Huge traffic drop after installing Cookiebot
I have originally posted in this topic: https://support.cookiebot.com/hc/en-us/community/posts/360012258433-Missing-data-missing-traffic-30-to-70- But I assume that since it is marked as "answered", it is not pushed to the top of the list. Since this issue has grave consequences for our and many other companies, I believe we need a more detailed / correct answer than the one given so far.
The post again:
It's hard to imagine that such an extreme drop is due to users not making a decision (therefore not being tracked). Here are the numbers:
In the 14 days from Feb 18 to March 2 we had 307 users on average per day on site 1 (as measured by Google Analytics). From March 3 we started using Cookiebot and *immediately* the number of users dropped sharply to 151 per day in the following 14 days from March 3 to March 16. The Corona measures only hit Germany hard from March 16, so this can’t be the reason. From the numbers we would have expected that in the “User consents” menu we would see around 307-151 = 156 opt-outs on avg. per day from March 3 to March 16. However, it only shows 81 on avg. per day in that timeframe. I would like to know why Google Analytics shows twice as many users lost compared to Cookiebot opt-outs.
While the former could, with some imagination, be attributed to users who did not make any decision, this cannot: On site 2 pre-Cookiebot we had around 300 users per day as per Google Analytics. After Cookiebot implementation we have about 90. That is a drop of about 210 users or 70%. Cookiebot, however, only reports about 25 opt-outs per day (ca. 10%)!
You also have to note that we use the banner which (annoyingly) stays at the bottom until a user makes a decision, so we assume most users make a decision at some point in their session.
-
Hi Marketing WBS,
Drops in traffic are not unusual when you give visitor the choice whether or not they want to be tracked.
Cookiebot will have an effect on how and when traffic is measured in your Google Analytics, mainly because of the fact that no data is registered for those users that ignore the cookie consent banner and leave the website without having either opt-ed in or opt-ed (it is not possible to register data and measure activity for those users).
You could find more about this here: https://support.cookiebot.com/hc/en-us/articles/360005476414-Why-is-there-a-drop-in-traffic-after-using-Cookiebot-
For example if the user opts in and gives a consent to statistics cookies, the visit will be registered in Analytics. But if the user opts out of statistics cookies, the visit will NOT be registered in Analytics.
You can see the number of opt in and opt out in the Cookiebot Manager though.
Learn more about this at the following link: https://support.cookiebot.com/hc/en-us/articles/360004223253?flash_digest=6182009ba060add7f341c03f9dd1c5cea3ae5a53
The consent method has also further impact on the analytics data because of the type of cookie consent banner you are using on your website. Using a banner with a high opt in rate, like the inline "allow all /selection" should give you a pretty good opt in rate, and thus more reliable statistical data.
If you are using a banner with the consent method 'explicit consent', the banner will stay on the website until the user submits a consent. This means, that if the user ignores the banner, it will not go away, no opt-in or opt-out will be made and no data will be recorded. This is what explains part of the "missing traffic" that some users report.
On the other hand if you are using a banner of the type 'active consent', the continued use of the website will constitute a consent. If you have enabled 'on scroll' for example, the user scrolling on the first visited page will constitute an opt-in and data will be recorded (if that opt-in includes statistics cookies).Regards,
Spas0 -
Dear Spas, thank you for your answer. I know all of the points you have raised, so unfortunately there are no really *new* leads here for me to follow and analyze. I guess I had hoped that the Cookiebot team will do a check on their backend to see whether the system works as intended.
However, you have given me the idea to check our bounce rate and/or new user rate on site 2. That a user who doesn't make a choice is neither tracked in Analytics nor shows up in the User consents statistics is old news to me, which I have always disregarded due to the fact that someone surfing the site would at some point definitely make *any* choice in the banner since the banner is so annoying. However: Site 2 draws in a big share of its users via ads and should not have many returning users. Ergo, a large share likely only views one page of the site and is thus not annoyed enough by the banner to make any decision. That means users not making a choice and therefore not being tracked could in fact be a relevant factor here.
I will analyze whether this is so and then get back to this topic. Thanks!
0 -
Dear Spas,
i would like to add a question, due to the fact, that we face the same issue with a huge drop of (tracked) traffic.
By using the consent method 'active/implied consent', is it still GDPR-compliant?0 -
Hi Daniel,
In oder to be sure that you are fully compliant Cookiebot recommends that you use explicit consent and no preticked checkboxes.
You can also use a banner with "allow all /selection" which will give you a pretty good opt in rate, and thus more reliable statistical data. Because all visitors will either opt-in or opt-out of all cookies.
You could find more detailed information by following this article:
https://support.cookiebot.com/hc/en-us/articles/360004413993-What-type-of-cookie-consent-banner-should-I-use-to-be-GDPR-compliant-Regards,
Spas0
Please sign in to leave a comment.
Comments
4 comments