ERC Research Projects

Richness and Accuracy of Online Consumer Profiling

Lennart Kraft, Klaus Miller, Bernd Skiera

Consumer profiles contain information about consumers which advertisers use to better target ads. Advertisers pay a lot of money to get access to consumer profiles while consumers are very concerned about their privacy. However, our study of 54,127 consumers on a large European ad exchange outlines that consumer profiles are relatively scarce, in particular for consumers who are tracked for a shorter period, and mediocre in their richness of information, especially for profiles that are young. The accuracy of demographic information is rather weak, though the accuracy increases over time.


Economic Loss of Cookie Lifetime Restrictions

Klaus Miller, Bernd Skiera

Over the past few years, regulators have begun to consider restricting a cookie’s lifetime as a way to protect consumer privacy. Most of this debate has taken place in the absence of any quantified cost-benefit analysis. To begin to fill this gap on the cost side of this discourse, we suggest a novel approach to estimate the potential upper-bound of an economic loss for digital publishers of lifespan restrictions on cookies. Our empirical study on cookies of 54,127 users who received about 130 million ad impressions over almost 2.5 years yields an average lifetime of cookies of 278 days (median 68 days) with an average value of €1.49 (median €.01). Only 21% of all cookies increase their daily value over time but their average value is almost three-times larger than the average value of all cookies. Restricting their lifetime to one year as the European Union proposes (two years as Google advocates) decreases their cookie lifetime value by 14.1% (6.8%), which represents a decrease in the value of all cookies of about 1.7% (.6%). In light of the €10.6 Bn. cookie-based display ad revenue in Europe, such restrictions would endanger each year €180 Mio. (€64 Mio.) or €.41 (€.15) per European Union Internet user.


The Impact of Consent and Transparency Requirements On Companies and Users

Julia Schmitt, Klaus Miller, Bernd Skiera

When processing user data, companies need to fulfill transparency and consent requirements by implementing privacy notices on their websites. The large variety of these notice designs poses the question of what the effect of implementing different notice designs are for companies. Using a large dataset of daily screenshots of over 7,000 websites, the authors first classify the privacy notices based on several characteristics, including visual information, content and clickable options for users. The paper then analyzes the effect these characteristics on the user engagement to find the optimal notice design for companies. The paper findings suggest that only meeting the transparency requirements is mostly optimal for companies while fulfilling the consent requirements has negative effects on user engagement for companies.


Economic Value of Online User Data

Rene Laub, Klaus Miller, Bernd Skiera

Firms collect a large amount of data by engaging heavily in the collection and storage of online user activity via various user tracking technologies. Recent policy initiatives aim at restricting this practice to protect consumer privacy. We study empirically the consequences of such restrictions for online publishers, such as news websites, because they strongly rely on revenues that are generated based on user data. We find a price decrease of ca. 30% for online ads when no data from user tracking is available. The potential revenue loss could be more than EUR 14 billion in the EU and more than USD 27 billion in the US.