Is Ad Blocking 10% Higher Than Commonly Measured?

A recent study by contentpass indicates that more than 25% of all ad blockers on desktop devices use the EasyPrivacy blocklist and are therefore invisible to common website analytics software.

Online advertising is delivered by so called ad servers which often also track website visitors across the internet. Most ad blocking software works by matching network requests and visual elements on a given web page against so called filter lists in order to block network requests to unwanted ad and tracking servers or to hide visual ad elements.

These filter lists are maintained by the ad blocking community and contain thousands of URL-patterns. A typical filter list entry looks like this:

||doubleclick.com^$third-party

This can be read as an instruction to block all requests to the doubleclick.com domain (Google’s ad server) whenever it is being called as a third-party, i.e. when doubleclick.com is not the main domain in the browser window.

The by far most popular filter list to block ads is the so-called “Easylist”. It is activated by default in popular ad blockers like Adblock Plus, Adblock or uBlock Origin and focuses on blocking ads both on a network- and on a visual level. Even the built-in ad blocker of Google Chrome uses this list.

More Than 25% of Ad Blockers are Concerned About Their Privacy

The same group of authors also maintain an additional list called “EasyPrivacy” which focuses on invisible third party trackers and is designed for users who are concerned about their privacy. While this filter list is technically compatible with most ad blockers, it is — in contrast to “Easylist” — active by default only in uBlock Origin but not in Adblock Plus or Adblock.

Because this list blocks all major analytics software (like Google Analytics and IVW), website visitors using this list are “invisible” to website owners and therefore there is little information available on how widespread the EasyPrivacy list is.

However recent numbers from our own research can give a clue about these invisible users: As part of our contentpass service we offer publishers a privacy friendly way of measuring, what percentage of their readers choose to block ads and unwanted tracking. We have written in detail about our measures to make this service as privacy friendly as possible.

Despite our efforts of building a privacy-by-design solution, the EasyPrivacy authors have chosen to block our service. While we still firmly believe that our software follows best practices with respect to protecting users’ privacy, we admit that the online publishing and advertising industry have mostly ignored user privacy interests in the past, with the latest revelations about Facebook and Cambridge Analytica being only the tip of the iceberg.

While EasyPrivacy users are now “invisible” to our service as well, we recently integrated our solution under the first party domain on a popular German IT news website. As a consequence of this first party integration the statistics about ad blocker usage were sent to a different URL, which was initially not being blocked by EasyPrivacy. It took about two weeks for the EasyPrivacy community to put the statistics URL of the first party domain on a filter list again.

These two weeks of unfiltered data allow us to get an idea of how many people use an ad blocker with EasyPrivacy activated (be it Adblock Plus/Adblock where the user manually activated EasyPrivacy or uBlock Origin where EasyPrivacy is activated by default).

Our data suggests that over 25% of all users with active ad blocking software on desktop devices use EasyPrivacy and are thus invisible to major web analytics software. In this specific case the true ad blocking rate on desktop was 37% while analytics software that is blocked by EasyPrivacy would only report what corresponds to 27% of ad blocking*. Or from a different perspective: 10% of the total desktop traffic on this website is not analyzed and counted by common third party analytics software. Historical data from the time where our service was initially added to EasyPrivacy suggests similar proportions on other sites and verticals.

The fact that a significant fraction of ad blocking users seem to be particularly concerned about their privacy is one of the reasons why we at contentpass have announced “privacy pass” which will allow users to browse publisher website without being tracked while still helping the publishers to sustain their services.

It is time for publishers to listen to their readers and offer privacy friendly ways to consume their offerings!


*Appendix: Correction of Ad Blocking Rates

At times where our statistics URL is blocked by EasyPrivacy we can of course not measure true ad blocking rates. However the numbers from the two weeks where our URL was not blocked allow us to correct the data from the other periods. Below we explain in detail how the data was corrected.

We assume that during the period where our statistics URL was not blocked by EasyPrivacy, the ad blocking rate that we measured was the true ad blocking rate:

P_true = (N_ad + N_priv) / N_total

Where N_ad is the number of page impressions (PI) where an ad blocker without EasyPrivacy was active, N_priv the number of PIs where an ad blocker with EasyPrivacy was active, and N_total was the total number of PIs irrespective of whether an ad blocker was active or not.

During the period where we measured the ad blocking rate with our URL being blocked by EasyPrivacy we were observing a skewed ad blocking rate since N_priv could not be measured by us at all:

P_skewed = N_ad / (N_total - N_priv)

Eliminating N_priv from the two equations above we can calculate the corrected ad blocking rate for the periods where N_priv could not be measured:

P_corrected = N_ad / N_total = (P_true - 1) / (1 - 1/P_skewed)

In the graph above the periods until 2018–03–19 and from 2018–04–03 onwards show P_corrected ≈ 27%, and the period in between shows P_true ≈ 37%. Web analytics software which is blocked by EasyPrivacy would report the skewed ad blocking rate which in our example would have been P_skewed ≈ 30%.