Second-party Data Could Unlock Ad Targeting
Digital advertising networks searching for new ways to target individuals across websites and devices could turn to second-party data sharing.
Internet cookies come in two flavors. First-party cookies store session data and facilitate website personalization by “remembering” a visitor and her preferences. For example, these cookies keep users logged into a website over distinct visits.
Third-party cookies track individuals across websites and have powered performance marketing for years. These tiny bits of code lead to relevant ad targeting and, for many businesses, superior returns on advertising spend.
Such tracking cookies, however, justifiably raise privacy concerns. Hence many observators, regulators, and companies have collaborated to remove most third-party cookies this year. This includes leading web browsers such as Google Chrome, Apple Safari, Mozilla Firefox, Opera, Brave, and Arc.
Despite the privacy enhancements, eliminating tracking cookies is not entirely positive. Visitors will see many offers for products and services they are not interested in, while advertisers — including ecommerce merchants — will spend more on ads to generate the same revenue.
A solution could be second-party data sharing, adopted already by many display networks.
Let’s explore three implementation techniques.
Programmatic Email Advertising
Programmatic email ads use hashed email addresses to identify specific people and show them relevant ads — without cookies.
The example below uses one of these ads for TactiStaff, a military-style walking stick.
Programmatic email advertising already works without cookies.
The ad appeared in a daily email newsletter that offers sugar-free dessert recipes. That might not seem like a good place for a walking stick ad, but the ad was not aimed at the context. It targeted the subscriber.
The ad uses a simple HTML structure within the email: an anchor tag wrapped around an image tag.
Both the link and the image path include a hashed version of the subscriber’s email address as a parameter, making the process privacy-compliant.
When it tries to load the image, AOL mail, for example, calls the ad server, which includes that hashed email address. The ad server matches the hash to a known identity graph and produces a relevant, targeted ad.
This targeting works because the newsletter publisher shares its first-party data — the hashed email address — with the ad platform.
Informed Web Ads
Continuing with the recipe newsletter, assume this same publisher also shares data via the links to its own content.
The publisher appended each link with the subscriber’s hashed email address or similar identifier. If she clicked to read a sugar-free brownie recipe, that subscriber’s information would be passed to JavaScript on the website responsible for showing targeted ads.
The script would send the hashed email address to the ad server. The ad server would compare that hash to its database and deliver a targeted ad without a tracking cookie.
This data-sharing technique is a current and popular practice.
Active Logins
Another cookieless technique, less popular, for sustaining ad performance involves active logins.
The process requires three and sometimes four parties collaborating to deliver targeted ads: a publisher, an ad network, a community software provider, and an email service provider.
The individual shown an ad must have signed up with the publisher’s community. The parties share hashed email addresses or similar unique identifiers. This can be complex, but it functions as follows:
- An email service provider appends an email hash or other identifier to every link in every message its customers send — likely billions of emails.
- When he clicks a link and lands on a publisher’s website with the community software loaded, a subscriber is automatically logged in to the community based on the identifier.
- Once the subscriber logs in, the publisher shares first-party data with the ad server and generates a relevant, targeted ad.
Impact on Ads
Programmatic email advertising, informed web ads, and active logins are examples of advertising networks sustaining ad relevance and performance when cookies disappear. Advertisers using leading demand-side platforms may currently benefit from these approaches without knowing it.
Thus eliminating third-party cookies will disrupt advertising, but targeting is far from done.
