | Breadth across represented organizations | Unattributed votes do not add an organization |
| Wezy Votes Last 30 Days | Recent eligible support | First activation, not every click or edit |
| Wezy Demand Score | Sort by the project’s active formula | Not a percentage or a delivery commitment |
| Wezy Weighted Organization Demand | Advanced weighted context | Non-applicable in Standard is not a measured zero |
These fields are read-only app outputs. Record their IDs in an internal configuration note; do not create lookalike editable fields or manually alter the values to make a query pass. The product definitions and edition boundaries are in Wezy’s field documentation.
Start with one known request
Before saving a large filter, inspect a request on which an authorized customer has confirmed a vote. Compare its current Wezy panel with the Jira fields, then check that Jira search returns it. Vote confirmation, field publication and search indexing are separate stages; there is no promised instant indexing deadline.
If the request is missing, isolate the problem in this order: account access, project/key, correct field ID, published value, search result. This is more informative than repeatedly changing the threshold.
Create a narrow shortlist
Here is an illustrative policy: review unfinished requests in one project when at least three active customer votes are recorded. The threshold of three is a team choice, not a Wezy recommendation or a statistical significance test.
project = "YOUR_PROJECT_KEY"
AND statusCategory != Done
AND cf[ID_CUSTOMER_VOTES] >= 3
ORDER BY cf[ID_DEMAND_SCORE] DESC, key ASC
Replace the project key and both ID_... placeholders with your site’s values before using this query. key ASC gives tied scores a stable secondary order; it does not mean the earlier request is more valuable. Atlassian’s JQL keyword reference explains ORDER BY.
The query deliberately selects unfinished requests with recorded demand. It does not prove that a request is currently enrolled, belongs to an allowed request type or is open for new votes. Review those conditions in Wezy. Add your own validated request-type/status restrictions when the queue should match a narrower business scope; Jira work-item types and JSM request types are not interchangeable labels.
Save the tested search, then use the relevant JQL in a JSM queue through Atlassian’s queue configuration workflow. Keep the operational incident queues separate.
Give missing data its own investigation view
A field with no value is not the same as a field containing zero. Build a diagnostic query instead of silently converting missing values into zero support:
project = "YOUR_PROJECT_KEY"
AND statusCategory != Done
AND cf[ID_CUSTOMER_VOTES] IS EMPTY
ORDER BY key ASC
This finds requests requiring interpretation, not proven publication failures. Some may never have entered the intended Wezy workflow. For a genuine zero-value view, use cf[ID_CUSTOMER_VOTES] = 0 and compare it with the empty-value view. Atlassian’s operator documentation matters here: negative conditions do not necessarily include empty fields.
Do not apply a missing-weight investigation to every Standard request. Weighted demand is not applicable in that edition.
Do not turn a current field query into invented history
A numeric current-value filter does not supply a vote event history. In its custom-field operator table, Atlassian does not support WAS or CHANGED for custom Number fields. Do not append CHANGED AFTER to a Wezy numeric field and present it as a working way to find every new vote. Use the documented recent-vote metric or available covered history for the question those measures actually answer.
Check the score field separately when a request meets the vote threshold but has no interpretable sort value. Missing Wezy Demand Score is not proof of zero demand, and changing the vote threshold cannot repair a wrong score-field ID. Keep the request in the diagnostic record until its published inputs can be explained.
Validate membership with a small fixture
Use authorized test requests and record the observed fields. This is a proposed acceptance table, not a report of tests performed on a customer’s site.
| Fixture | Published active votes | Workflow category | Expected shortlist result |
|---|---|---|---|
| A | 2 | In Progress | Excluded: below the example threshold |
| B | 3 | In Progress | Included |
| C | 7 | Done | Excluded by the team’s unfinished-work policy |
| D | Empty | In Progress | Excluded; appears in the missing-data investigation |
| E | 3 | In Progress | Included; ties are ordered by request key |
Run the search with the actual reviewer role, not only a Jira administrator. A saved filter does not grant request access, and a wider sharing setting is not a repair for missing search results.
What a passing queue proves—and what it does not
A passing check means the query selects the expected accessible requests from the published fields at the observation time. It does not establish a complete customer census, a representative sample or a commitment to build the first result.
Keep the field map, query, test date, role and active scoring profile together. When scores appear inconsistent, use the score-change reconciliation procedure before editing the filter. When repeated review is justified, the Automation guide shows how to add a reversible review marker instead of sending repeated notifications.
For teams needing this signal on existing portal requests, review Wezy Customer Voting on Atlassian Marketplace. JQL and queues are native Jira capabilities; Wezy supplies the customer-demand fields used here.
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.