A Brief History of Third-Party Advertising IDs and Why They Are Going Away

The third-party cookie fueled digital marketing for decades. Cookies matched what brands knew about their customers to what publishers knew about consumer behavior to enable personalized ads.  Everyone has a similar anecdote or example about this — you browsed for a lamp, and then you saw ads for that lamp.

We’re now at the cusp of a new era. Consumers noticed bad actors misusing or neglecting their data, and started to question how their data was being used. In turn, this helped lead to government regulations, as well as browser changes. More than 50% of the internet is now cookieless, and soon it will be 100% with Chrome deprecating the third-party cookie this year. 

This naturally leads to: what are marketers supposed to do now? How are they supposed to know who to prospect and how to personalize their ads without those signals?

Before we can understand where the industry is headed and the best strategies for thriving in the post-cookie world, it’s helpful to reset and understand the context of how exactly we got to where we are.

In this post — the first of two — I walk you through the evolution of personalized media, starting in the 1960s (yep, you heard that right), and how the history of the industry has impacted the way programmatic marketers have achieved their ultimate goal: targeting the right consumer with the right message at the right time. Stay tuned for the next two posts, which outline tactics for achieving that same goal in a cookie-less future.

The Slow Rise of the Third-Party Cookie

Data-driven marketing isn’t new. Marketers have always known something about their consumers, and so have publishers. The two parties have been combining their data for a long time.

Take direct mail, for example. Back in the ‘60s, marketers were already leveraging and supplementing their first-party customer data with third-party datasets (sourced from magazine publishers, survey owners, and market measurement firms) to ensure each piece of mail contained ads pertinent to recipients' interests.

So when digital advertising came along in the ‘90s, there was even more hunger to take what was already a data-fueled industry to the next level.

At first, the market was heavily influenced by publishers. They leaned into context clues, assuming things like, “If someone reads Yahoo! Sports, they are probably interested in an REI ad,” and using those generally correlative predictions to sell sponsored ad placements.

In-house brand marketers began to notice patterns in consumer behavior, too. Things like consumers who bought item A in the past tended to buy item B in the future. But the data didn’t always line up.

Some consumers might look at a shoe three times, put it in their cart, and still not buy it. Clearly, the buyer had intent, so what was the blocker? How could they encourage that customer to purchase?

Connecting consumer intent, first-party data, and insights from thousands of different publishers was the key to unlocking those answers. And soon, there was a way to do it: the third-party cookie. Cookies were identifiers specific to a browser so that publishers and brands could not only build audiences based on the web behavior of their consumers but, most importantly, connect those audiences across sites — in layman’s terms, if you kept looking at the lamp, the brand could understand that and then use that to buy ads on a publisher you visited that evening.

Third-party cookies enabled connection across the entire internet - while third-party cookies were not created for this purpose, they enabled all parts of the digital advertising ecosystem to talk to each other. Brands gained an unprecedented level of insight into customer behavior and preferences and the ability to personalize content. Publishers, in turn, were able to better monetize inventory, and this innovation fueled the vast expansion of the internet over the 2000s. The unintended downside was that consumers often had little idea what data these companies had about them, or how they were using it. Sometimes, when these companies mistreated or lost control of this data, this put consumers’ information at risk.

With the rise of smartphones and mobile apps in the early 2000s, mobile advertising IDs (MAIDs) emerged to provide yet another layer of intelligence around customer behavior and intent.

For the next 15 years, marketers leveraged the powerful combo of these IDs to spin up highly advanced campaigns targeting (and retargeting) specific audiences with personalized messaging, making digital media the most precise and valuable marketing medium in history.

In fact, cookie data was so powerful that Index Exchange, one of the world’s largest advertising marketplaces, found that ad requests unable to tie a cookie-based ID to an ad placement received 99% lower bids than the same ad request with the identifier.

Take that in for a moment.

That means 99% of an ad’s value was in the cookie. Measured against estimated programmatic ad spend in 2023, third-party cookies created $552 billion in value for publishers and intermediaries.  This revenue powered content - publishers could hire journalists and pay those journalists to create more content for consumers, powering the free open web. In effect, the third-party cookie helped build and finance the modern internet.

In 2017, that would all change. The all-powerful third-party cookie was going away.

The Cookie’s Downfall

A confluence of several events caused the cookie to crumble.

The first — public mistrust around advertisers, publishers, and vendors’ use of their private data — came to a spectacular crescendo in the United States during the 2016 election and associated Cambridge Analytica scandal.

To put the significance of these events into context, let’s run through a bit of background. In the early 2000s, the Internet was an awesome engine that was getting even more awesome. No one asked questions about how or why it was getting better. And publishers never shared how they were making their money. The consumer was in the dark as to the value exchange that was occurring. 

We’d later see proof of that in a congressional testimony on the Cambridge Analytica data leak. US senator Orrin Hatch, who, presumably was briefed for the occasion, asked Mark Zuckerberg how Facebook makes money.

So when the Cambridge Analytica scandal happened, data privacy became a huge topic of conversation. How does Instagram serve me ads for things I never knew I wanted? Is my iPhone listening to me? And if it is, what conversations is it picking up? Who is listening to them?

In general, the public became much more sensitive to the types of data companies collected, stored, and connected to them individually. At that point, two things happened:

1. Governments stepped in. Once GDPR and CCPA were signed into law, nearly every other country adopted similar data privacy and security regulations.
2. Apple (very astutely) opted to protect the consumer. They eliminated all default tracking IDs across their devices and in their Safari browser. And they leaned into the privacy sales narrative. Billboards with “What happens on your iPhone stays on your iPhone” were plastered all over major cities.

Both of these events pushed other companies to follow suit. Mozilla Firefox launched Enhanced Tracking Protection (ETP) in 2019, and Microsoft launched its tracking protection in 2020. This put immense pressure on Google, and they finally announced that they’d sunset Android IDs and Chrome tracking IDs would be removed in a phased approach.

All is Not Lost

Although these sweeping changes create challenges for marketers and publishers, there are still ways to excel in this environment — but your strategy will need to change and fast. 

There are two main approaches marketers will need to adjust:

1. Personalization. Publishers are still selling digital media, but by continuing to operate using outdated approaches to marketing, you, as a marketer, don’t know how useful it is. By 2022, The Interactive Advertising Bureau’s State of Data report already showed 50 to 60% of programmatic inventory no longer had a user ID associated with it. If you don’t want to waste your ad budget, you need to find new ways to connect publisher inventory with your data.
2. Measurement. Third-party cookies helped provide significant functionality linking intent signals and conversion events, and losing them makes it harder to tie ad exposure to purchase behavior. You’ll have to find ways around that, too.

Want to know how to overcome these two challenges? That’s where we’re headed next. Read the second post in my series here.