Fewer cookies, more first-party data
It’s been looming on the horizon for Digital Marketers for some time now. The end of cookies.
Watch this 4-part series to get the lowdown on how it will affect marketing reporting:
Our Head of Growth, Juuso Lyytikkä, chatted with János Moldvay, CEO of Adtriba, Dr. Time Wiegels, VP of Data at Free Now, and Gianluca Binelli, Co-founder and CEO of Booster Box Digital. Together, these experts share advice about what you can do to prepare your marketing.
Why are third-party cookies disappearing?
When talking about a cookieless future, we've got to say this phrasing isn't entirely accurate. But, it's how people know it best! So just to clarify, when we refer to the death of cookies, we're talking about third-party cookies.
Janos and Juuso quickly cover what has been happening in the tech world regarding user data, cross-site tracking, and privacy (GDPR, CCPA) in the past few years. Janos gives a brief history and dives into the why of it.
First-party cookies, as far as we know and can imagine, are here to stay. These are cookies generated by the vendor that owns the website and provide a better user experience.
In the phase-out of third-party cookies, you'll still have access to all of your first-party cookies and data, and this will become critical data for you to feed into the ad platform algorithms.
Most web browsers like Mozilla Firefox, Safari have already made the move. The dominant Google Chrome, however, keeps pushing back the deadline for phasing out third-party cookies (it was March 2022, and now it's in 2023). We don't know when it will happen, but we've got a feeling it will come soon.
How will your attribution reporting be impacted?
Janos, our attribution expert, talks through what's changing regarding attribution reporting, what you should do to prepare, and recommendations for making the shift to Marketing Mixed Modeling.
As data granularity fades away with third-party cookies, marketers will need to shift their attribution models to be more privacy-friendly. And the way businesses collect data will change.
It means relying on your first-party data and available aggregated ads data, as well as other external factors, like seasonality. This aggregated data doesn't allow you to pinpoint and track a conversion on a user level. Still, it does allow you to create meaningful cohorts to help you attribute revenue to certain channels or initiatives.
What we need to start using or incorporating is MMM (Marketing Mixed Modeling). MMM includes external data (like seasonality) with your first-party data and your aggregated ads data.
Janos mentions the Adidas case where they could see not enough was being spent for brand campaigns. If Adidas can use MMM, so can you.
Janos recommends using both when you're starting, but the focus will be more on MMM.
How will tracking, targeting, and retargeting be affected?
Dr. Tim and Gianluca discuss how the loss of third-party cookies will impact tracking, targeting, and retargeting. Dr. Tim explains how the Free Now team is preparing and the importance of first-party data.
Tim and his team have an OKR to make all marketing decisions based on aggregated data. And it's all due to the third-party cookie phase-out.
Retargeting is going to be more complex and sometimes impossible. The approach to ad serving will change - simple as that! Ad platforms are working on alternatives for targeting in cohorts, but those solutions seem far from ready (like Google’s FLoC).
If a user doesn't allow you to track them in an app and block third-party cookies, then there's no way to serve a targeted ad to that person. So sadly, there isn't much you can do but depend on each ad platform and their black box to sell your products and services.
Some of his recommendations are to:
- Look at a higher level rather than at a user level.
- Do more baseline modeling and tracking. Ask "What would happen if I didn't do this campaign?"
- Do more brand campaigns and try being more local.
So do your brand campaign and check if you get more users, especially those you were expecting to target.
Testing, optimization, and reporting without third-party cookies
The guys tackle questions like, "How can I still test and optimize my campaigns?" or "How am I able to show ROI as a marketer (or as an agency)?"
Gianluca believes you'll start seeing more hyper-local advertising as data like device and location will still be available. Combine that with the first-party data you have from people visiting your website, and you can do some great campaigns.
One thing that will get harder for testing and optimization is A/B testing. It will be way more complicated because marketers will lose track of who is who and will need to rely more on the ad platforms for this information.
One of the biggest concerns many SMBs (Small-Medium Businesses) have is that they will start seeing conversions go down because they don't have enough data to feed each ad platform's algorithms. Some recommendations to combat this are:
- Feed all the first-party data you have into the algorithms.
- Have a limited number of campaigns and consolidate things as much as possible.
- Start optimizing for conversions at the top of the funnel. Then slowly test further down the funnel until you find the sweet spot that works for you.
- Use cohorts to look at the data. Find ways to visualize it so that it's meaningful for your team and accessible for them to draw insights from.
Since Gianluca runs an agency, we had to ask him how he feels the death of third-party cookies will impact his industry. He thinks that it's been a trend for years that the world is becoming more difficult for agencies. However, you never know how things will turn out.
Tim made one significant point that agencies could pool a lot of aggregated data to gather insights. And this could help companies struggling with having enough data to make decisions on optimizing their marketing strategy.
Gianluca believes three types of agencies will survive in the future: agencies that focus on creatives, ones that focus on strategy, and others that focus on data. And the ones that are focusing on data will probably be the ones that win.