Troubleshooting
Table of contents
In short
If there are problems related to loading the editor, saving variations or tracking variations, this page lists possible errors and describes how to solve them.
1. editor or preview variants do not load
If you call up your website via the Varify.io dashboard but the Varify.io editor does not load, this is usually because the Varify.io script is not correctly integrated on your website, your website is using a content security policy or the necessary vairfy parameters have been removed.
Is the script installed correctly?
The easiest way to check if the snippet is implemented correctly is to open the browser console:
1. click with the right mouse button on an empty area of your web page
Select "Inspect" or "Inspect element".
Select the "Console" tab, type the text varify and press Enter.
If you see your account ID in the "iid" field and the value "loaded: true" is displayed, the script has been implemented correctly.
In the following example you can see a common error. In front of the included Varify.io script is another open tag. This has the consequence that the Varify.io script cannot be loaded correctly.
Content Security Policy (CSP)
Web sites can implement a content security policy (CSP) by using the HTTP response header Content Security Policy use or a <meta>-day with the http-equip Value "Content Security Policy. specify
ℹ️ Instead of blindly trusting everything a server delivers, the CSP defines the Content-Security-Policy HTTP header, which defines an access control list of sources with trusted content. It also instructs the browser to execute or display resources only from these sources.
Instead of blindly trusting everything a server delivers, the CSP defines the Content-Security-Policy HTTP header, which allows you to create an access control list of sources with trusted content. It also instructs the browser to execute or display resources only from these sources.
How to recognize CSP errors
You need to update your CSP to use Varify.io if:
a) you see an error like this in your browser console:
This page uses security features that are incompatible with the varify.io editor.
b) you see an error like this in Google Developer Tools:
Refused to execute editor.varify.io because it violates the following Content Security Policy directive...
c) You see an error like this:
Content Security Policy of your site blocks the use of 'eval' in JavaScript.
How to fix CSP errors
If your website uses a (CSP), the following values must be entered there:
- https://app.varify.io and
- https://editor.varify.io
In addition, the script-src must contain the directive "unsafe-eval" to be added.
In order to use the responsive device switcher, you must also provide your domain in both the frame-ancestors and frame-src directives.
Example
“`
Content-Security-Policy: frame-ancestors {YOUR_DOMAIN}; frame-src {YOUR_DOMAIN}; script-src 'unsafe-eval' https://app.varify.io https://editor.varify.io;
“`
Notice:
The unsafe-eval value must be set to allow custom javascript to be delivered by the variation.
Varify parameters are removed
When loading the editor or displaying the preview variants via the corresponding preview links, important parameters are appended to the URL.
If these are removed or blocked by your website, the editor or other functions, such as the preview links, may no longer work properly.
How to fix the error
So that no more important parameters are cut off from the URL, you should white-label them. The following parameters can be appended to a URL by Varify:
varify-editor
varify-token
varify-preview
varify-force
varify-mode
varify-variation-id
varify-variation-name
2. variants are not saved
You can make changes with the visual editor, but you can't save variations? Then it could be because the Varify.io script has not been implemented correctly.
Wrong Account ID
Check if the account ID in your snippet matches the your account ID in the Varify.io dashboard.
Varify.io script is not integrated correctly
See here
3. experiments cannot be started
If an error message appears after clicking on "Start Experiment" and the test does not start, there may be various reasons for this. As a rule, the GA4 Report Integration is active. Please check the following conditions:
- The authorization to create the audience in GA4 is missing for the connected account.
Solution: Have the authorization granted to the account or connect a Google account to the authorization instead. - The GA4 property already has 99 or 100 audiences and no further audiences can be created
SolutionDelete old, no longer used audiences from the GA4 property. - If 1. and 2. are not given please contact the Varify.io support and we will help you asap.
4. the Responsive Device Switcher does not display the website
If you would like to display and edit your website responsively, but your website fails to load in the responsive device mode, it could be due to framing restrictions.
Solution: Use of the Varify.io® Chrome Extension
So that you can still display and customize your variants for mobile devices, you have the option of bypassing this framing restriction with the Varify.io® Chrome Extension.
1. download the Varify® Chrome Extension
2. activate it in your browser and store your Varify® code snippet
3. activate the "Allow Responsive Device Switcher" switch
4. reload the page - Ready!
The Responsive Device Switcher should now work.
A more detailed explanation of installation and activation, as well as other functions, can be found here: Varify® Chrome browser extension
Customization of the X-Frame options on your own website
If you see an error message like in the screenshot above, a framing restriction is not allowing you to load the page in the responsive device switcher.
The normal desktop view of the visual editor is not affected by framing restrictions.
If your page is delivered with the X-Frame Options response header with the value "deny" you will not be able to use responsive editing.
A reasonable solution in this case would be to restrict to X-frame options: “sameorigin" .
X-Frame-Options: "sameorigin" allows your pages to be embedded in a frame from your domain, but does not allow them to be embedded from other websites.
5. tracking: no or missing data in GA4
Custom definition is missing or incorrectly named
Before you can receive data in GA4, it is necessary that a "Custom Definitions" has been created in Google Analytics 4. It is important that you follow the instructions GA4 Direct integration or GTM Integration follow.
GA4 Integration:
In GA4 integration, the user-defined definition must be linked to the event parameter "varify_abTestShort" be created.
GTM Integration:
For GTM integration, the user-defined definition for the event parameter "varify_abTestShort" be created.
It is important that the spelling is correct. Please pay attention to upper and lower case.
The event parameter can also be selected directly in the dropdown. However, it will only appear if you have previously selected:
- have started a first experiment (feel free to set a A/A Test on)
- There are test participants in the experiment, so data is also sent to GA4
- The first test data run in only after approx. 24 hours
6. varify.io - Test events in the real-time report
To check whether the Varify.io events arrive in Google Analytics 4 (GA4), you can use the real-time report. There are minor differences depending on whether you have connected your tracking via the direct integration or the Google Tag Manager.
GA4 Direct integration
1. Navigate to "Reports" and click on "Realtime".
2. Trigger a Varify.io event. For example, by calling up a page on which an A/A test is running.
You can now see directly in the real-time report whether the expected events are being recorded. In the "Event Count by Event Name" tile, after a few seconds the corresponding event "varify" will appear. You can then click on it to check the event parameter. Here you should now see "varify_abTestShort" be visible.
Google Tag Manager Integration
1. navigate to "Reports" and click on "Realtime".
2. trigger a Varify.io event. For example, by calling up a page on which an A/A test is running
You can now see directly in the real-time report whether the expected events are being recorded. In the "Event Count by Event Name" tile, after a few seconds the corresponding event "Abtesting" will appear. You can then click on it to check the event parameter. Here you should now see "varify_abTest" be visible.
7. pages Flickering - Flickering effect
This effect occurs because the A/B testing tool first loads the original version and then places the variation on top of it. During this transition process, there may be temporary changes in the design, layout or content of the page that are visible to the user. This leads to unwanted "flickering" or "flickering" on the user's screen.
The flickering effect can be frustrating and have a negative impact on the user experience.
To counteract this, an anti-flicker script can be used, which prevents the flicker effect for your users. You can find detailed instructions here: Anti-flicker script.
Visitors are redirected to other variants during a session / after accepting cookies
In this case, the CMP provider ensures that the browser's session or local storage is emptied as soon as tracking has been accepted. If this is the case, you only need to create and categorize Varify.io® or app.varify.io as a provider in the CMP tool. Here is an example of Cookiebot:
9. google analytcis 4 connection to Varify.io is frequently disconnected
It can happen that the connection between GA4 and Varify.io is interrupted, which is indicated by a corresponding error message. Such an interruption can have various causes, for example if the authorizations for the GA4 role of the connected account change.
The Google account can be reconnected at any time without any loss of data in the existing audiences.
If the account is disconnected regularly, for example every 24 hours, this could be due to a re-authentication setting in Google Cloud Session Control.
In this case, the frequency of re-authentication can either be extended or completely deactivated.
10. too many audiences in GA4 - experiment does not start
The number of audiences in GA4 is limited to 100. Since a separate audience is created for each variant of an experiment when using the GA4 Connection in Varify.io, an experiment cannot start if no new audiences can be created due to the limit. The following measures can help to solve this problem:
- Archiving of audiences in GA4: Audiences from old campaigns that are no longer required can be archived in GA4. Please note that no new visits can be assigned to these archived audiences.
- Archiving of past experiments in Varify.io: Experiments that have already been completed can be archived in Varify. This automatically leads to the archiving of the associated audiences in GA4. The data for the archived experiments is still stored in Varify.io and can be viewed.
Find out more here: Archive experiments