Speaker 1 (00:01):
Welcome to The Next Big Whatever, a podcast that discusses the current trends and how it impacts marketing. Good afternoon, JZ, and welcome to the The Next Big Whatever. Thank you so much for stopping by today.

Speaker 2 (00:16):
Thank you for having me. I really appreciate the time.

Speaker 1 (00:18):
So today we're going to talk about what's happening in the world of third party cookies, and there's been a lot in the press lately, specifically about cookies going away and what the future holds for us. So if we get started, let's talk a little bit about what are those third-party cookies and why is everyone freaking out?

Speaker 2 (00:45):
You know, I always like to start with a little bit of education, make sure everyone's on the same wavelength, but you know, a cookie is really just a small piece of data stored in a user's computer, uh, by the browser, right? Whether that's Firefox,Chrome, Edge, I mean, you know, a variety of browsers out there. Every browser uses cookies in a variety of ways to store activity. And really that's about understanding, you know, someone state, you know, did they click a button? Are they logged in? Really just functional pieces of the website and their use has really become pretty critical to the functioning of pretty much every website and what we've seen over the years is that the number of cookies has greatly greatly increased; I think, well beyond what most people could have ever dreamed of back when they were developed. So it's a pretty big shift. Uh, third-party cookies have certainly been around again that same, and have been used to track people across, uh, different websites, right? The, the beauty of a third party versus a first party is that a first party cookie can only really track you and understand your activities on the website that it's being placed on. Third-party can track your entire journey across the web. And so, you know, this change is going to significantly impact many, many companies out there.

Speaker 1 (02:14):
So for those of us who are not familiar with cookies and how we use them in everyday life, they are right the means by which we target most of the audiences and digital media, correct?o

Speaker 2 (02:26):
Yeah. You know, the cookies, uh, in a third party sense serves as really a, you know, unique identifier by those companies that are you using them to understand, you know, who you are, um, and, and what you do across the web. And they use that information, um, to understand whether you are maybe in market for our car, right. Based on your browsing history, um, you know, then they are able to attach data to you, right? If you're, um, you know, they're able to map like, say, you're, you know, you're a male or, you know, you're 35 years old or, you know, whatever that is. And so they developed this rich profile based on, in a year, identifier your cookie. And they use that either from an analytics point of view or selling you, uh, advertising, or, you know, the, probably a multitude of different ways, uh, in a third party sense, leveraging that sort of holistic profile. Exactly.

Speaker 1 (03:28):
And so there's a lot being written about Google. Google was the last browser to make this change. Um, they have said that the cookie will be going away in 2022. Um, talk a little bit about why you think it took them so long to make the change. And then also to like, after that, we're done with that, let's talk a little bit about their privacy stand box that they have, uh, decided to move forward with in, in place of the cookie.

Speaker 2 (03:59):
I would say that Google really took the longest because they have the most to lose, uh, when it comes to changing this sort of infrastructure level, you know, technology. And a lot of that is because Google makes money based off of advertising. Mozilla is a as a not-for-profit. So you basically donations and, and money that they get from having different search partners, uh, search revenue based on their browser and things like that. Of course, Microsoft has their browser. Uh, Microsoft does not make any money off of their browser, just a means. And of course, safari with apple, right? Apple makes money selling you devices and apps and things like that. But Google's revenue stream comes almost exclusively, uh, you know, 90 plus percent from the sale of advertising. And that advertising is based in large part or at least in a meaningful part, um, on third-party cookies and their ability to track and understand you across your entire web journey. And so that's the reason that they were sort of, you know, quote unquote behind, um, nothing sort of spurred this other than the companies themselves and, you know, consumer privacy becoming sort of a more out in front concern for most people. But, you know, they just had to figure out a lot because they have to not just change their browser, but the core of a lot of their products in order to adapt to something else.

Speaker 1 (05:32):
So they came out with the name or the acronym flock. I think it was a few weeks ago, tell people a little bit about what flock is and how that fits in their privacy sandbox. Sure.

Speaker 2 (05:45):
Flock is basically Google applying machine learning algorithms to their identifiers, their users, to figure out what group and cohort you belong into. Right? And so instead of having all this granular information attached to your browser and all that, they have really started with a beta of like, I think it's like a one to 3% of Chrome users that you will eventually be able to opt out of, but you cannot opt out of currently, but they will use that to inform whether the machine learning models are accurate compared to what they're seeing, of course, in the live environment with cookies today, right? Because cookies, third party cookies still exist. They're still highly functioning and run relevant piece of, of technology for Google. And so they're using their machine learning to start, uh, beta testing, whether the algorithms they've developed internally are matching or how closely they are matching what the third party cookies are currently providing.

Speaker 2 (06:51):
And I think, you know what we'll see over the next year or so. Uh, and again, Google has no legal requirement to deprecate third-party cookies on Chrome in 2022, they can decide at any moment to push that to 20 23, 25 or 2030. So, but if, if they come to a point where the machine learning is able to mimic and replicate what the third-party cookies are doing, that is when I think they'll feel comfortable in finally deprecating that, uh, knowing that they essentially can replace that knowledge from a marketing standpoint and targeting standpoint in their own, in their other platforms to, to target those users with, with relevant advertising. But of course now in a new privacy, more privacy centric way.

Speaker 1 (07:44):
So explain to people the idea of a cohort, because it's different than just segmenting people. And I think it's really important for people to understand what a cohort is and, um, how this is different from our current targeting.

Speaker 2 (07:59):
Yeah. Uh, a cohort is really, um, think about it as like a membership, right? Memberships are not a, a transient thing. They, they stick with you, uh, for, you know, some amount of time, right? You're a member of something there's always a subscription for some, you know, amount of time, whether it's a month, a year, you're right. Think of a cohort as the same thing. Now, sometimes when people develop cohorts, it is, it is permanent. You are a cohort based on, let's say your initial behavior. When you first came in and bought something from a company that could be one cohort that a person develops other times it could be something where it's more, you know, last three months, last six months kind of cohort. And that's something where, you know, in the last six months you've done something. And I think that's where Google is going to have to take that a lot on the cohort itself though.

Speaker 2 (08:57):
You know, if it's something like, uh, they're trying to do an in-market segment or in market cohort, that's going to have a much different behavior than let's say a cohort of a different variety, which is, let's say you certain browsing of content, right? You like a certain type of content. People tend to re re browse the similar content every single day. So your membership in that cohort is likely to be of much longer lens, just because your behavior doesn't change that often versus one type by car. I no longer in the market for a car. So you would hope that the cohort would eventually adjust myself out of that,

Speaker 1 (09:37):
But at this time, and granted, you may have read something, uh, different than me, but Google isn't really sharing. What's going to be in the cohorts. Is that your take as well? Like, do you think we're ever going to know what the cohorts are made up of?

Speaker 2 (09:52):
Uh, no, I, I, to be honest, I think this is, this is a bigger benefit for Google. It makes their data that they have on you, right. Via your login and things like that. It makes it even more valuable because they have the keys to the kingdom. They can understand that now with the privacy sandbox, supposedly you were able to identify with cohorts. Someone is in, but I don't think we'll ever really know much like SEO. We don't really know how Google's algorithm exactly works, unless you're probably one of the dozen people who can actually piece together the entire algorithm. I think the cohorts will work very similarly. So markers will have to, um, test and learn themselves into whether the cohort is continuing to be relevant for your business or not.

Speaker 1 (10:42):
So they're going to have cohorts trade desk. It said they're going to have unique identifiers. I mean, obviously people are, are looking to different ways in order to mitigate the going away of the third-party cookie on the different browsers. Um, what does that mean from a measurement and analytics perspective? I mean, how will this change, how we measure? I think in some

Speaker 2 (11:06):
Ways it's, it changes everything. And then in other ways, I think, well, structured measurement plan. It doesn't really change anything. You know, I think this idea of having a singular platform that can measure reasonably accurately, uh, users across a variety of platforms, uh, the entire customer journey, things like that, the, you know, in the past a lot of these larger attribution platforms, I think those days are gone. I think those, those we can, we can say we're essentially sunsetting that ability. And again, you know, if one company's the solution possibly is big enough, it still is possible theoretically, but I would say it's unlikely and it's, it's still going to essentially a panel at that point more than it is a true one-to-one across your entire customer base. But I think a lot of times when it comes to measuring business performance, the online data itself and digital interactions were really limited any anyway, I mean, unless you ran a e-commerce site, but if you were anything, but e-commerce, you always had to resort to tying online data to some sort of offline activity, right?

Speaker 2 (12:20):
Whether it's appointments or, you know, uh, products sold at the, at the store, those were always required. AB testing, split market testing, advanced modeling, which is aggregated aggregated data, right? Advanced modeling doesn't necessarily require that one-to-one data on hundreds of thousands of your customers. You can do it in an aggregated fashion and get a pretty good insight into what you maybe should be doing. And so, you know, some of the technologies like attribution will see themselves changed in a significant way, but things like marketing mix modeling, uh, we'll probably see a resurgence in popularity because really that's, what's kind of left in order to tie online behavior to offline, uh, in a scalable way. Again, all these companies will come out with their own measurement solutions. Live ramp is an example of that, where you can upload your customers to live ramp and they'll help measure, you know, and understand the overlap. It's certainly not causal, but the overlap between people who saw advertising digitally, and those who showed up in the store, that's the attempt at that. But again, their match rates will never be a hundred percent and it will never be as scalable as it once was with when third-party cookies were at their heyday, you know, a handful of years ago.

Speaker 1 (13:45):
So my takeaway from exactly what you just said, that I want everybody on this podcast to realize as that attribution will be forever changed. And we as an industry need to move away from attribution and to other ways of measurement. And I think that's really important for us in our clients moving forward, as I said, attribution will no longer be a viable measurement technique. It hasn't been for quite some time, but it's still been part of the conversation.

Speaker 2 (14:13):
Right? And the other thing that people to realize is that we're not necessarily replacing at this time. That was something new, um, marketing mix modeling has been around for quite a number of years, um, a whole generation longer than digital attribution modeling. And so, you know, what we're able to do is really rely on something that has proven itself time and time again, and its effectiveness and accuracy, and not have to worry about, you know, well, is the new thing gonna be as good as the old thing? Well, the old thing is actually what's new again,

Speaker 1 (14:46):
The whole industry it's incredibly circular. And it's really interesting to see that the shift is taking us back to some of those tried and true techniques that we had, you know, 10 years ago when we first started digital, um, before all of this happened. But let's talk a little bit about, I'm a client, you know, I'm worried about third-party cookies going away. I'm reading about it everywhere. You can't open up one of our trade publications or an email without seeing third party cookies are going away. How should I prepare? What should I do for this?

Speaker 2 (15:17):
I think it's, I think it starts with having a good plan. Let's start there. Right? We got to, and not just like a measurement plan or something like that, but a really solid acquisition or retention strategy for your, for your clients and your, um, and your business. And so, you know, from there we'll know that, you know, a lot of the one-to-one, uh, kind of targeting and things like that are going away, right. And so how do we prove effectiveness in a world where it's not as easy as just opening a report and being like, oh, this is it worked great. You know, how do we develop a testing strategy to know whether our new investments are actually paying, paying themselves off? Right. And we can work backwards knowing our new limitations. And I think if we do that in a systematic way, we'll come out with a really smart plan, probably the plan we should have had all along to be honest with you, but we'll come up with a really intelligent plan where, you know, we know exactly how, and when we're going to test things, then we'll know based on those tests, exactly what the ROI is in real time.

Speaker 2 (16:30):
You know, it's always the downside with, you know, the one downside with marketing mixed modeling is that it is still directional. Um, you still have to actually do something in market to understand whether it is 9%, 7%, 11% or somewhere in between. And so having this testing plan, having these activation plans that don't rely on one-to-one targeting, but maybe rely on contextual targeting, you know, rely on split market testing, rely on, you know, tried and true techniques that will get us the insights that we have always needed. But now we absolutely, you know, are cornered into getting one way or another. And then from there we can, we can take that and continually optimize the marketing plan moving forward.

Speaker 1 (17:18):
All right. So a lot of change, a lot of things on the horizon for us, but if somebody weren't going to have one key takeaway from our discussion today, what would it be?

Speaker 2 (17:30):
I would say the key takeaway is that targeting isn't going away, but one-to-one is not something that can be done at scale anymore.

Speaker 1 (17:40):
And that's really, really important for us because that means that moving into the future, we are going to have to look at different ways to reach the audience. And it may not be as direct and distinct as it is now. Correct. Absolutely. All right. Well, thank you so much for joining me today and talking all about cookies and those are not the cookies that you get at the grocery store. It's those third-party tracking cookies that drive a lot of how we target and measure online right now. Thanks so much.