Cross-Browser Testing with Testsigma: Setup, Execution, and Reporting

Cross-Browser Testing with Testsigma: Setup, Execution, and Reporting

Cross-browser testing means running the same test suite against multiple browsers and checking that your app behaves consistently. Testsigma handles the infrastructure — you define which browsers to target, and the platform provisions them in the cloud and runs your tests in parallel.

This guide covers how to configure cross-browser test plans in Testsigma, interpret results, and handle browser-specific failures.

Supported Browsers

Testsigma supports the four major desktop browsers:

Browser Versions supported
Google Chrome Latest, Latest-1
Mozilla Firefox Latest, Latest-1
Microsoft Edge Latest
Apple Safari Latest (macOS only)

For mobile web, you get:

  • Mobile Safari on iOS (iPhone and iPad)
  • Chrome on Android

You don't manage browser installations. Testsigma's cloud maintains updated browser versions across its execution infrastructure.

Configuring a Cross-Browser Test Plan

Step 1: Create a Test Plan

In Testsigma, a Test Plan ties together your test suites with execution configuration. To add cross-browser coverage:

  1. Go to Test Plans → Create
  2. Add your test suites
  3. In the Execution Configuration section, enable Cross-Browser Testing

Step 2: Define Browser Combinations

Testsigma calls these "Machine Configurations" — each one defines an OS + browser + version combination:

OS Browser Version
Windows 10 Chrome Latest
Windows 10 Firefox Latest
macOS Ventura Safari Latest
Windows 11 Edge Latest

Add as many combinations as you need. The test plan runs your suite against each combination, either sequentially or in parallel (parallel requires a plan that supports concurrent execution).

Step 3: Set Parallelism

Under Advanced Settings, configure how many machines run simultaneously:

  • Sequential — one browser at a time, no extra cost, slower
  • Parallel — all browsers simultaneously, faster, uses more concurrent session credits

For a suite of 50 tests across 4 browsers, parallel execution can cut total time from 40 minutes to 10 minutes if your plan allows 4+ concurrent sessions.

Running the Test Plan

Trigger from the UI: Run Now.

Trigger via API (for CI/CD integration):

curl -X POST \
  "https://app.testsigma.com/api/v1/execution_results" \
  -H "Authorization: Bearer $TESTSIGMA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "testPlanId": 42,
    "environmentId": 1
  }'

Poll for completion:

RESULT_ID=123
while true; do
  STATUS=$(curl -s \
    "https://app.testsigma.com/api/v1/execution_results/$RESULT_ID" \
    -H "Authorization: Bearer $TESTSIGMA_API_KEY" \
    | jq -r '.result')
  echo "Status: $STATUS"
  [ "$STATUS" = "SUCCESS" ] || [ "$STATUS" = "FAILURE" ] && break
  sleep 30
done

Reading Cross-Browser Results

The test plan results page shows a matrix of test cases × browser combinations. Each cell shows pass/fail. You can:

  • Click any failing cell to see the step-level failure with screenshots
  • Filter by browser to isolate all Chrome failures or all Safari failures
  • Download an HTML report for sharing

Patterns to look for:

  • Fails on Safari only → likely a CSS or JavaScript compatibility issue
  • Fails on all browsers → logic bug, not a browser issue
  • Flaky on Firefox → timing issue, consider adding waits
  • Passes locally, fails on cloud → environment difference (cookies, screen resolution, font rendering)

Handling Browser-Specific Failures

Safari Quirks

Safari has stricter cross-origin policies and slightly different JavaScript behavior. Common issues:

  • SameSite=None cookies require Secure flag — Safari enforces this strictly
  • WebRTC and some media APIs behave differently
  • CSS gap property on flexbox has patchy older-Safari support

In Testsigma, you can add conditional steps that run only for specific browsers using custom functions or test data conditionals. There's no built-in "skip this step on Safari" syntax, but a workaround:

If environment variable "BROWSER" is "safari"
  Skip verification of audio player controls
End If

You'd set the BROWSER environment variable in each machine configuration.

Firefox and Geckodriver Differences

Firefox uses Geckodriver instead of ChromeDriver. Some selectors that work in Chrome (particularly shadow DOM queries) behave differently. If you see failures in Firefox that pass in Chrome, check:

  1. Shadow DOM element access
  2. window.scrollTo behavior
  3. File input handling

Edge

Edge is Chromium-based, so test failures on Edge that pass on Chrome are rare. When they happen, check extension-related features or IE compatibility mode.

CI/CD Integration for Cross-Browser Testing

GitHub Actions

name: Cross-Browser Tests

on:
  push:
    branches: [main]

jobs:
  cross-browser:
    runs-on: ubuntu-latest
    steps:
      - name: Trigger Testsigma test plan
        run: |
          RESULT=$(curl -s -X POST \
            "https://app.testsigma.com/api/v1/execution_results" \
            -H "Authorization: Bearer ${{ secrets.TESTSIGMA_API_KEY }}" \
            -H "Content-Type: application/json" \
            -d '{"testPlanId": ${{ vars.TESTSIGMA_PLAN_ID }}}')
          RESULT_ID=$(echo $RESULT | jq -r '.id')
          echo "RESULT_ID=$RESULT_ID" >> $GITHUB_ENV

      - name: Wait for results
        run: |
          while true; do
            STATUS=$(curl -s \
              "https://app.testsigma.com/api/v1/execution_results/$RESULT_ID" \
              -H "Authorization: Bearer ${{ secrets.TESTSIGMA_API_KEY }}" \
              | jq -r '.result')
            echo "Current status: $STATUS"
            if [ "$STATUS" = "SUCCESS" ]; then exit 0; fi
            if [ "$STATUS" = "FAILURE" ]; then exit 1; fi
            sleep 60
          done

Local vs. Cloud Execution

Testsigma's cloud handles cross-browser execution, but you can also run tests locally against a specific browser using the Testsigma local agent. This is useful for:

  • Testing against a browser version not available in the cloud
  • Running against a locally deployed app (not publicly accessible)
  • Debugging failures with a live browser session you can watch

Install the local agent, register it with your Testsigma account, and select it as the execution target in your test plan.

Costs and Session Limits

Cross-browser testing is session-intensive. Each browser + test combination is a separate execution session. A test plan with 100 test cases across 4 browsers = 400 sessions minimum.

Testsigma pricing is based on concurrent sessions and run minutes, not per-test charges. Check your plan limits before setting up aggressive parallel cross-browser grids — hitting concurrent session limits causes tests to queue, eliminating the speed benefit of parallelism.

Summary

Testsigma handles cross-browser test infrastructure out of the box. You define browser combinations in test plan configuration, run them in parallel from the cloud, and read per-browser results in a matrix view. The main value: no Selenium Grid maintenance, no browser driver management, no separate BrowserStack subscription required. The main limitation: fewer browser version options than dedicated cross-browser platforms like BrowserStack or Sauce Labs.

Read more

Start now free