Issue 1: Direct Traffic Increases, other Channels Drop; the Rate of Clicks to Sessions Drops
Cause: Allowing users to browse the website without making a consent choice
Sometimes, when implementing a Consent Manager, a decision is made to allow users to move through one’s website without making a consent choice. This normally means the tags are blocked from firing until the moment a user does grant a consent. In this case, the utm/gclid/other parameters will be lost as the user moves beyond the first page. Google Analytics uses the parameters on the landing page to determine the origin of the session: if they are missing, the session will be classified as ‘direct’.
There are several ways to remedy this, one of them being by making users choose before they browse. This solution is not without its challenges, those who take this route are met with a flurry of questions ‘why are users from Campaign X landing on pages Y and Z if there are no ads directing them there?’. Providing training and information about the expected behaviour can help overcome this issue.
Cause: Firing the pageview tag on the next page after the consent is made
Sometimes when Consent Manager is implemented users are indeed forced to make their consent choice before they can browse the page. Occasionally the pageview triggers are not adjusted correctly in the Tag Management System, which means that the pageview fires when the next page is loaded, rather than when the user clicks the ‘accept’ button. Because the URL parameters are only on the landing page, the information is lost by the time the pageview is fired – and the session is attributed to the ‘direct’ channel.
The solution to this is to fire the pageview tag on click of the ‘accept’ button. This way the first page will correctly register as the landing page, including the URL parameters, and the traffic will be attributed correctly.
Data Issue 2: Traffic from non-EU Countries Dropping Significantly
Cause: Failing to review Consent Manager behaviour in non-EU locations
There are multiple methods to implement a Consent Manager, and often things can go wrong. It is crucial to check that the implementation is working correctly, not just in the ‘opt-in’ countries, but also in the regions where consent is not required. If your tracking is triggered by a variable which is only present if a user opted into the tracking, you need to ensure that alternative logic is arranged for users who do not require this consent – otherwise you might miss on a significant portion of traffic. Many companies will already have VPN available internally, but there are also free VPN resources – so make sure you check the implementation from another region.
Data Issue 3: Nothing has Changed Before and After the Implementation
Cause: Implementing a Consent Manager but not reviewing the tags
If you’ve just implemented a Consent Manager and there were no changes to your data, we would recommend double-checking the implementation for issues such as tags firing before a user accepts the consent, or tags continuing to fire even if the user declines being tracked.
Consent Managers are often called ‘cookie managers’ – a term is used as a natural extension of ‘cookie notifications’. This naming leads many to think that it’s only cookies which need to be classified, which is not the case. If you implement a Consent Manager, your tags should also fire depending on the level of consent given. This is because GDPR doesn’t relate to cookies, but personal data – which can be collected by more means than just cookies.
Please note: the regulation is different between European countries, for example the French legislation does allow for firing of the tags before consent is given in certain cases. The GDPR is different from CCPA and other regional regulations, which is why many companies opt-in to install a third party consent manager to alleviate this issue.
Data Issue 4: The Traffic Dropped
Cause 4: Implementing the Consent Manager
This isn’t actually an issue, but many brands experienced a drop in traffic post Consent Manager implementation which can be alarming. If your traffic drops by 30%, this is normal – and what we found to be the average drop-out rate for our clients. This rate will vary by country and the type of website/business, but it is perfectly normal to have some drop off.
At the end of the day, you are offering users opt-out from something many are distrustful of – some will surely take this option!
—
This blog post was a result of a presentation given on MeasureCamp, by Incubeta’s Analytics Manager Monika Mesnage, and the discussion which ensued afterwards. Click here to watch the full recording; Implementing a Consent Manager.