At some point, the browsers are going to unilaterally put an end to this debate about online privacy and advertising tracking. Now Internet Explorer has announced that it is going to automatically turn on do-not-track settings by default as of IE 10, which will ship with Windows 8 in the fall. I believe that we'll see the other browsers follow suit and possibly escalate this to help differentiate their products. I personally believe that the only problem with this is that IE has not gone far enough. That opinion can make me a little unpopular in some quarters.
I've gone on the record repeatedly about this issue, and I've laid out numerous reasons why third-party tracking without an opt-in is unethical. I completely understand the hardship that this would cause for third-party tracking companies, and I understand that many in our industry are concerned about the future of their businesses. I am not. I believe that there is a path through this that is both completely ethical and ultimately will significantly improve the results of online advertising.
At my core, I'm a product guy. And as a product guy, I essentially review the business scenarios that would be solved for customers through the use of a product, then work with engineers to innovate around those scenarios in ways that invent new technologies and intellectual property. This skillset is really about empathy (being able to put yourself in the customer's shoes) and logic (if this, then that). And since I'm not building a data company right now, I thought I'd lend my product expertise to the industry and make some specific recommendations to ad-tech companies in the data space, as well as to publishers. We can work together to reinvent the way that targeted advertising functions, in order make it both highly profitable and ethically defensible.
First: The distinction between first party and third party
First-party tracking and targeting takes place on web pages owned either by an advertiser or by a publisher. These are pages that consumers have decided to visit. In the case of consumers visiting an advertiser's website, they are there of their own volition and attempting to either research a product or service, or they're trying to interact with the brand in some way (perhaps reach customer support or find out about other products that the company offers). If the consumers are customers of that advertiser, they have an explicit relationship with the company. They can "vote with their feet" by not buying their products and not visiting their website.
In the case of consumers visiting a publisher, they've clearly come to consume content of some kind. In these cases, there's been a long-standing quid pro quo between the audience and the content owner regarding a trade-off between free or reduced-cost content and advertising being shown. Tracking of the delivery of ads is required in order for the publisher to sell the advertising; therefore, tracking is a true business requirement. Likewise, when a consumer visits specific categories of content (only on that publisher's site), the publisher can track this and then offer the consumer ads for relevant content on other sections of the site. This is the very genesis of behavioral tracking and targeting.
Third-party data companies don't have this direct consumer relationship; they track consumer behavior across numerous properties and follow the consumer around on these properties in such a way that a consumer can only escape their notice by opting out. (At least, that's the case today.) Because these companies perform a service across the entire industry to help publishers sell their ad inventory for a higher price, as an industry we've tried to protect them from being regulated. But because they don't have a direct consumer relationship, this has been hard.