Turn the UAT test cases you run by hand (work orders, purchase orders, inventory) into automated tests with the AI step writer, run them from your spreadsheet data, and keep step-by-step screenshot evidence. No developers needed.
3 design-partner places: a free 60-day pilot. Or start on the Free plan
A live demo on a representative Maximo flow, and what your IT team would need. Your own flow comes next, in a co-build session.
ACME-SUPPLY✓1004✓12✓6100-200-350Found by fallback: label "GL Debit Acct"!BMXAA4205I Record has been saved✓Everything in the AutoTestX platform, plus what Maximo’s lookups, generated ids and BMXAA messages need.
Paste the steps from your UAT spreadsheet. The AI step writer drafts them using the Maximo fields your saved tests already find (lookups, tabs and table rows), and learns your project’s wording as you keep tests.
Maximo regenerates ids and classes. The Generated ids healing profile trusts the label, heading and table row instead, and, when look-alikes can’t be told apart, AI healing suggests which one is meant, for a person to approve.
A BMXAA…E message classes the failure as a Maximo business rule, so nobody “fixes” a test for an application rule. AI diagnosis explains the rest in plain language.
Guide: from Excel test cases to automated tests, with a Maximo work order example →
Screens from our demo environment, with sample data.
Click through Maximo as you normally would. AutoTestX captures each step with several ways to find every field. When one stops working, the run is flagged before anything fails, and a renamed or moved field gets a proposed fix.
Attach an Excel sheet of test data. One test runs once per row: ten vendors, ten runs.
Run the whole set against your test environment from the app, or from your CI pipeline with the command-line runner.
Every run produces a report: what was done, with which data, and a screenshot of every step.
Re-run the whole regression set as one suite the day a fix pack lands in test.
How it works →Every automation script, new field or workflow change gets checked against the flows it could break.
How it works →Build the tests on today’s Maximo. They become the baseline for the full regression pass the migration needs.
How it works →Hand over a report with the data used and a screenshot of every step, not screenshots pasted into Word.
How it works →Lookups, tabs, frames and table rows: the parts of Maximo that break generic recorders.
GL accounts, vendors, items. Table rows are found by business key (the PO line for item 1004), not by position.
The recorder turns system messages into assertions: exactly what testers check by eye today.
When a step finds its field by a fallback method, the run is flagged warned before anything fails.
A small agent records in your tester's own browser, on Windows or Mac. Outbound HTTPS only, no admin rights.
A failure with a BMXAA…E message is classed as a Maximo business rule, so nobody "fixes" a test for what is really an application rule.
A plain-language view of every test, per-step screenshots and full run history.
Record the flows you already test by hand, read them as plain steps, and fix them yourself.
See what ran, on which data, and what failed, across every application team.
Catch regressions before a change reaches production, without a new platform to run.
Runs are deterministic: each step does exactly what is written, and no run acts on an AI answer. AI is optional. It can draft steps from plain text, and, if you turn them on, suggest a heal or explain a failure, always as a suggestion a person reviews. An organization admin can turn AI off altogether.
It heals, and you decide who changes the test. When a field can’t be found, AutoTestX compares the page with the last passing run, without AI. If one match clearly wins, the run carries on with it and the fix is proposed with its evidence; if not, the step stops and the fix is proposed. A person approves it, or, on Business and Enterprise, a rule your Test Manager sets applies safe fixes on its own. Risky steps like Delete or Approve stop instead of continuing on a healed field. An AI suggestion never lets a run carry on and no rule ever applies one. Every decision is kept in an audit trail and can be reverted.
The agent installs per user with no admin rights, and makes one outbound HTTPS connection: no inbound ports, no firewall change. It is not code-signed yet, so endpoint protection may flag it; the IT pre-check says what to expect.
For unattended runs, yes. Runs from the app, schedules and suite runs execute on the AutoTestX server in AWS Canada, so it must reach your test environment, for example by allowlisting its fixed address. Recording and watched runs use the tester’s own browser and work behind your VPN. The IT pre-check sets out the options.
The agent runs on macOS too. Until Apple has signed and notarized it, it installs with one Terminal command instead of a download. Runs from the app or your CI pipeline need no agent at all.
That is when you need regression tests most. Tests built now are the baseline for the migration’s full regression pass, so the comparison is ready when the move happens.
If it works well on your Maximo flows, keep it. If scripts need developers or struggle with lookups and tables, the Free plan or a pilot on ten of your cases is a cheap way to compare.
In 30 minutes: a live demo of AI drafting, healing and explaining tests on a representative flow, and what your team would need for a pilot.