How do you balance personalization and privacy when targeting online audiences?

Increasingly, we’re seeing self-serve advertising platforms like Meta and Google use AI to find the right audiences for us. Whereas before you would need to manually set up campaigns for specific target audiences, new buying methods (such as Meta’s ‘Advantage+) removes this option. Instead, advertisers can upload up to 50 different ad formats and the platform will use machine learning to show different combinations of ads to what it believes to be the right audiences, and automatically optimise as it learns what’s working. This means the advertiser can focus more time and energy on creative and let the platform optimise in real time. A handy and time saving tool.

In the streaming TV industry, contextual targeting is proving to be a winning strategy. TV viewing is frequently a household activity and advertisers can't be sure who is in front of the set. But contextual, which targets viewers based on the type of programming that is being watched can solve that. If you are a DIY home repair brand you can safely assume that anyone watching a "home fixer upper" type show will be interested in your product. Better still, contextual targeting can be combined with traditional targeting to reach people in specific geolocations.

There's nothing creepier than the shoe that follows you around the internet because you accidentally fat-fingered one of their mobile ads. Brands need to be very cautious in terms of re-targeting and respecting consumer privacy. With television, we've found that consumers are somewhat less concerned about privacy because what shows they watch does not seem overly personal, but it's important to remember that they do still care and brands and streaming services need to be very transparent about what data they are collecting and how it is being used.

First I would like to start with the various types of advertising and their relationship to user privacy. -Contextual Advertising - Content of the advertising is targeted based on the website where it appears and may not know anything about the individual user. -Behavioral Advertising - The fundamental concept is the ad is served based on something that is known about individual user's behavior. Advertiser targets ads to users in certain demographic categories such as location, age, preferences etc. -Ads Retargeting - Advertiser knows a lot more about the individual user.

Meaningful consent and usability problem For example a user is logging into their bank account and looking at their statement. In this context, the user has clearly consented to this relationship with their banking service. My money is with this bank, this is a service for me to be able to view my current balance. This clarity is lacking for the user when those hundreds of cookie consent banners pop up.This brings us to the usability problem on the internet. Consent prompts appear left and right, interrupting the user flow and provide unclear expectations.Most users do not even understand the consequences of rejecting all cookies or accepting all cookies. They just want to get past this prompt in order to complete their task.

Global Privacy Control (GPC) and Universal Opt Out signal What we need is informed legislation putting guardrails around universal opt out signal. We really need the cross jurisdiction guidance and a method to scale GPC adoption across organizations. We need consensus around standardization for an organization to honor the Global Privacy Control. There needs to be provision for a ramp up period and clear implementation guidelines in the upcoming regulations. The consequences of not honoring a GPC need to be stated upfront by the regulators.