The Interactive Advertising Bureau has released for public comment a telling document that illustrates why Congress and the FTC need to develop some rules to protect consumers.Â Take a look at the definitions the IAB has embraced on targeting and data collection–and ask yourself.Â Based on what they say, can this really be–as the IAB claims–non personal information? Here are some of the definitions from the Networks & Exchanges Quality Assurance Guidelines [Feb. 2010]:
*Audience Targeting:A method that enables advertisers to show an ad specifically to visitors based on their shared behavioral, demographic, geographic and/or technographic attributes.Â Audience targeting uses anonymous, non-PII data.
*Behavioral Targeting:Â Using previous online user activity (e.g., pages visited, content viewed, searches, clicks and purchases) to generate a segment which is used to match advertising creative to users (sometimes also called Behavioral Profiling, Interest-based Advertising, or online behavioral advertising).Â Behavioral targeting uses anonymous, non-PII data.
*Attribute â€“ A single piece of information known about a user and stored in a behavioral profile which may be used to match ad content to users.Â Attributes consist of demographic information (e.g., age, gender, geographical location), segment or cluster information (e.g., auto enthusiast), and retargeting information (e.g., visited Site X two days ago).Â Segment or cluster information is derived from the userâ€™s prior online activities (e.g., pages visited, content viewed, searches made and clicking and purchasing behaviors).Â Generally, this is anonymous data (non-PII).
*Behavioral Event â€“ A user-initiated action which may include, but not limited to: searches, content views, clicks, purchases, form-based information and other interactions.Â Behavioral events are anonymous and do not include personally identifiable information (PII).
*Clickstream Data – A Clickstream is the recording of what a computer user clicks on while web browsing.Â As the user clicks anywhere in the webpage or application, the action is logged on a client or inside the web server, as well as possibly the web browser and ad servers.Â Clickstream data analysis can be used to create a user
profile that aids in understanding the types of people that visit a companyâ€™s website, or predict whether a customer is likely to purchase from an e-commerce website.
*Cookie â€“ A small text file sent by a websiteâ€™s server to be stored on the userâ€™s web- enabled device that is returned unchanged by the userâ€™s device to the server on subsequent interactions.Â The cookie enables the website domain to associate data with that device and distinguish requests from different devices.Â Cookies often store behavioral information.
*Cross-site Advertiser Analytics – Software or services that allow an advertiser to optimize and audit the delivery of creative content on pre-bought publisher inventory.Â Data can range from numbers of pages visited, to content visited, to purchases made by a particular user.Â Such data is used to surmise future habits of user or best placement for a particular advertiser based on success.
*Deep Packet Inspection – A form of computer network packet filtering that examines the data and/or header part of a packet as it passes an inspection point. In the context of online advertising, it is used to collect data, typically through an Internet Service Provider, which can be used to display targeted advertising to users based on previous web activity.
* Retargeting (or re-targeting) – The use of a pixel tag or other code to enable a third-party to recognize particular users outside of the domain from which the activity
was collected. See Creative Retargeting, Site Retargeting.
*Creative Retargeting:Â A method that enables advertisers to show an ad specifically to visitors that previously were exposed to or interacted with the advertisersâ€™ creative.
*Unique User â€“ An individual user that has interacted with online content, which is smaller than or equal to the number of cookies observed.Â The number of unique users to a website is usually an estimate.Â Â