Making the most of first-party data with Josh Perlstein

In this episode, Chris Mercer shares his Plan-Build-Launch framework to help you future-proof your marketing measurement.

You'll learn

  • What's first-party data
  • Why it's important to collect first-party data
  • How to qualify the data before adding it to your data warehouse
  • How to analyze and extract insights from first-party data

Subscribe to the Marketing Intelligence Show

Learn from Supermetrics' experts how to use data to fuel growth and maximize the ROI of your marketing spend.
Subscribe to the Marketing Intelligence Show


Anna Shutko:
Hello, Josh, and welcome to the show.

Josh Perlstein:
Thank you. I’m honored to be here.

Anna Shutko:
I’m very, very happy to have you on the show. Today we have yet another exciting topic. We’ll be talking about data, the definition of first-party data, and why this type of data is such a valuable brand asset. So like I mentioned, my first question to you, Josh, would be, what is the definition of first-party data? Can you please explain what it’s all about to our maybe less technical folks?

Josh Perlstein:
Absolutely. And first-party data is something that we’re passionate about at Response Media because it’s unique. It’s a type of data. And, well, for this conversation, I’ll define first-party data as data that a marketer owns. So often, people talk about first-party data and zero-party data. If you think about the context of first-party data as data a marketer owns, that could be behavioral data you’re picking up from your website or social presence or even on location in person kind of data, and you’re picking up behavioral data. But it also for this conversation will include what often folks will call zero party data, data that a consumer or a customer gives a marketer. So things that I offer about myself through a survey or a registration process.

And first-party data, if we look at it as widely data marketer owns, is the most valuable type of data that a brand can get. Much more valuable than data I’m buying or data I’m, let’s say, renting from some media provider or data provider. I can do what I want with it in most markets because I own it. There are a lot of opportunities for first-party data capture, as well as a lot of value behind first-party data now more than ever.

Anna Shutko:
Yes, this is indeed very true. Now first-party data is getting more and more valuable. So my next question here would be that you’ve briefly mentioned that the data is becoming valuable and that it’s much better to own your data than to rent it or to buy it. So what makes first-party data so valuable to a company? How can its value increase the brand’s value overall?

Josh Perlstein:
That’s an interesting way you ask that question because, inherently, first-party data has value because you own it, and you have the opportunity to do what you want with it. But it has more value now than ever in the past because it becomes the key to unlocking someone’s identity, whether a customer or a prospect. When you capture information from declared first-party data, someone gives you their email address, for example, and consent to use it. You can leverage that email address and use it to identify that person across devices across the entire web where someone authenticates. And now is an interesting time in the world of data-driven marketing because, for example, Apple has said, “We’re not going to allow you to use identifiers to find someone on our iOS system.” So you can’t necessarily identify somebody just because you’ve seen them before. You must capture data and use it to identify them again on iOS browsers.

Google, as an example, has announced sometime in 2023, and they’ve actually made this announcement a couple of times and pushed the date back, but they’ve announced that you’re not going to be able to use cookies to identify folks who are using Chrome. The same goes for Android devices. So, first-party data will be increasingly important in the future to identify folks for retargeting ads and personalization outside your own website.

Anna Shutko:
Right. And I like your point about this type of data being key to unlocking someone’s identity, which is a great segue to my next question. So now that the marketers start to collect this type of data, which actually helps identify a person across different devices, as you mentioned, how can they collect and clean this first-party data ethically and securely to make sure it doesn’t leak and to make sure the data integrity is preserved?

Josh Perlstein:
So it’s important to answer that question and then a couple of sections. First, set yourself up for success in collecting first-party data at scale. You have to have the right platforms and policies in place. Let’s start with a privacy policy. First and foremost, you need to make sure that you have the proper consent and the proper overt policies in place to tell consumers you’re collecting their data and you have the ability to use that data. So that means you need to look at your privacy policy. That means you may need permission to capture that in markets like the EU and California. In addition, you need to have the right secure platforms to hold that data. Secure login access. You need to have the right procedures in place to ensure that data doesn’t leak.

So once you have the right infrastructure and permissions to capture that data, it’s important to lean forward into your strategy. So that means you’re tagging your website appropriately to be able to capture data from a behavioral perspective. Your database is set up appropriately, so you’re capturing what data is important. And in your initial strategy, we encourage clients to figure out what data is most important and isn’t as important. So you can prioritize. For example, transactional history, what you’ve bought, what you’ve browsed, and what you’ve put in shopping carts for an eCommerce retailer, for example, might be a valuable thing to capture vs. what categories I’ve browsed very briefly. I don’t want to capture every page a visitor looks at. Only those pages will be meaningful and support my strategy for targeting them in the future.

On the declared data side, we often encourage clients to look at every opportunity to capture information and permission from the consumer. That means a call to action to get someone to give their email address, give permission to get the email in the future from a given marketer and give other data points. I’ll use the eCommerce retailer example, like what kind of products they’re interested in when they plan to buy. Asking those questions and capturing that data can be critical in the success and how you market back to that person again in the future.

So every opportunity to capture that data front and center, we encourage the use… Well, intelligent use of things like light boxes or popups to capture information at critical times. And I say intelligent lightboxes. We’ve all been there when we’ve seen a lightbox or a popup over what we’re trying to do on a webpage annoyingly multiple times. Today’s technology serves those kinds of things, maybe after someone’s been on a website for some time. Only once or twice in a session, only if they haven’t given you that data in the past.
So we encourage marketers to think about capturing data at every given opportunity and thinking intelligently about how to use that data moving forward in a way that’s toeing the line between intelligent and not too creepy.

Anna Shutko:
Yes, I think that’s a very, very interesting answer. So there is a lot to unpack there. You’ve mentioned that mergers obviously shouldn’t cross the line. So data should be collected very carefully, very ethically. And you’ve mentioned a couple of ways marketers can do so.

So now let’s talk about one of them a little bit more. So you’ve mentioned that, first of all, there is a certain toolbox that a company has to have, and that’s a place where data can be secured correctly. So why should a brand validate the data it collects before adding it to this secure place? For example, as you mentioned, a data warehouse. And maybe you could add how exactly what is the course of action according to which a brand can validate that the data that’s added to the warehouse is correct.

Josh Perlstein:
Absolutely. And these kinds of data validation options have been around for a while. I’ll go with the old tried and true. When you capture someone’s permission to send an email, often, you’ll get an email back from that marketer that says, are you sure you want to get an email from us? Please confirm. The old school has been called double opt-in. That’s not required, and it can be quite cumbersome, actually.
So doing things like sending an email to someone to make sure that email is correct before you add it to your database is an important element of making sure that email is real because if you send bad emails to some ISPs, they’re going to start blocking your emails. So really important to confirm that information’s correct either in an outbound way, like sending emails, or there are validation services that connect in real-time, and we use quite a few of them.

But those validation services will make sure an email address looks like a real inbox, make sure syntax and setup are correct on an email address, it has a dot com or dots whatever, it has an @ sign in the right place.

It also can triangulate things like this person is coming from this IP in country X, but they said they’re in country Y. Is this potentially fraudulent? Does this look like a bot? A variety of real-time validation services can reduce the amount of bad data a marketer gets. And we think it’s a critical piece. They’re usually priced on a cost-per-transaction basis. So it’s not like a marketer has to pay a lot for these validation and verification services. They only pay when they use them. So we found a lot of value in that, and we found that it cuts down on the amount of bad data, bad data being fraudulent registrations, bought traffic, or bought registrations enormously, which ultimately won’t only make your marketing more accurate, but it will save you money over the long term by not serving ads to or sending emails to bad addresses.

Anna Shutko:
All right. Excellent. Thank you so much for sharing. I think this is a very useful algorithm that many marketers out there can use. And previously, you mentioned that the next step after data collection is actually using the data. So let’s talk about this. What kind of insights can analysts get from the first-party data after it’s collected? And in a follow-up, how much more accurate is the first-party data compared to the third-party data?

Josh Perlstein:
So there are many ways that marketers can use first-party data. First, I’ll explain that it’s not just about insights and analysis, but you can leverage that data for better media targeting, which is important. And the difference between first-party data and what you’re getting from second or third-party data, second being data you’re using from a platform directly or some other owner, third-party might be data from some other source altogether you might be used to do advertising or marketing on another platform.

So when we think about how you use first-party data, you can take real customer data, like folks are interacting with you directly or giving you information directly. Turns out that that’s a lot more, not surprisingly, a lot more accurate than data you’re getting from somewhere else. You can take that data and build models to target your media better.

I’ll give you a use case for that. We have a baby formula client. And that baby formula client often looks for expecting a new mom’. And there’s a handful of sources we can buy that data, but when we’re using our own data, we’re collecting from website traffic, from folks who have registered on our website. Our models to find those individuals are at least 30% more accurate because real data is connected directly to a consumer. On the analysis and insight side, again, we’re talking about real data vs. data we’re bringing in that could be collected in aggregate.

You’re not sure exactly where it’s coming from all the time. And that allows us to build real-time customer profiles. What does a customer actually look like? What is their real behavior?

I’ll continue with that baby formula example. When we were thinking about and taking information from our clients about who they thought their best customers were those customers, they had thought interested in very relevant content like baby nutrition, and feeding turns out when we looked at the data in analysis, what did our best customers really want to hear about in content from this brand, turned out it was more about preparedness and emotional and mental state of mind after having your first child.

We started to create more content around that using our first-party data as an analysis point, like are these people buying more if we give them more content that their behavior showed they wanted vs. what we thought they wanted? Turns out they did buy more. They interacted more frequently with us.

So it’s kind of like I’m trying to think of a good analogy, and a good analogy might be you’re talking to a group of friends. You think they’d really like a joke about X. You can try that joke about X, but when you see that they don’t laugh, you pivot and try other things, or maybe you ask them a little about what they think is funny.

It’s a terrible analogy. My apologies. But real data and feedback give us much better insight into who someone is or who they perceive themselves to be than using assumptions or data from some unexpected source. And that often is where third-party data comes from, random or aggregated or collected sources. And it’s important to keep asking questions and keep testing and learning about first-party data to give you a better insight into how to build better relationships between brands and their customers.

Anna Shutko:
All right. Thank you very much for sharing. These are great examples. And I did like your joke analogy, actually. So now I want to clarify the first-party data collection process a bit more.

So you’ve mentioned that, of course, the data the company collects and the user market collect is far more accurate, and 30% is a very, very surprising statistic, but how can marketers rely on this data? And what I mean by this is how can they know that they’ve collected enough data for the results of their analysis to be statistically significant? Is there a particular threshold rule of thumb?

Maybe you’ve observed your clients collecting a particular amount of data, and then they started making relevant conclusions based on the data analysis. What can you tell me more about this?

Josh Perlstein:
So my disclaimer to my answer will be that I’m not a data scientist or an analyst. So I’ll give you a layman’s marketer view of this. And the best way to answer your question is I believe that it’s all relative. So we have B2B clients as an example who may be selling hundreds of pieces of software a week because it’s expensive, it’s a long considered purchase process, and they have a smaller data set than, for example, my consumer package goods clients who may be selling thousands if not tens of thousands of units per week.

So that’s a question better suited to the world’s data scientists. But I’ll tell you. Most marketers have a gut feeling. If we can get enough data to feel like it’s giving us marketing direction, it’s good enough. So that may mean hundreds of survey responses or at least a few 100 anecdotal interaction points. That can give us enough confidence to spin tests and hypotheses and then prove them.

So there is no magic number per se. But I think there’s an iteration process that marketers must go through to develop insights based on their data. I’ve seen X number of people visit this page and act this way. I’ve seen X number of people give us this data point, so let’s come back and try this because it seems like that’s where the data are taking us. And then, we keep testing and iterating until we see the incremental lift or not. That will allow us to feel comfortable leaning forward even further.

So data’s interesting in that it doesn’t lie, it tells us the truth, but it can lead us to some unexpected conclusions in some cases. The baby example I mentioned earlier where our content was way off, and we were looking at content that wasn’t directly relevant to our product as being more desired by our customers was completely unexpected and surprising. And we find stuff like that all the time that challenges our conventional view of where we thought we were going.

So I wish I could give you that magic formula, but it’s just a process of testing, learning, iterating, and testing again. And data gives us the right answers to lead us ultimately to the right place.

Anna Shutko:
All right, Josh. Thank you so much for telling us more about the first-party data and sharing all your response insights. And now, if the audience would love to learn more about you, where can they find you?

Josh Perlstein:
Sure. I’d love it if folks want to connect with me on LinkedIn, Josh Perlstein, CEO at Response Media, or come to our website,, to learn more about me or what we do in helping marketers leverage data to make marketing work better.
And a couple of things that I also want to mention that are really important as it relates to first-party data.

And I mentioned this as a concluding point because I want folks to walk away with some actionable steps. This perception is that first-party data can’t be collected at scale. And again, we mentioned scale as a relative term, but first-party data can be if you focus. And focus means making sure your website and all your owned properties are set up to collect data and put it into your database. Make a conscious effort to capture important declared data, registration data, or survey data, wherever you can at reasonable times, and budget for it.

We often get introduced to clients who say, “I want to do more first-party data collection and usage,” but they haven’t set aside budgets specifically dedicated to that. It needs to be earmarked and budgeted specifically.

And last but not least, now more than ever, make sure that you’re collecting consent. I don’t know where privacy legislation on data collection may go in the US or other global markets. Still, it’s critical when you’re capturing data and planning to use it that you give your visitors, customers, etc., the opportunity to raise their hand and say, “I’m okay with you using my data. I expect value in return, but I’m okay with you using my data.”

Collecting that consent, whether you need it or not, and you don’t need consent in some markets, the US is one of those. It’ll keep your first-party data practice future-proof because I assume that more privacy legislation will come in the future. So it’s important that the folks you’re collecting data on are okay with that, and they’ve given you clear overt consent. Just wanted to leave you with a couple more points.

Anna Shutko:
Excellent. Thank you so much for sharing these, and thank you so much for coming to the show.

Josh Perlstein:
My pleasure. It was an honor. Thank you.

Turn your marketing data into opportunity

We streamline your marketing data so you can focus on the insights.

Book Demo