Third-party cookies will go. When they do, what’ll happen to your marketing?
Our webinar on preparing your marketing strategies and measurement for when third-party cookies vanish brought many questions.
I want to share some of the frequently asked questions - here’s our take on it.
FAQ on web cookies
What's first-party data?
Unlike third-party cookies, first-party cookies are generated by the site you're on and track user behavior. These cookies help the browser remember important info about you, such as what items you add to your shopping cart, your username, password, and even language preference.
Other examples of first-party data are purchase history, demographic information, location, company, occupation, etc.
What's second-party data?
Second-party cookies are somewhat controversial. In general, second-party data is some first-party data shared between partners.
For example, an airline company could sell its first-party cookies (with data on consumers such as names, email addresses, etc.) to a trusted hotel chain to use for ad targeting. The transfer is done from one company to another that benefits both.
Want to learn more about first, second and third party data - watch the explainer video.
What's third-party data?
Third-party cookies are cookies placed by other sites onto a site you're visiting.
How does not having third-party cookies affect the data available to me?
First-party cookies will still be available to you. You’ll still have access to data about what your users do on your website.
However, tracking and targeting people across sites and applications won't be possible, mainly because data from 3rd parties won't be accessible. For example, Google will still have all user-level data to be utilized in advertising -- it just won't be available outside of Google.
It’s also likely that more granular data will be restricted. Right now, we don't know what level of granularity will be lost. But we should prepare to work on a more holistic or campaign level.
What's the best way to get first-party data?
Lead generation forms are great. With gated content, you can collect customer data while also offering value in return. Just make sure you remain interesting and relevant enough with your content to reach users.
Also, consider asking your customers for data. Think of post-checkout surveys, "How did you find out about us?" Promote users to become members or log in to an account so you can accurately understand behavior.
Our recommendation is to be open about what data you need and why you would like to collect it.
How does this affect my targeting and retargeting?
Audience size will decrease for specific audience types. It won't be the case for audiences built using data within one ecosystem, like a Facebook audience built using Facebook data. But for audiences built around 3rd-party cookies, that strategy won't be viable any longer. And for audiences retargeting users, we might find it harder to reach the people we want to with less cookie acceptance.
Therefore, it's essential to consider the move to contextual advertising to reach people in hyper-relevant placements. For example, show ads on TV-review YouTube channels for those in the market to purchase TVs.
Note that Google is looking into replacing this more user-level targeting with something they're calling "interest-based" advertising. Google's FLoC is creating cohorts of people with similar interests and providing them as an audience group to target.
For retargeting, we can't be sure or give the best answer. However, we believe retargeting within the Google universe will still be possible - not by cookies but by internal Google IDs. Other platforms will also have to work similarly.
How does this affect my marketing reporting?
Metrics tracking actions in a platform should see no change. That is, you can still optimize for clicks, CPCs, and CPMs in the channel.
What’s changing is the sharing of data across platforms. It means less demographic data, less audience information, and if users are converting on your site, it might be more challenging to feed this data elsewhere.
You might have noticed a change towards this with the recent iOS14 update. Now it’s harder to see specific breakdowns for performance. Also, you might find it harder to share data across platforms.
In short, marketers now need to adjust to more holistic and aggregated marketing data. And they need to be better at using their customer or CRM data to understand their audience and its behavior.
How does this impact my attribution reporting?
With the lack of granular data available and detailed user journeys, you'll need to switch from Multi-Touch Attribution models to a Marketing Mixed Model.
Multi-Touch Attribution is based on tracking-level data and user touchpoints, both of which will be impacted. Marketing Mixed Modeling, however, is based on aggregated ads data, your offline data, and any other external trends you want to add. You can read more about it in our blog post here.
How should I adjust which KPIs my team looks at?
It depends. Platform level KPIs should not see any impact (like CPCs, CTRs, CPMs). When you want to match that with events outside platforms, like cost-per-conversion, you might need to find another source of truth, such as your CRM or Analytics platform.
Also, you might need to abandon specific KPI breakdowns, like per country or region. Facebook and Google could optimize towards it, as the algorithms expected us to place more trust in them. But, they might not make that data accessible to us while we use their so-called walled gardens.
How do I ensure I have reliable tracking in place?
The only way to ensure your tracking is in place is to constantly test and quality check. Look at trends to find outliers or strange and sudden behavior changes; they might point to some technical issue (as behaviors tend to change gradually over time).
How does my team prove the ROI of advertising?
Attribution is a big topic and will become increasingly challenging as we can't track and tie together the full user journey. Identifying a source of truth is essential to establish consistency and a reliable, shared method to measure performance.
For some people, this is Google Analytics. For others, it’s their CRM. Once you’ve got a source of truth, ensure tracking is as best it can be. You’ll need accurate UTMs and cookies that are allowed to best attribute conversions within your source of truth.
Another key area to dive into is testing. You’ll want to understand the impact and correlation analysis. For example, “What’s the impact on conversions as you increase spend on display ads?”
How does this impact analytics tools like Google Analytics?
Google Analytics relies on first-party cookies for tracking and attribution, which won't be affected by sunsetting third-party cookies. But in GA, any cross-browser tracking or cross-device tracking can be misattributed (e.g., clicked an ad to the in-app browser and converted from Google Search on Chrome).
As web analytics is less affected, you need to ensure you’re packing as much first-party data as possible into your UTMs. Make sure your UTM data is clean. And ensure you’re packing information about your campaigns down to the ad level to help view and analyze performance in Google Analytics.
The likely main impact on Google Analytics won’t be due to 3rd-party cookie blocking, but an overall increase “in any cookie” and script blocking can be expected from major browsers and add ons.
As for how it works with your Google Ads attribution, Google Ads does integrate natively with GA. Meaning, the attribution for Google Ads campaigns doesn't rely solely on the 1st-party cookies and is more "aggressive" than other platforms. Consequently, more attribution credit is given to Google Ads than other marketing channels.
If you'd like to learn more, read here.
Is there any way to still understand my user journeys?
Understanding user journeys is already becoming increasingly challenging, especially the longer the journey is. The reason for this is cookies can expire or be removed after a certain amount of time.
The alternative to a full breakdown of user journeys here is mathematical modeling or correlational analysis.
Modeling involves taking what information we do know and using this to predict behavior where we don't have tracking. Google and Facebook are already working on this solution.
For marketers wanting to understand the impact of different channels on the final conversion, consider trying correlation analysis. For example, looking at how conversions were impacted when you increased spend in a specific channel.
How should I prepare for the death of third-party cookies?
Continue to build strong relationships with your customers from the data you own.
Find out what first-party data you’re collecting today and what information you’ll need to strengthen customer relations.
Ensure the data you collect benefits your customer, that you’ve got active consent, and that you safeguard their data and privacy.
Leverage first-party data to create personalized and highly targeted customer experiences. For example, welcome a customer to your website, give them recommended products and services based on their shown interest, and send them a cart abandonment reminder, etc.