Online advertising has been relied on cookies for far too long. Now its reign is coming to an end.
Cookies have been a mainstay of online advertising almost since the very beginning. They’re one of the go-to tools in the toolbox.
But their time seems to be running out.
As more and more users become privacy-conscious, the reputation of the cookie has been irreparably tarnished. So tarnished in-fact that browsers are beginning to block the kind of cookie that is the most useful
What is a third-party cookie? A third-party cookie is a cookie that is placed on the browser of a user by a third-party site. That is to say, a site that the user didn’t visit places a cookie on the browser. This usually occurs through some kind of pixel.
- The Many Uses of Cookies
- Why the Cookie is falling out of favor
- What will replace the cookie?
As the third-party cookie sings its swansong, the world of online advertising is adjusting to the emerging reality of a cookie-free future.
Let’s quickly examine the extreme importance of these tiny text files.
The uses of the cookie
Cookies have a lot of uses. They can be used to keep you logged into a site, or they can maintain your shopping cart after you leave. Beyond these little conveniences, however, cookies are also heavily utilised by advertisers.
And herein lies the problem.
With advanced third-party cookies and cookie tracking, users can be identified and tracked across sites. And that tracking potential only grew with the introduction of cookie-syncing.
Cookie-syncing is an important technology that allows different platforms to exchange cookie identifiers for various users. That way a platform can see user 342rez6 from CoolDSP and realize this is user 35654rrr in their own database.
This allows the various platforms to interact with each other and understand exactly which user is being discussed.
One of the primary uses of this kind of this effective communication and tracking is “retargeting.”
Retargeting is a very popular digital advertising technique whereby users that have converted in the past or partially engaged with a product are specifically targeted.
Even if you’ve never used retargeting in your campaigns before, you’ve almost certainly been on the receiving end of retargeting.
Have you ever visited a random website only to see ads for products you had in your cart elsewhere? Or even just products or games that you’ve viewed recently?
This is retargeting.
Essentially, the advertiser knows that you’re interested in these products and have at least partially engaged with them. So, by targeting you individually, the advertiser dramatically increases his chances of producing a conversion.
Of course, eCommerce is one of the simplest use cases of retargeting. The technique, however, has much broader, cross-industry appeal to advertisers.
A SaaS platform might retarget customers who are nearing the expiration of their contract term. Or they might target existing customers in the hopes of convincing them to upgrade.
Third-party cookies provide a means of following and identifying users even as they move from site to site.
These cookies, along with cookie-syncing, form the bedrock on which modern retargeting is built. That is, at least outside of the major walled gardens like Facebook and Google.
Cookie technology made retargeting and advertising to specific users particularly easy.
So easy, in fact, that it made users uneasy.
Why is the Cookie falling out of favor?
As consumers have become increasingly privacy-conscious, they have begun to take better care of their data. Or they are at least more interested in having some manner of control over with whom it is shared.
This movement toward privacy was born out, at least in part, out of new privacy laws such as the General Data Protection Regulation (GDPR) in Europe. Further, the major browsers
But as the stars cross for the cookie, new – perhaps even better – technologies are rising to replace it.
Cross-site retargeting has never been perfect. Without a universal ID, there is always data loss with Cookie Syncing. This can have even graver effects downstream in the cookie syncing system.
So users weren’t always recognizable to ad networks, DSPs, and other programmatic services that use user recognition to place relevant ads.
What’s going to replace the cookie is still something of an unknown; third-party cookies are still alive and kicking all over the web, even as their death knell rings.
Most of these technologies are not 1-to-1 replacements. They offer some similar tracking and targeting capabilities, but they are still different in scope.
One of the ways that cross-site advertising could work in the future and, indeed, partially works now, is through the association of a single user’s various accounts.
This way, behavior and preferences and data (that are shared or allowed to be shared across sites, that is) can be used to build a better profile of the user than one would be able to build just using the data from a single site.
This could allow for tracking and retargeting that is relatively similar to modern cross-site tracking solutions that are currently based on cookies and cookie syncing.
While this technique is on some level analogous to cookie-syncing, the other technologies differ greatly.
Device recognition/device ID
Of all the tracking systems that do not rely on cookies, user device ID is probably the easiest to use and the most accurate.
Each Android and iOS device has a specific ID that is connected to it. This is set when the device is powered on for the first time and generally doesn’t change for the lifetime of the device.
This ID is easily accessible to app developers and advertisers that use in-app advertising (it is generally unreachable from mobile web).
Since Device ID is constant and easy to access, it has become the go-to method for mobile user tracking.
Further increasing its value to advertisers and app developers is the fact that it isn’t nearly so maligned as cookies. While, in theory, device ID can be reset (therefore rending a lot of tracking completely useless), it is not so common. Most people don’t even know that Device ID exists, much less that it can be changed.
With cookies, on the other hand, practically everyone that uses the internet knows that they can be cleared – and many often do precisely that. This can wreak havoc on user tracking systems and render large swaths of carefully maintained cookie sync tables useless.
As users rarely, if ever, reset their device ID, this is hardly a problem for digital advertisers focusing on mobile.
The future is mobile
As a larger and larger percentage of digital advertising occurs on mobile (and most of that occurring in-app), the future is beginning to look, well, mobile. This transition is being born out in developing and developed countries alike.
Naturally, that means that the future of advertising will be equally mobile. As such, this tracking method will likely be the most important of all those mentioned insofar as replacing cookies.
Now, that said, smartphone operating system developers are already starting to try to limit device ID tracking to a certain extent.
But regardless, Device ID is here to stay and will likely remain the default way of tracking mobile users in the near and medium-term.
Of all the technologies listed here, behavioral recognition is by far the least developed – but it might have the greatest potential in the long term.
Behavioral recognition isn’t truly user tracking. Instead, it’s an application of behavioral monitoring, Big Data, machine learning, and artificial intelligence.
The idea here is to observe, understand, and sort user behaviors within a relatively restricted field of view.
So, a given website might not have a lot of access to actual, reliable information on a user, but it can track everything the user does on a given site.
What behavioral recognition promises is two-fold 1) It promises the ability to discern with relative certainty user attributes based merely on their on-site or in-app actions. 2) Determine which behavior patterns tend to occur before conversion.
One thing that behavioral tracking offers is the ability to A/B test what amounts to behavioral modification.
Once you have a good understanding of your user behavior, you can begin adjusting the content they see, the way your page or product is designed, and inserting different interactive elements that, through experience, you know to produce an effect that increases the likelihood of conversion.
Individual users with certain behaviors identified by your behavioral tracking and recognition system might respond differently to different stimuli. These stimuli could then increase (or decrease) the likelihood that they move onto the next step in the conversion process.
Once users are recognized and understood, the content that the site presents to them can be tailored to them individually.
This new, specific path can be engineered through machine learning to identify the ideal conversion path for each user then present that path to each user. Ideally, the result is more time spent on the site and a (much) higher likelihood of conversion.
There are, of course, other kinds of advertising that can be similar to behavioral modification. A clear example of this is Instagram Influencers. A company might identify an influencer who has a following that overlaps significantly with its intended audience.
The user himself might not want a particular watch or pair of shoes or have any interest in a given brand. But, once he sees someone that he wants to emulate wearing something from the brand, then he’s suddenly interested in buying it.
This pattern of inventing desire is different from traditional advertising. It is “influencing” what the person actually wants rather than just trying to present him with something he already wants.
Behavior tracking has limited applications for real retargeting
As you might have noticed, behavioral tracking isn’t going to be an adequate replacement for cookie-based retargeting. No matter how good the AI and algorithms get, there is no chance that they will be able to identify what users left in their cart at their last site.
So behavioral tracking is going to have more of an impact on user experience and conversion once a user has hit a landing page and begins his on-site conversion.
It will also be extremely useful for publishers as behavioral recognition will be able to, in theory, associate certain user behaviors on a site with user attributes (age, sex, etc.), thereby making targeting more effective.
But as a retargeting replacement, DeviceID and account linking will be significantly more useful for advertisers.
The cookie is heading for hard times, and the third-party cookie seems to be reaching the end of the line. The industry, however, is not quite ready to give up on the techniques that it has built on top of cookies and cookie-syncing technology.
That is to say, user tracking and retargeting/remarketing have become standard tools in the digital advertising repertoire. Indeed, these are some of the main reasons that advertisers are placing more and more of their total ad spend on digital with each passing year.