BlinqIO Documentation

AI Recovery Case Study On Salesforce ​

This case study explains how AI Recovery helps maintain test stability when UI changes break existing test flows. It walks through a Salesforce test scenario for account creation, demonstrates how the test fails after a layout update and how AI Recovery adapts automatically without any manual intervention.

Existing Test ​

The automated test is designed to create a new Salesforce Account.

salesforce-cs-1

Failure Due to UI Change ​

After a Salesforce update, the form layout changes.

image-20250710154613852

The screenshots above show how labels and field structures changed in the updated form. These layout changes causes the test to fail when trying to locate elements using the original selectors.image-20250710160423731

image-20250710160458843

The Reports page highlights the failed step, showing the error message, root cause, and retraining trigger.

image-20250710160540775

Execute the Scenario with AI Recovery ​

When AI Recovery is enabled, the AI Test Engineer automatically detects the cause of the failure, regenerates the broken step based on the updated UI, and re-runs the test using the new logic all without any manual input.

Note

Refer Execute scenario with AI Recovery to learn more.

image-20250710160633126

This process happens in three stages:

1. Initial Run – Failure Detected ​

The test begins execution using the original recorded steps. It fails when it cannot find a UI element due to a structural change.

image-20250710170630862

2. Root Cause Analysis – AI Relearning ​

AI Recovery analyzes the failure. It inspects the updated DOM and surrounding context to understand why the original locator failed, then regenerates a new locator and updates the step logic.

image-20250710170734803

image-20250710171119782

Note

Refer Root Cause Analysis to learn more.

3. Final Run – Recovery and Success ​

The scenario re-runs using the AI-updated selector. The test proceeds successfully, and the step is marked with an AI Recovery badge in the report.

image-20250710184211617

The recovered step appears with a green status and an Error fixed by AI badge, indicating successful self-healing.

image-20250714173853421

This Salesforce case study demonstrates how AI Recovery enables your test automation to adapt to real-world changes in the application UI without requiring any manual rework. Even when recorded selectors break due to layout shifts, AI Recovery intelligently analyzes the failure, regenerates the required logic, and allows the test to complete successfully.

By reducing manual maintenance and preventing test failures, AI Recovery keeps your CI/CD pipeline resilient even as your application UI evolves.