Creating Test Cases with Architect

The Functionize Architect is used to create automated test cases by navigating through your website or application workflow. Architect eliminates the need for scripting in almost all cases, generating fully functional cross-browser test suites.

Powered by machine learning, Architect works best when it has access to as much data as possible. It leverages this data to interact with the UI during execution, producing robust and maintainable tests. Simply stated: Architect enables you to create tests with just a few clicks, even with limited technical experience.

Prerequisites

  • Functionize Account: You will need a Functionize account to create test cases.
  • Architect Enabled: If Architect is not already enabled for your account, your Team Admin or Site Admin can grant access.
  • Minimum System Requirements:
    • Processor: i3 or better
    • RAM:
      • 4GB on Linux
      • 8GB on Windows
      • 8GB on Mac
    • Note: If these requirements are not met, you may experience slower performance when using Architect. Test creation speed also depends on the complexity of the site under test.

Best Practices Before Using Architect

  • Architect is supported only in Google Chrome.
  • Chrome must be up to date.
  • Chrome Incognito mode is not supported.
  • For best results, create a dedicated Chrome profile with no other extensions installed (to avoid conflicts).

Installing Architect

  1. Go to the Functionize Architect Extension in the Chrome Web Store.
  2. Click Add to Chrome and follow prompts.
  3. Once installed, log in to Functionize to begin.

Alternative Path:

  • Log in to Functionize → Go to a Project → Click New TestArchitect.
  • If the extension is missing, you’ll be redirected to the Chrome Web Store to install it.

Launching Architect

  1. Log in to Functionize and select your project.
  2. On the Test Case Listing page, click New Test → Architect.
    • If missing, contact Functionize Support to enable Architect.
  3. In the Create New Test pop-up, complete the following:
    • Starting URL – enter or confirm your AUT (Application Under Test).
    • I confirm that no production credentials, PII, PHI, or other sensitive information will be used in this test case – check to provide your confirmation.
    • Start recording automatically – recommended if your AUT redirects from the initial URL.
    • Region – defaults to the nearest Google Cloud data center. You can override if necessary.
    • Orientation – Portrait or Landscape.
    • Desktop/Mobile – select platform (Desktop by default).
    • (Optional) Save these settings for future use.
  4. Click Start to launch Architect.

Recording a Test Case

  1. Architect opens with your AUT URL.
  2. If you didn’t enable auto-recording, click Record.
  3. Handle cookies/storage:
    • Architect may prompt you to clear cookies, HTML5 storage, or IndexedDB.
    • Choose whether to clear or retain data (note: retained cookies won’t persist in executions unless injected via custom actions).
  4. Recording begins with a Pageinit Action (launch step).
  5. Walk through your workflow step by step.
    • Wait for each page to fully load.
    • Click slowly and in the center of elements.
  6. When done:
    • Click Save,
    • Give the test a name, and
    • Select whether to save as a Test Case or Page Object.
  7. Tests are stored in your Functionize account and can be run across browsers.

Architect Options

  • Collapse/Expand Panel – toggle panel view.
  • Add Instruction – document your test with plain-language notes.
  • Insert Action – open the Actions sidebar to add steps.
  • Settings – configure recording preferences (see below).
  • Edit Action – click action dots to open details, then update and save.
  • Reorder Actions – drag and drop steps.
  • Delete Action – click the X next to an action.
  • Verify Action – add checks for unique/static elements.
  • Visual Check – compare page state with screenshots (Baseline vs. Step Comparison).
  • Hover Action – record hover interactions.
  • Pause/Resume Recording – pause recording with the red button, click again to resume.
  • Cancel Test – stop recording and choose from exit options.
  • Save Test – finalize your test.

Configuring Architect Settings

Architect offers settings to fine-tune test creation:

  • Faster Performance – reduces ML data collection during recording (not recommended unless testing very large pages).
  • Coverer – must be enabled for Double-Click actions.
  • Mouseover/Mouseout – control hover behavior recording.
  • Focus – prevents AUT from forcing focus away from Architect.
  • Long Text – removes 250-character input limit.
  • Force 100% Zoom – use if a site enforces zoom levels.
  • Enable AI Assist – on by default; gives guidance during recording.
  • Disable Service Workers – prevents caching issues that interfere with recording.
  • Enable Tab Keypress – adds Tab as a recordable action.
  • Enable SVG Children – allows interaction with SVG elements.
  • Enable Pointer Events – enables clicking elements with disabled pointer events.
  • Always Clear Browser Data – ensures a clean session before every recording.

Best Practices While Recording

  • Always start Architect recording before interactions.
  • Wait for full page load before clicking.
  • Add verifications after:
    • Page loads
    • State changes (e.g., login complete)
    • End of test (final verification is required).
  • Verify unique/static elements rather than containers.
  • Keep tests modular by verifying one element at a time.
  • Use descriptive names for steps and verifications.
  • Avoid hard-coded data; use variables or dynamic data wherever possible.
  • Only use stable selectors; handle dynamic elements carefully.
  • Validate actions with positive and negative checks.
  • Keep each step simple and focused.
  • Remove redundant steps for maintainability.

Summary

Architect makes it possible to build robust, ML-powered test cases without scripting. By following setup requirements, recording best practices, and using verifications effectively, you’ll produce stable and maintainable automated tests that evolve with your application.