The third-party cookies have been a hot topic for many months since Google announced its plans to phase out support for third-party cookies in Chrome and a lot has been written since then. After a bit more than a year since the announcement, we think that now there is enough perspective to have an in-depth summary about this topic.

How is it affecting me?

First off, let’s see how this could affect your current media investments:

a) Cookies just work in browsers, so all your media budget destined to mobile app inventory is not affected by the consequences of this announcement.

b) Third-party cookies are used to uniquely identify users for better targeting and reporting on web display ads, i.e. video and banners shown on websites. It does not affect your search ads (except GDN), nor your social ads investment/performance as these are mostly app-based.

So all in all, the phasing out of third-party cookies in Chrome will mainly affect your web display media activities.

Why is Chrome (and other browsers) doing that?

Us marketers have had a lot of fun before we reached this point, haven't we? We’ve flooded advertiser’s websites with little scripts sending data all around the web so we can have at our fingertips every tiny detail about the performance of the campaign and the user. We've bought, merged and targeted user's interests and affinities as we pleased - all this at the expense of the user’s privacy. 

We did it to a point where users were so frustrated that companies saw an opportunity on developing Ad Blocker browser plugins. It was exploited to such an extent that website owners didn’t even know where their user’s data was being sent. We went a bit too far and are now, rightfully, paying the consequences.

User’s privacy is a legitimate concern, it should be addressed and there is no reason why we can’t carry on with our job in the web ecosystem by learning from our past mistakes. It’s our opinion at Uptimal that we, the marketers and the industry in general should applaud this change towards a more private and safer web.

Cookie replacement

Chrome aims for a sustainable and more secure web advertising ecosystem. There are many online companies that depend on their advertising revenue to survive, so cutting them off, as other browsers did, wouldn't be good for the industry

Note that the importance of this statement relies on the dominance of the Chrome browser, owning 70% of desktop and 60% of mobile’s web activity. Therefore any change they make has a huge impact on the web ecosystem in general.

Being in that predominant position, Google has released a draft of their Privacy Sandbox project, a toolset to gradually replace third party cookies by putting the user’s privacy first. In this project there are many matters relating to third party cookies, however, we will just address those related to digital and specifically to media:

1. Ad Conversion Measurement

Cookies aren’t essential to publish an Ad, but if we want to measure its performance, right now, cookies need to be in place. Ad Conversion Measurement is the tool (or API) that will mandate how to track Ad performance without affecting user’s privacy on Chrome (and hopefully, to other browsers as well). For this to work correctly, this tool relies on a couple of specifications, or sub-tool (API's), to accomplish that:

1.1 Conversion attribution (conversion reporting)

We all agree that being able to answer questions like "did the ads served lead to conversions?" or "is this campaign providing a good ROI?" are essential to have effective marketing campaigns. Without it, we would be back to optimizing campaigns towards clicks instead of conversions. So how does the Google Chrome team see conversion attribution working with user’s privacy coming first? Check out the image below:

1) Ad loads on a publisher site where the click link contains custom attributes defining the identity of the publisher, the identity of the advertiser and the identity of the AdTech platform where data for the click will be sent:

<a href="https://advertiser.example.com/products.html"
  conversiondestination="https://advertiser.example.com" 
  impressiondata= "200400600"
  reportingorigin="https://adtech.example.com">
...
</a>

2) A user clicks the Ad and as long as above’s "conversiondestination" parameter matches with the domain where the user is being sent (the advertiser website), a Click Id (representing all the click’s data) is saved into the user's browser. This here is the fundamental change. Data isn’t sent directly to Adtech providers (i.e Facebook), instead, it’s safely stored into the user’s browser - the user owns the data.

3) Later on, a conversion happens and an AdTech pixel (such as the Facebook tracking pixel) is dispatched informing the AdTech Ad Manager's platform about the conversion:

<img src="https://adtech.example/conversion">

4) The Adtech Ad Manager server responds by requesting the user’s browser to store a conversion with a certain value (the number of products, some product characteristic or the event’s name).

5) At a later stage, the browser sends conversion data by batches to the Facebook Ad Manager server in an anonymised and aggregated manner. 

Realize that this entire process would allow us to have campaign’s data such as "the creative A got 200 clicks and 3 conversions" but no user’s data (no user id, no user browser, no IP, no user browser’s history, no nothing!), preserving that way the user's privacy. If you want to get more details about how this process works, check out the next video from the Google Chrome dev’s team:

As a final note here, it’s important to highlight that Conversion Attribution is still not able to track post-view conversions. More on this will be revealed in the coming months.

1.2 Aggregated Reporting  (reach reporting)

Conversion attribution is just giving us the post-click attribution conversions, but what about the reach? And what about features like frequency capping? Google is aware that this information is crucial for us to optimize our campaigns and the Aggregated Reporting tool (API to be more precise) accounts for this.

Let’s see how the Google Chrome team envisions this tool/API:

1) Ad loads on a publisher site - nothing new here.

2) At the very moment this occurs (a view), data about the impression is stored into the browser - again, very important: view data is not sent to the AdTech provider, instead is stored in the user’s browser. The user owns the data.

The HTML snippet that displays the Ad, besides having the anchor link shown in the Conversion Attribution section above, also has a few lines of javascript enabling storage of the view’s data in the browser:

const entryHandle = 
window.writeOnlyReport.get('campaign-123');
entryHandle.set('country', 'usa');
entryHandle.append("visits", "1");
entryHandle.reportAfter(2 * kMsecPerDay);
entryHandle.expireAfter(7 * kMsecPerDay);

3) Later on, after a still-unknown-amount-of-time, the browser sends the data to a central Aggregate Service. This Aggregate Service is still in a conceptual stage, but basically, it aggregates all the view data from all the user's browsers that have seen the Ad.

4) Once enough data is gathered from an Ad to guarantee its aggregation whilst maintaining the user’s privacy, the AdTech platform can request the report.

Like with the Conversion Attribution, this process will enable us to measure the reach of our campaigns, a very important piece of data, especially for branding campaigns. Regarding the frequency capping, the proposal is to offer it per-publisher instead of per-user. As this is still a proposal, it may vary over time.

Summary

So far we have covered how the Ad Conversion measurement (attribute clicks to conversions, campaign reach and frequency capping) would work in the very near future. Keep in mind that Google seeks the agreements of other Browser vendors (Safari, Firefox, Edge…) to extend this Sandbox Project to all these browsers. So do not be surprised if things change a little bit, although it's our opinion at Uptimal that the fundamentals are pretty well defined and will remain as they are.

No, we did not miss the Ads Targeting section, in our next post we will demystify how this important feature will work in a world after Cookie-geddon. Stay tuned!