Tracking with Varify.io
Table of contents
In short
Under A/B Testing Options, you can specify which web analytics tool you want to use to evaluate your A/B test results. On the other hand, in the A/B testing setup, you determine whether the assignment of test participants to groups should only be valid for the duration of a session or whether it should be permanent.
Analytics Connection
Varify.io® provides the corresponding experiment and variant information. After a short setup phase, you can run your experiments directly in your web analytics.
Tracking with Google Analytics 4 (direct integration)
Should you use GA4 - directly without the Google Tag Manager - you can see test result data directly in GA4 without further configuration. The data is visible approx. 24 hours after the test has been started.
You can find detailed instructions with a step-by-step tutorial here:
Important: Important: The option can only be used if you do not use Google Tag Manager on the website. In this case, please use the "Google Tag Manager" option.
Tracking with Google Analytics 4 (Google Tag Manager)
Select "Google Tag Manager" in Analytics Connection if you use the Google Tag Manager on your website. The direct integration is then unfortunately no longer possibleas GTM uses a different DataLayer format than GA4 for direct integration.
You can find detailed instructions with a step-by-step tutorial here:
Note: You can also use the Google Tag Manager to send the data to another web analytics tool. If you have any questions, please contact Contact with us.
Tracking with Matomo
Matomo can currently be linked to Varify.io via the Matomo Tag Manager. Select "Matomo" under Analytics Connection and adjust the Matomo Tag Manager accordingly.
You can find detailed instructions here:
Important: Unfortunately, you can currently only run one experiment per URL at a time using this method. Otherwise there may be problems with the evaluation.
We are already working on a way to analyze several experiments at the same time.
If you have linked Matomo and Varify.io correctly, you can find out here how to analyze experiments in Matomo: Matomo - Results analysis
Tracking with Piwik Pro
You can set up the tracking of your experiments in just a few clicks using the Piwik Tag Manager.
You can find detailed instructions here:
Tracking with Snowplow
Currently only the Snowplow integration via Tealium is supported. Integration instructions will follow soon...
Tracking with etracker
The integration with etracker currently only works via the Google Tag Manager. It is set up in just a few steps using a custom JavaScript.
You can find detailed instructions here:
Data saving mode (beta)
Without "Data safe mode", the Varify event is triggered every time the experiment is played (Page and Audience Targeting return true). If "Data safe mode" is activated, the Varify event is only triggered once per visitor per session. This has the advantage that fewer "requests" are generated, which can lead to cost savings when using BigQuery, for example.
Wait for GA4 consent update (beta)
If the Varify event is fired in the DataLayer before the Consent Update, it can happen that no or hardly any A/B test data arrives in GA4.
Here it can help to activate the "Wait for GA4 consent update" mode. As soon as this is activated, the Varify event is held back until the consent update is initialized in the DataLayer.
This ensures that experiment data is only transmitted to GA4 if consent has been given.
The playout of the experiment is not affected by this, as this takes place independently of the "Wait for GA4 consent update" setting.
AB Testing Setup - Saving the group assignment
Configure whether the group assignment of your test participants should happen again after each newly started session or whether the group assignment should happen permanently.
By default, Session Storage is set. With Session Storage, the group assignment is repeated as soon as a session is restarted. This happens, for example, when the visitor closes the tab on your website or the browser. With Local Storage, on the other hand, group assignment is performed once. Test participants remain assigned to the initial group even when they return, as long as the test is active.
