Know if you're ready to ship — before anyone asks.

Hawzu is a modern test case management platform. Plan and organize test cases, run them, and track requirements, defects, and releases — with built-in analytics and AI across the whole testing lifecycle, in one place.

QA and engineering teams who want a clean system of record for testing — whether you test manually, run your own automation, or both.

Keep whatever you already run — Selenium, Playwright, Cypress, any framework. Hawzu reads JUnit, TRX, NUnit, Cucumber and Allure reports and maps every result onto your test cases. It manages and analyses your testing without owning your engine, so switching frameworks never means switching test tools. No script yet? Record the test once in Chrome and Hawzu keeps it as Playwright (TypeScript) and pytest + Selenium (Python) scripts you run like any other.

Yes. Point a test run at a Jenkins job, a GitHub Actions workflow, or a GitLab CI pipeline and press Run — Hawzu starts the build, watches it, downloads the report it produced, and maps each result onto your test cases. Pipelines that already run on their own schedule can push to Hawzu instead. See how CI/CD automation works.

By the test case code — stated in the report, or simply embedded in the test's name, like CHK-101 user can log in. Matching on names alone breaks the first time somebody renames a test, so the code always wins. Results that match nothing are recorded separately and never affect your pass rates or release readiness.

Yes. Point it at a requirement, a pasted spec, or a PDF and it writes full test cases with steps and expected results. It also maps the coverage first — up to forty one-line scenarios grouped by risk lens, with the requirements nothing was proposed for named on screen — so you see the gaps and pick what's worth writing before it writes anything. Nothing is saved until you accept it. See how a run works.

Give it your own documents. Canon is the set of specs, PRDs, standards and exit criteria the AI writes from — add a document and every generated test case names the document and section it used, so a reviewer checks a claim in seconds instead of working it out again. It's on by default and can be turned off for a single run. Canon reads documents you supply (PDF, Word, Excel, CSV, Markdown, text) — there is no crawler and no wiki connector — and when nothing in it matches a topic it says so plainly rather than filling the gap. See how Canon works.

Yes — that's Oracle. Type it the way you would say it, like automated test cases created in the last 10 days that have never been executed, and it answers with the matching records, a chart, an answer drawn from your own documents, or the right create form opened ready for you. Every condition it applied is printed above the answer in plain terms — "the last 10 days" shown as the actual date — and taking one away re-runs the answer instantly without using AI again, so correcting it is free. It reads your permissions from the same rules that guard each screen, and it never sees the contents of a record: it is shown what your fields are called and the values they hold, not your test case titles or defect descriptions. Meet Oracle.

Ask Atlas. It keeps a curated map of your application — areas, screens, behaviours, the routes between them and the journeys people actually walk — built from your test folders and the requirements you link, and optionally drafted from Canon for you to review. Pick the part that changed and it returns a worklist in three tiers, where every test case states why it is listed: it sits on the changed part, it's on a critical journey, or it covers a linked requirement. It's ranked in tiers rather than scored, because a number nobody can predict is worse than a bucket they can. Atlas does not read your code. See an impact answer.

Hawzu stands on its own — requirements and defect tracking are native, with a full Kanban board and a lifecycle linked to the failing step. If your team already lives in Jira, GitHub, GitLab, Azure DevOps or Linear, connect them: file the issue in your tracker as you raise the defect, or link one that already exists, and the defect shows its live status either way.

Yes — the board is the default view for a project. Drag defects between columns, group by status, assignee, severity, priority, release, or your own dropdown field, split it into swimlanes, and cap a column with a WIP limit that warns or blocks. Every drop is a real update, with the history, notifications, and tracker sync to prove it — and the same defects are one click away as a sortable table. See the defect board.

Yes, and it is the whole product — five people, three projects, no credit card and no seat minimum. Nothing is held back for a higher tier: the paid plans give you more people, more storage and more history, never more features. Growth is $20 per member per month billed yearly when you outgrow it.

Yes, and you run it yourself. The import wizard takes TestRail, Zephyr, QASE and Excel/CSV exports directly — map your columns, preview a sample, and it rebuilds your folder hierarchy as it goes. Coming from something else, send us the export and we'll map it with you. Nothing is a big-bang migration and no history is lost. Talk to us about migrating or compare Hawzu.

Role-based access control, audit logs, and scoped access tokens are built in, and your test data stays yours. See our Security page for details.