The behavioral targeting technique is used in online marketing to adapt advertisements based on the user’s behavior. The relevant data are usually collected using cookies that are previously saved on the user’s computer after visiting a website.
Behavioral targeting is a popular form of targeting among advertisers, which is partially controversial among privacy advocates. For advertisers, this form of targeting can significantly reduce divergence losses.
A “corner store” can be used to describe how behavioral targeting works on a website. The customer is the online user. The website is the corner store owner who performs this type of targeting.
If the customer comes to the corner store, the owner often knows what the customer wants in advance. This is because he/she has known his/her customers for a long time and is therefore aware of their buying habits. Similarly, he/she also knows a little about their life and can therefore offer them appropriate products.
While the store owner keeps a profile of his/her regular customers in his head, websites and advertising networks work with cookies that either mark the corresponding device or IP address and record relevant data about the visit duration, bounce rates, viewed products, etc. If the user now visits other pages, where the advertisement program is also used, he/she is shown the appropriate advertisements that are adapted to his/her surfing habits.
Similarly, a shop can use the collected user data to better customize its range of products and, for instance, also send appropriate newsletters to its customers. Some programs, such as Sticky, analyze the user behavior on the website and can use appropriate coupons to entice users to return to the page.
Behavioral targeting is more effective if more data about the user are available. For this purpose, different user profiles are anonymously created and assigned to specific customer groups.
An in-depth analysis of the website’s visitors is usually done after the user profiles have been created. Common analysis programs, such as Google Analytics, can be used to provide very precise reference points. Larger multi-variant analysis methods, such as Log File Analysis, can also be used. Based on the individual profile, various channels can therefore be set up for which special landing pages, dynamic advertisements, etc. are created.
Behavioral targeting should provide a better user experience. Here, it is of utmost importance to ensure that users only find the content they are looking for. Advertisers often aim for an increased number of conversions. With increased spending for online advertisement, targeting of advertisement campaigns, or web content using behavioral targeting becomes a matter of cost savings. One goal of behavioral targeting can also be to minimize the CPA (costs per action).
Advertising networks and affiliate networks no longer work with manually created user profiles but rather collect data using the so-called Ad Servers and have self-learning systems that can create the profiles and set up different user channels. At the same time, tests are also performed in order to ensure users are shown the most appropriate advertisements.
The predictable behavioral targeting also uses collected data sets about the user behavior but, in addition, links the data sets with results from surveys or customer management systems. Here, the so-called big data can arise. These are massive databases that can contain additional socio-demographic information. In the end, algorithms determine the corresponding sample user based on this data set.
The advantages of this method lie in the fact that the marketing channels and themes can be linked with each other since the advertiser/advertising network is aware of the interests that the user can have in the current offer. Predictable behavioral targeting is thus suitable for cross-selling campaigns, etc.
Search engines, such as Google and Bing, also use forms of behavioral targeting. In particular, the web protocol and the Google account are, just like Gmail, powerful instruments in the analysis of user behavior and prediction of, for instance, what the user wants to search. Google also receives relevant data through Google Instant and Google Suggest. Here as well, the balance between the large benefit for users and the increase of advertising revenue through more targeted advertising is critical. Specific targeting takes place with the help of Google AdWords or Gmail sponsored promotions.
Like in almost every form of targeting, this method of customizing advertisements does not just help improve the user experience. Particularly when the advertisements or ad content are adapted to the needs of users without their consent and for laypersons, it is not comprehensible how this is done.
For data protection advocates, the massive collection of data by advertisement companies and groups is often a major source of criticism. The fact that user-related data can theoretically be easily combined with personal data is seen as a major invasion of privacy of Internet users. And although groups, such as Google, constantly claim not to collect personal data, e.g., through the anonymization of the IP address, online shops, etc. can do this when they combine the data of their customers with the actual user behavior from the log files and data from Web Analytics.
In addition to the discussion on data privacy, there often are active debates such as the one on the so-called “net neutrality”. Here, it is argued that the user should have the largest possible freedom, e.g., to visit all available websites that he/she wishes to. However, if search engines, such as Google or advertising networks, adapt the content to the user behavior, a part of the Internet diversity can be lost in the process since the results are filtered in advance.