Understanding the ID Problem In AdTech
When people travel internationally, their passports allow them to be identified by government departments in different countries.
If this form of persistent and portable identification didn’t exist, you’d no doubt experience unbearable delays at the border while the various countries scrambled to uncover your true identity.
Luckily for travelers, this isn’t the case. For AdTech, on the other hand, there’s a real big identification problem.
But what’s behind the identity (ID) issue in AdTech, how is it handled currently, and why is a new solution to the identity problem needed?
In this article, we reveal all.
What Is Identity In AdTech?
Put simply, identity in AdTech is the ability to recognize a user via a persistent ID.
Mobile devices have them via device IDs, making in-app mobile advertising less exposed to this problem. The same can’t be said for web browsers on mobile and desktops/laptops.
The only way for an AdTech company to identify a user as they browse the internet is via cookies, but as we’ll uncover in this article, identifying users via IDs stored in cookies is an ineffective solution to the identification problem.
Overall, identifying a user in online advertising is important for a couple main reasons:
Targeting: Knowing who to target and who not to target means the difference between an advertiser selling a product and wasting an opportunity (and money). This not only applies to identifying your target audience, but also knowing whether that user has been exposed to the same ad too many times within a certain time frame (i.e. frequency capping).
It’s also important for publishers to know who’s visiting their website, as they’ll be able to earn more money from their inventory and customize the user experience for their own benefit through content or product personalization.
Attribution: Did last month’s $10,000 display-ad campaign reach the right audience? Should we increase or decrease the budget? Answers to these questions rely heavily on knowing who viewed the ad.
The other side of the identity issue revolves around walled gardens.
Facebook’s and Google’s closed ecosystems mean that they know who’s who each time a user logs into their platforms. This allows them to offer advertisers a way to reach their target audience without the identification issue — something many independent AdTech companies can’t offer currently.
It’s clear that identity is important for everyone in the online advertising ecosystem, from the publishers, advertisers, and agencies who use it, to the independent AdTech vendors that sell it.
How Is Identity Handled Today?
In the beginning of the online advertising industry, targeting was mainly based on information passed on from the user’s browser and computer, such as browser type, operating system, language, etc.
With the introduction of real-time bidding (RTB), AdTech companies started using cookies (the third-party variety) as a way to collect data and create profiles that included information about a user’s web history and interests, among others.
These detailed profiles allow advertisers to display highly targeted ads to highly engaged audiences, but only if they can identify them.
The problem with cookies is that they can only be read by the domain that created them. So when an SSP creates a third-party cookie and stores an ID inside it, that ID can’t be accessed by DSPs.
This also means that if a DSP has previously created an ID for that same user, they can’t automatically match the ID sent from the SSP, as there’s no point of reference.
In order to solve this problem, cookie syncing was born.
As the name suggests, different AdTech platforms including SSPs, DSPs, and DMPs, sync each other’s cookie IDs. The benefit of this is that they can identify the same user, even though all their IDs are different.
The actual technical process that powers cookie syncing is quite complex and I won’t go into in this post, but what I will tell you is that it’s far from an ideal solution. Read more about cookie syncing here.
That brings us to the why.
Why Do We Need a Solution to Cookie Syncing?
Although cookie syncing helps solve the identity issue, it has three main drawbacks.
The first is that the match rates between cookie syncs are often not very accurate nor consistent.
Things such as cookie churn and the location of the two AdTech platforms involved in the sync can negatively impact match rates, which can range anywhere from 40% to 90%.
However, if those platforms then go on and sync with other ones in hopes of identifying the same user, the rate would decrease further.
The second reason is that cookies, especially third-party cookies which AdTech vendors rely on, are increasingly becoming more unreliable due to browser mechanisms that block third-party cookies (e.g. Apple’s Intelligent Tracking Prevention), higher deletion caused by browsing in private or incognito mode, and high adoption rates of ad-blocking tools.
AdTech’s reliance on cookies is becoming weaker by the day.
The third drawback is the impact on the user experience. As this process requires tags to load and fire in the background and for extra connections to be made between AdTech platforms, it increases page-load time.
According to research conducted by ID5, the most popular news websites in Europe have around 68 user-sync pixels per page, which take an average of 12 seconds to run.
The evidence for a new form of identification is clear, and luckily for the AdTech world, there are a few being passed around, including the Trade Desk’s Unified ID, the IAB Tech Lab & DigiTrust ID, the Advertising ID Consortium, and ID5.
These initiatives are not intended to compete with one another, but rather, work together and offer advertisers a choice when it comes to selecting an ID solution.
But just like with every other initiative in AdTech, the key to the success of these ID solutions is adoption.
Will the various independent AdTech vendors come together and solve the ID problem?
Only time will tell.
