Smart Collections

Smart Collections automate one of the most time-consuming parts of testing in Jira: building test cycles. Instead of manually selecting test cases and assigning testers for every regression run or sprint cycle, you define a collection once and reuse it to generate cycles with a single action.

A collection holds test cases added in two ways. Manual selection lets you browse or search your test case library and pick specific tests. Rules use the Visual search editor to define criteria that dynamically match test cases - dropdown selectors for fields, operators, and values with auto-complete make it easy to build conditions like "Tag = regression AND Module = Payments AND Priority >= High." You can combine both in a single collection: add a handful of critical tests by hand and let rules handle the rest.

The payoff comes at cycle creation time. BesTest evaluates every rule against the current test case library: new tests that match are included automatically, and tests that were deprecated or no longer match are excluded. Your regression suites and recurring test cycles maintain themselves - nobody has to remember to update the list as the catalog evolves.

Default assignees take this further. For each rule - and for individually added test cases - you can set which tester should be assigned to the resulting executions. When the cycle is created, BesTest fills in every assignee automatically: payment tests go to your payments specialist, security tests to your security tester, API tests to your automation engineer. For teams that run weekly regression cycles, this alone saves significant setup time every sprint.

Typical collections include a "Quick Smoke" suite (tests tagged smoke with critical priority) for hotfix verification, a "Full Regression" collection for major releases, a sprint-scoped collection matching tests linked to current-sprint stories, and module-specific collections for focused runs after one area of the product changes. Smart Collections also work alongside BQL (BesTest Query Language) for ad-hoc, JQL-like searching across all your test data. A step-by-step walkthrough of building your first collection is linked in the resources below.