Martin Kihn is Senior Vice President of market strategy for Salesforce Marketing Cloud. During a recent Dreamforce virtual presentation (full presentation, log in required), he described the future of data and privacy in a cookie-less world, and offered 5 tips for how marketers can thrive in it. What follows are excerpts from Kihn’s Dreamforce presentation...
The background: Cookie-less is coming
Don’t forget, it's not just about the cookie. “Whether it’s third party, first party, or any party, the cookie is just a random string of letters and numbers that identifies the browser for a single domain. Ultimately, the cookie is a means to an end. So even when cookies go away, there’s going to be other means to the same end that might be even better.”
What marketers need. “Marketers need the ability to find and reach prospects and customers, i.e., addressability. We also need the ability to measure and improve our tactics, which is accountability.
The end of third party cookies is a kind of turning point - an existential moment. Cookies are just one part of a much bigger story, which is actually a major rethink of what's going on in marketing right now.”
A major rethink re: cookies and tracking. “In a recent version of its iOS operating system, Apple released the app tracking transparency framework. It requires an explicit opt-in for “tracking” for all app users from their store. The opt-in rates have been pretty low, so this diminishes the effectiveness of the IDFA or the in-app ID for mobile. Both Apple and Google have proposals now to obscure IP addresses from the header web pages. Apple also recently released privacy protection for email making open rates and deliverability harder to measure on email campaigns. So basically a lot of the identifiers that you've relied on as a marketer, both to target and measure at a personal level at a unique individual level on the open web are already in the process of changing.”
What's currently being considered to replace 3rd party cookies. “There's a whole lot of activity and debate happening now in forums like the W3C, which is the worldwide web community, the standards bodies for the internet, platforms like GitHub where engineers go to chat with one another about the future of addressability and accountability, targeting and measurement browsers, and apps. If you're interested in following any of these debates, just Google “privacy sandbox,” and you can read some of the proposals. Here are 3 of the ideas being considered:
- FLoCs. This is the most famous and it stands for federated learning of cohorts and it groups people into cohorts or groups based on their browsing behavior. And it's done by the browser. So it's using machine learning or algorithms.
- User Control (Ad Topic Hints). There's another more recent proposal that came from a Facebook engineer called add topic. This one would let people themselves explicitly indicate what kind of marketing they'd like to see.
- Interest Groups (FLEDGE). And there's a whole other set of proposals that would let publishers and brands put people into groups using tags. This is more akin to the way things work today. And that would help marketers create audiences for things like retargeting around product categories and interest groups.”
5 tips Marketers should be doing now
First tip - Don't panic. “The changes to cookies may sound like a big deal, but a lot of them are very gradual and you're going to have time to prepare. So keep using cookies for now -- and remember that we're all in the same boat, including your major competitors. Let's decide collectively to frame this in terms of what we're gaining rather than what we're losing. What we're gaining is a web where everybody knows the rules and consumers are comfortable with how their data is being used. We're going to be using groups and cohorts more, especially outside of our own channels, but this isn't new. Anyone who's done digital marketing for the past 10-15 years knows that in the past, this was how marketing was done.”
Second tip - Get comfortable with groups. “It’ll be more about segments and micro segments, rather than chasing the elusive dream of one-to-one marketing, at least for acquisition and upper funnel work. A big part of this is going to be better analytics. Now is a really good time to get your team to map out the differences in observed behavior between the type of people who opt in to tracking and the type of people who don't opt in.
This will tell you how, or even if, you can use data from users as a kind of mini sample for your user base. Think about how you would apply cohorts rather than one-to-one for both targeting and measurement across your campaigns. For example, retargeting in the future is going to be at a higher level, certainly not at the product level, but at the category level and subcategory level of groups of products.”
Third tip - Understand that measurement is going to be hyper-important. “Mixed attribution models are going to have to be more sophisticated. You're likely going to have to do more experimentation, hypothesis testing, to validate assumptions. You’re going to need advanced tools, like consolidated data for different platforms, so-called walled gardens (like Facebook, Amazon etc.) that are a growing part of media plans and that are very difficult to measure together.
Marketers who invest in data science talent, who invest in tools and do more testing are going to be more successful, they’ll see better results.”
Fourth tip - Rely on first party data. “First party data is data you get from your customers and prospects with their consent. First party data is absolutely critical to how you build trusting relationships. When you get better data, you can build better models and more personalization, you then get better engagement, which gets you more and better data. That’s a flywheel effect. Data tends to attract more data.”
Fifth tip - Rely on Customer Data Platforms/CDPs. “The Salesforce CDP customer data platform was released in 2020 and it's really taking off with our customers right now. It provides tools for first party data management analysis with the main uses being communication, privacy, insights, analytics for marketing, and beyond. CDPs are built to unify data from different MarTech and adtech systems.”
Reach out to us today to learn even more about Salesforce Pardot and/or Salesforce Marketing Cloud and how to thrive in a cookie-less future.