A Guide to QA Testing Your Experiment (Without Preview)

Created by Ethan Sagan Chang, Modified on Mon, 17 Nov at 3:06 PM by Ethan Sagan Chang

Our users usually know about QA testing the display of the variant on their live site using the preview function.

However, if you would like to test how the variant displays without using preview, follow the steps below (or alternatively watch our video: How To QA Your Test: A Step by Step Guide):

1. Ensure both your computer and mobile phone are connected to the same Wi-Fi network.

2. Exclude your IP address (under the Targeting step) to prevent test contamination:

3. Set a URL parameter that ensures only you can see the test (make sure to save it!):

4. Set the traffic allocation to the variant temporarily to a 100% and publish the test:

5. Open the URL (with the URL parameter, e.g. www.example.com?utm_campaign=qatest) in an incognito window and/or a new browser on your computer/phone.

Note: If you use your mobile network to test (rather than the same Wi-Fi connected with your computer), you may be included as a user in the test. To prevent this, simply head to https://whatismyipaddress.com/ to extract your mobile network's IP address and exclude it in the test.

Note: You can restart the QA process and enter the test as a different user by using a different IP address (e.g., through a VPN).

Note: If you do QA a test with any URL parameter rules using preview, you must add the URL parameters to the end of the preview URL. For example: https://www.mida.so/?test-preview=17735&test-variant=Variant_1&utm_campaign=test

Note: If you don't see the variant, try the following:
1. Open dev tools (F12) then head to the Console. From here, type in 'mida.version()' to see if your experiment is showing the Control or Variant.
2. If you've still been bucketed into the Control, try opening the URL with another browser in incognito mode. You may also add 'mida-refresh' to your URL to clear your browser cache and Mida data. This will allow you to refresh and be eligible to be re-bucketed into a new experiment variant.
3. Double check the page targeting rules for your experiment to ensure that it's targeting the right page(s).

You may also use our free browser extension to see the Control or Variant on your live website: https://chromewebstore.google.com/detail/midaso/ejlcplfphenlembnohkibbfiebeeofmm?hl=en

When you're done testing and are ready to publish the test, simply set the traffic allocation back to distribute equally and remove the URL parameter.

If changes still need to be made and you need another pair of eyes on the test, simply pause the test and request a QA - we'll be more than happy to assist you!