Protection of personal data, consumer consent and accountability are the main focuses of Bill C-27, with artificial intelligence (AI) also an issue that’s getting a lot of attention. AI is being used to create images, sound, video and text, but also has the ability to create data, something that’s of major importance to media and marketing. It’s used to create synthetic data and new databases from aggregate sources such as Statistics Canada to mimic what is seen in the real world.
Chris Williams, chief marketing officer at Arima Data, says data science is on the cusp of a third era built upon a synthetic foundation. He explains that it doesn’t mean that panels and first-party census level data goes away; instead, they become part of what gets fed into and used within a synthetic population for marketing. “As we talk to agencies and media vendors, they start to see what the implications are of how that data can be used, and it means they don’t have to do certain things around consent management. That issue goes away.”
Arima has created tools on top of its synthetic population database that can query the database, to target based on criteria such as women aged 25 to 54, and then find out more about them such as their education and household income, and also to investigate their media habits. “We’ve made it very simple to do this so it looks very much like tools that any media planner has used in the past. What’s different is that underneath it is a synthetic population,” Williams says.
Glassroom was the first media agency to integrate Arima’s synthetic data into its planning flow. Scott Stewart, general manager of the agency, says Glassroom has also made significant customizations to the application, specifically around how the agency approaches cross-media planning and how the application of synthetic data replicates what personal level data does.
“The big thing for us is that we recognized the value of synthetic data as a way to plan as discreetly as we wished, but also have the utmost respect for consumers’ personal information and the ethical use of data in the marketplace. You have 38 million Canadians, and you have 38 million Canadians as a synthetic society. What Arima has been able to do is integrate 23 different data sets to replicate that.”
Marketers can also merge their first-party data into the synthetic database, but to eliminate the privacy risk and consent issue, Arima can look at the patterns in the first-party data and create a synthetic version of it or any first-party data set.
Stewart says, “More work needs to be done around the use of AI as it pertains to digital marketing. A lot of the times, AI is used specifically to maximize performance, and personal information is often an afterthought. If we truly care about consumer privacy protection, then we do need more transparency around how AI is being used as well as good governance around conduct.”
