By now, you’ve probably heard that Google Chrome is doing away with third party cookies, you may also have heard that the most used browser is now delaying that decision to 2023.

It’s nothing new – Safari, Firefox and Brave did it years ago. Despite the delay, we felt it apt to start thinking about the future and what the changes will mean for you, and how to adapt to it.

There are plenty of things you might already be using third party cookies for, directly or indirectly:

• Frequency Capping in your display campaigns


• Targeting audiences with third party data


• Retargeting users visiting your website


• Tracking conversions, like sign-ups or first-time deposits

Without third party cookies, we can no longer do any of these things in the same ways. But we can find workarounds.

This isn’t the end of data-driven advertising. Far from it.

Digital marketers have lots of alternatives at their disposal, like:

• identity graphs


• household-level data


• first party data, including login data


• second party data


• consented third party data


• contextual data


• user cohorts.

So there’s no shortage of opportunities here. But there are some hurdles to clear too.

Some of the solutions we have are more mature than others. And new tools and vendors are popping up all the time. Which makes it tricky to figure out what you should be integrating into your digital marketing game, and how.

Let’s take a quick look at what options are out there, and how to use them.

First party data

This is some of the most valuable data you can get. It includes login data, like email addresses, and any other info your users give you directly. You can use first party data to build deterministic identity graphs.

The pros: It’s accurate, consent-based, and you can use it to train machine-learning models.

The cons: It doesn’t scale much and you have to be careful not to run afoul of privacy policies.

Second party data

This is data you get from sharing information with other publishers or operators (data collaboration). Data gets merged safely, anonymously and securely.

The pro: It’s very scalable.

The con: People might not know how their data’s being shared.

Identity graphs

An identity graph is a mix of different things you know (or can confidently assume) about your users. It includes things like IP address, device, device ID, email, location and website usage data. An identity graph stitches them all together to give you a general profile of each user.

Identity graphs will still be viable. But they’re mostly based on third party cookies, so some serious tweaks will be needed.

The pro: Without third party cookies, identity graphs may well be stronger and more reliable when they’re built on other identifiers.

The con: Building identity graphs takes a lot of resources for stitching the profiles together.

Household-level data

Household data is an old game. It was around in data-driven marketing long before digital concepts like cookies even existed. It works especially well for ‘full-funnel’ activities like brand building. But it also works for performance tactics.

The pro: You can combine it with offline activities, like post mailings, out-of-home or TV.

The cons: It only tells you about a household – not the individuals that make up that household. It also relies on IP addresses, which could prove a sticky wicket in future when it comes to privacy.

Contextual data

Contextual data is all about tracking outside influences that affect your users’ behaviour. Where are they when they use your app? What’s happening in the sport they’re betting on?

Like household data, this isn’t a new technique. In the pre-internet era, retail chains would factor in weather patterns or current affairs to better understand their customers’ shopping habits.

The pros: Contextual data is a good proxy for a relevant audience. It’s mature and privacy safe.

The cons: It’s harder to do on an app, compared to a website. Your reach only extends as far as the content you have available – if you want to target a Premier League fan, for example, there’s only so much Premier League content you can use.

Cohorts

With Google announcing the end of third party cookies, they introduced FLoC (Federated Learning of Cohorts). Google isn’t the only company working on this, but it’s probably the most prominent one.

Cohorts are anonymous groups based on common behaviour and interests. They’re not tied to individuals like cookies are. You could have a cohort called ‘sports fans’, ‘Arsenal fans’ or anything else, being as specific as you like. You won’t know which individual you’ve reached, but you’ll know what cohort they’re from.

The pro: It’s privacy safe, assuming the cohorts are big enough.

The con: This is a new idea. Google’s FLoC experiment (LINK) relied on cookies, so we don’t yet know how cohorts will work when they’re based on non-cookie data. But it’s an idea worth exploring.

Look out for more advice in future newsletters

A lot of these ideas are speculative, so there are many unanswered questions. Things that make sense on paper don’t always work in practice. And with new solutions turning up all the time, there’ll be plenty of developments to keep an eye on.

We’ll stay up to date with everything that’s happening and keep you informed with our future newsletters.