How to Create Custom Reports in GA4 using Mida's Data

In case you haven't integrated Mida with GA4 yet, please refer to this article: https://mida.freshdesk.com/a/solutions/articles/154000090813

As mentioned in our previous article, you can transfer event data from Mida to GA4 smoothly. This means you can analyze user engagement and website performance in-depth and generate more comprehensive reports within GA4.

Mida sends two types of events to GA4:

1. mida_pageview 

This event is automatically sent whenever a visitor who has been exposed to any Mida A/B test views a page. The event includes the following properties:

• id: Your test ID
• name: Your experiment profile name
• variant: Indicates the variation name (Control / Variant 1 / Variant 2, etc.)
• uuid: Mida's unique identifier for the variant
• timestamp: Unix timestamp of when the pageview occurred

2. mida_conversion

This event is sent whenever a visitor has converted on their primary Mida goal. The event includes the following properties:

• id: Your test ID
• name: Your experiment profile name
• variant: Indicates the variation name (Control / Variant 1 / Variant 2, etc.)
• uuid: Mida's unique identifier for the variant
• timestamp: Unix timestamp of when the pageview occurred

For simplicity, we will only be using id and variant to demonstrate the creation of a custom report.

Creating Custom Segments

1. Head to the Explore tab on the left panel.

2. Click on the blank exploration.

3. Click on segments.

4. Click create a new segment.

5. Then, click on user segment.

6. Click on add new condition then type in mida into the search box. You will see two mida events: mida_conversion and mida_pageview. We will create the segment for the total no.of users in the control first. As such, click on mida_pageview.

7. Search for id then click on it. A list of integrated experiment ids will be presented in the drop-down. Select the experiment id you'd like to create a report for.

Note: Always ensure your Google Analytics 4 integration is activated in the CONFIGURATION settings whenever you're setting up an experiment.

Note: If you're unsure what your experiment id is, simply head to the experiment dashboard and look at the URL. An example would look like: https://app.mida.so/experiment/13379/results

Note: If you don't see any experiment ids in the drop-down, you may have to wait 24 hours before you see any data (custom reports can only be created up to yesterday's date).

8. Next, we will select the variant type. Click on add new parameter and search for variant, then select control. It should look like below:

Give the segment a name and make sure to click Apply to save your changes.

9. We will now create the segment for conversions in the control. Hover over the control segment and click on the kebab menu (vertical 3 dots) to duplicate the segment, and change the condition to mida_conversion. It should look like this:

Give the segment a name as before, and click Apply.

10. You should now have two segments for the control - one for the total no. of users (mida_pageview), one for the no.of users who have converted (mida_conversion). Next, simply duplicate the control segments, and change the variant to variant 1 for both segments. You should then have 4 total segments. 

Congratulations! You've created your custom experiment segments for the control and variation. Next, we will focus on obtaining your experiment data using metrics and dimensions.

Extracting Experiment Data

1. Let's start with getting the total no. of users and conversions for the control and variation. Click on metrics.

2. Then, look for the Users tab. Click on it, and you'll see a long drop-down of metrics. The one we are interested in is Total users. Click the checkbox and hit Confirm.

3. You'll then see Total users under Metrics. Now, scroll down the Settings tab and drag it to Values. 

You should then see some data on the right side of your screen - in the example above, 120 represents the total no. of users in the control, while 11 represents the total no. of users who have converted in the control for our main goal. The same applies vice versa for the variation.

You can then add other metrics and dimensions that may be useful such as user engagement or first user source to get deeper insights. Feel free to explore the different metrics and dimensions GA4 has to offer to get the data you need!

Note: Be sure to check your experiment live date and your GA4 report date are synced up to get accurate data!