What will be the effects of anti-tracking policies on ad retargeting channels for e-commerce merchants? It’s a little soon to say, but here’s how things are shaping up.
The growing trend toward anti-tracking policies
Mozilla’s Firefox browser actually kicked off the anti-tracking policies trend. In June 2019 Mozilla announced that ad tracking would be turned off by default for all Firefox users. They could opt to turn it back on, but they had to dig a little to find the controls.
In June 2020, Apple announced that its newest IOS release would let users opt out of cookie tracking on their Apple devices. It delayed the rollout until April 2021 to give app developers and ad retargeting engineers time to adopt new, privacy-protective solutions.
And then in March 2021 David Temkin (Google’s director of product management for ads privacy and trust) shook up the e-commerce advertising world with a blog post announcing that Google was following suit. He declared that “once third-party cookies are phased out, we will not build alternate identifiers to track individuals as they browse across the web, nor will we use them in our products.”
Facing blowback from developers and advertisers, though, Temkin wrote in July 2021 that Google’s privacy changes will not take effect until late 2023.
Tracking cookies, like real ones, actually generate a lot of excess fat. 
It often happens that tracking cookies are missed when transferring to an adtech site. Because of this, cookie match rates range are only about 40 to 60 percent. This is why when you are searching for red Nikes on Zappos, you might get retargeted with ads for blue swimsuits.
Bad targeting from fragmented cookies both wastes advertisers’ money, and it annoys users with irrelevant ads. And cookies are device-based, so when a user moves to another device the cookies tend to crumble.
Cookies can also cause trouble with suppression triggered by a purchase. So you end up wasting money on retargeting people who have already converted on a different device.
Another important exception to note: the new anti-tracking policies for the most part are aimed at third-party data. They don’t affect the ability of publishers to track and target their own visitors. Apple and Google continue to support first-party relationships on platforms for partners. If there’s a direct connection with their own customer, they’re good.
Developing alternative strategies for ad retargeting
People-Based Targeting
The shift to mobile has been hampering the effectiveness of cross-device remarketing for a while now, since most mobile devices and apps don’t accept cookies. And cookies are device-specific. So when someone goes from their work computer to home or switches from desktop to their phone or switches between browsers, retargeting opportunity is lost. People-based advertising bridges that gap.
Introduced to the marketing world by Facebook, people-based advertising relies on a unique identifier that is related to the user, not the device. Facebook’s Pixel has a wide collection of tools for tracking user ID for ad purposes. But it can only track their activity on Facebook, Instagram, WhatsApp or any of their other acquired or not-yet acquired social apps. To get a full-spectrum view of behaviors on and off Facebook, you can use Google Analytics integrated with Pixel user ID data.
Contextual, content-based advertising
What’s old may be new again. Some in the retargeting industry are saying that contextual advertising is making a comeback. Advertisers still need to get the word out about products, and they still have ad budgets to spend. But many are expected to re-allocate budget and shift back into context-based advertising. After the advertising herd moved on from keyword targeting, they’ll be back in it.
Contextual advertising refers to ads relevant to the main content on the screen. If they’re reading a sports article, they’ll see ads for team paraphernalia. From the average user’s perspective, contextual ads are more relevant and less creepy than behavioral retargeting with cookies. They accept that ad algorithms know what they’re seeing on-screen at any given moment, but knowing everything about THEM breeds suspicion.
Marketing Land outlines the four key steps publishers and advertisers can use to develop alternative forms of targeting. For example, sponsored content is effective because everyone dislikes ads, but they do like to be informed and entertained. Relevant offers made within sponsored content are more appreciated, and prompt more conversions. See our two-part article on content marketing for e-commerce for more ideas.
Google labs at work
Google’s Temkin said the new privacy policy doesn’t spell doom for advertisers. Through “advances in aggregation, anonymization, on-device processing and other privacy-preserving technologies,” he says there are other ways to get the performance benefits of digital advertising.
Google is testing a new approach called Federated Learning of Cohorts (FLoC), in which ads are targeted at large groups of users based on common interests. He said Google will begin testing FLoCs with advertisers in the second quarter of this year.
As the dominant web browser with > 60% market share, Chrome’s new policies in their “Privacy Sandbox” will have a huge impact on digital marketing and advertising. It used to be said that “as General Motors goes, so goes the US economy.” Now, it’s Google rather than GM that’s driving the flow of commerce.
