In a press release issued last month, the Office of the Privacy Commissioner of Canada announced it is holding public consultations into the ‘online tracking, profiling and targeting of consumers by marketers and other businesses’ this year. The consultations are part of the review process of the Personal Information Protection and Electronic Documents Act (PIPEDA). The office has opened up its call for submissions of comments or papers from interested parties who wish to contribute to the discussion, the deadline of which is March 15. The first consultation will be held in Toronto on April 29, 2010.
As the issue moves forward into the public sphere, Chris Williams, managing director, Media Contacts, says that the needs of Canadians also need to be balanced with the concerns of Canadians publishers, marketers and media professionals. In this op-ed piece, he proposes a middle ground built on the same anonymous profiling system that postal codes currently address.
I believe the term ‘behavioural targeting’ is a misnomer, since marketers use more data than just consumer online behaviour to target advertising. Behavioural data, in addition to other data, is necessary for the online advertising industry to function in Canada. Without it, publishers and advertisers would not be able to make the fundamental distinction between a Canadian and an international user.
This has two major implications, the first being that it would give US-based publications an advantage and the second being that it would mean the loss of a significant tool for the promotion of Canadian culture to Canadians in the online environment, a tool that other media enjoy. The result, I believe, would be a severely handicapped Canadian online publishing industry with no significant impact on Canadian’s online privacy.
Let’s look at the first implication: Canadians spend significant amounts of time consuming online content originating from US sources. In [Media Contacts’] own research looking at Canadians visiting US sites, we found that about 40.5% of Canadians’ time spent online, was spent on US based sites such as Yahoo!, MSN and Google (ComScore, January 2007).
Advertisers evaluate all these media with respect to their ability to reach the right consumer in the right environment and the right price. Without data-driven targeting abilities, the Canadian publisher must sell all of its inventory based on the editorial environment (i.e. a sports section) while the US publisher offers not only the editorial environment but is able to break that inventory down into smaller chunks based on guesses derived from data generated through browsing behaviour and other estimated sources.
In other words, not only is the US publisher offering a more targeted ad, the marketer pays less by reaching hundreds of targeted consumers instead of thousands of undifferentiated consumers. Marketers discount such types of buys, leaving publishers earning less money due to the fact that their inventory is finite. Given the choice between the Canadian publisher and the US publisher in an environment where data driven targeting is eliminated through legislation, the marketer will choose the more efficient and cheaper US publisher every time.
Favouring US sites with data-driven targeting options has rippling effects. US-based sites could service the Canadian market through US sales staff, eliminating sales jobs in Canada. Or worse, some global clients may go as far as to centralize their Canadian marketing spend at the global level, cutting jobs on the client side leading to a loss of jobs at the agency and then with Canadian publishers as well.
If the first implication doesn’t severely hurt the Canadian online content industry, the second implication removes the Canadian government’s ability to promote Canadian content creators. Data-driven targeting may enable a publisher to push Canadian content to more prominent positions on a page, however without the corresponding allowance for marketers to pay extra for premium targeting, the publisher’s incentive to support Canadian content will be diminished.
For instance, imagine the CBC is covering the Royal Winnipeg Ballet’s cross-country tour with stories and video on the CBC site. A person from Moncton maybe identified through data-driven targeting as likely being very interested in this editorial content, based on their geography and previous interest in ballet or the arts. Without data-driven targeting for the ad, this person is just as likely to see advertising for snowblowers (at $5 cost per thousand) as they would for ballet venue ($25 cost per thousand). This is not just a lost revenue opportunity for the CBC but also a loss to the Canadian cultural industry, because it is denied the ability to connect Canadians to Canadian content produced by Canadians.
If a marketer must keep an opt-in record for all of its data-driven targeting, even if that targeting is based on non-personally identifiable information (non-PII), then he or she might well ask for either personally identifiable information (PII) or the right to fuse non-PII with PII databases as part of the opt-in. This leads to a counter-intuitive erosion of privacy and a higher risk of accidental information exposure.
The need for balance between citizens’ privacy concerns and publishers’ and marketers’ targeting needs can be met in the same way that data is handled with postal codes. By having minimum target cell sizes, citizens are not personally identified and publisher and marketers can identify and isolate audiences who are seeking specific types of content. The current consumer ‘opt-out’ and data-driven targeting management options provided by some publishers are powerful and easy to use. Legislation that goes further than PIPEDA must be developed in coordination with industry experts so that the implications can be forecast before the law goes into effect.
Do you have an opinion on the Canadian mediascape that you’d like to share with Media in Canada? Or a response to this one? Contact us.
