The industry has been moving towards integrating AdTech, MarTech, and research data for quite a while, and that trend has only accelerated as companies scramble to find solutions to take the place of the cookie, adapt to privacy regulations, and/or overcome data siloing. As a result, we are in the early stages of a new marketing ecosystem powered by personal data, and many companies in the research space (especially panels) have been moving to try to take advantage of the opportunity.
The bottom line is that the marketing ecosystem is powered by data. That data is used to drive advertising and targeting, which is how the system makes money. Companies need data to engage, understand, and activate consumer relationships, and suddenly lots of consumer data is being removed from the traditional system by moving behind the walled gardens of privacy-first platforms with limited to no mechanisms for brands to engage with these user populations. That is a huge issue for both advertisers and the companies that exist to serve them via data. This new marketing data ecosystem has created a significant vacuum that isn’t easily filled by the current players, and everyone is feeling the economic impact of that gap already.
Therein lies the opportunity, and the focus of today’s interview.
I’ve known John Dick, CEO of CivicScience for over 12 years and have followed them closely as they charted their own unique course within the insights space. As the first company to scale online surveys as a content gateway, as well as pioneering “data chunking” models for surveys, CivicScience has been pioneering the future of survey research for a long time. With that in mind, it is no surprise that the company has decided to take advantage of its massive data assets to develop a new model for ad targeting. Here is the announcement that signaled CivicScience’s next move:
“Built on CivicScience’s fast-growing, industry-leading market intelligence platform, the new company will leverage first-party, privacy-compliant data from over 100 million survey respondents and relationships with hundreds of leading digital publishers and Fortune 500 advertisers who use the CivicScience platform. As privacy regulations and the elimination of third-party cookies accelerate, the company is uniquely positioned to deliver ethical, consumer-friendly advertising on the open web.
“There are few things skyrocketing faster in our data today than people’s privacy concerns,” said CivicScience CEO John Dick. “The beauty of this solution is that it won’t just protect people’s privacy, it will actively engage and empower them in the advertising process.”
By relying on the billions of permissioned, self-reported poll responses CivicScience gathers every year inside the content of premium publishers, the new company will power high-fidelity, anonymized ad targeting that is fully privacy-compliant and not dependent on third-party cookies. As publishers work to expand their own first-party data capabilities, this new platform is designed to allow them to compete more readily with the walled gardens for advertising dollars.”
In this interview, John and I explore this move in-depth, what it means for the industry as a whole and what happens from here. It is an important interview because I firmly believe the direction CivicScience is heading is a big part of the future of the industry and we’ll be seeing many other companies moving in similar directions very soon.
One note: We did have a bit of an audio glitch at the very end of the interview, but the interview was wrapping up so nothing of substance was missed and we just went with it as is.
The text has been edited for clarity
Lenny Murphy: Hello, everybody. It’s Lenny Murphy here with another one of our CEO Series interviews. And all of these are kind of special, but this one is particularly because I’ve known John Dick, the CEO of CivicScience for, what, John? 12 years?
John Dick: Right.
Lenny Murphy: Right after you launched, I was working on a startup called BrandScan. John was working on CivicScience. There were plans to partner and utilize their data to drive BrandScan. BrandScan didn’t happen in any meaningful way, but CivicScience kept ongoing. So, it’s really great to talk about what we’re going to talk about today and kind of how far you’ve come. That’s my very long-winded way of saying, welcome. And John, why don’t you tell the audience who may not be familiar a little bit about yourself and a little bit about CivicScience, and we’ll go from there?
John Dick: Yeah, absolutely. And it is hard to believe it’s been 12 years. And first of all, thank you for having me today, and thanks for all the support you’ve shown the company over the decade-plus we’ve known each other.
But yeah, I’m John Dick. I’m the Founder and CEO of CivicScience. CivicScience is a research company based in Pittsburgh. Started the company through Carnegie Mellon about 13 years ago, sort of alive 12 years ago or so.
The general premise around the company was, could we find a better way to gather survey research at scale that didn’t rely on the prevailing panel methods that we all know about today? A lot of the research that we had done uncovered some things about that world that we felt there were some areas for improvement.
When you’re compensating somebody to answer a survey, it introduces certain kinds of biases into the exercise, and kind of psychographic biases that we were seeing a lot. The higher proclivity to eat at value menus and use rewards programs and all these kinds of things. And also too, there was just sort of a scale limitation that we felt we could do better.
And so the general idea we had was to harness polls and quizzes people were already answering sort of organically around the web inside of content sites. Might be the poll of the day you read on your local newspaper website, or it might be a quiz you answer on BuzzFeed about what wine or Simpsons character you are. We know tens of millions of people every day are answering those kinds of polls and quizzes, and we thought there was an opportunity to harness that data and/or control the experience in such a way to do a couple of pretty cool things.
First of all, engage just massive amounts of people in survey research, many of whom weren’t otherwise inclined to participate in, sort of what we would consider to be survey research. Secondly, we could do it in a way that sort of bartered with the publishers for space in a way that allowed us to gather all that data effectively for free, which is what we do. And I’m happy to explain that in a minute.
And then thirdly, by virtue of gathering all this data at scale for free, we could build what was primarily a syndicated business, which we know generally has much larger scalability and higher profit margins, right? So anyway, so we did that. Took a lot longer than any of us expected to build a large enough network of publishers to do that, to be able to achieve not only representativeness but the scale that we needed to achieve. And that took us the better part of almost seven years, I think.
We then take the aggregate of all of that data from hundreds of sites and we build research products on top of it. Primarily syndicated, but we also have a pretty robust custom business as well. So our customers are large brands, large media companies, and a lot of hedge funds, private equity, and hedge funds looking for signals in that data.
We do upwards of about seven million of these polls a day, which is just enormous; we have at any given time several thousand questions running through that system. And we sell a subscription service on top of that to, again, large corporate enterprises primarily.
Great business. Growing really healthily over the last few years. Particularly since COVID, there’s a certain real-timeliness to our business that’s been pretty compelling since the pandemic started. Our ethos is that everything affects everything, and everything’s constantly changing, so we study everything constantly. That’s our approach.
The last big number is we just crossed our 100 millionth respondent. So we’re dealing at a different scale, I think, than your typical kind of panel business might be, just by the sheer multitude of people that we’re able to reach. And yeah, that’s the general background. Happy to go deeper on any of those – about your audience and what they’re into.
Lenny Murphy: Yeah. Well, let’s talk about – and again, because of history. So one of the things I thought was always interesting – and you were the first in the market with this – was the idea of data chunking, right? I mean, you used this model before Google Surveys came into this model, as well as all of your respondents are anonymized.
So contra to traditional panels, you’re taking – if I get it right, and I think I do – you’ll ask the same question of 1,000 people, ask the next question of another 1,000 people, whatever the case may be, and build out this massive data set without knowing the specifics of who that individual is, but leveraging this fiscal science to be able to say, look, we are sure this is representative of the overall population based on this data.
One, did I explained that correctly, and two, how has that driven the evolution of the business? Because I think that, one, there was a lot of resistance originally to that idea and how you’ve overcome that resistance because obviously, you have.
John Dick: Well, I mean, there’s a couple of things. First of all, we explicitly ask the demographics of these people so we’re not inferring or imputing anything about folks. Traditionally, we used what everybody else used was third-party cookies to identify the same user over multiple sessions. So our typical poll or quiz might only be four questions, right? Four or five. And they’re generally non-contiguous. We’re not asking somebody a string of things, right?
One of the things we like to say is that every time you add a question to a survey, you increase the bias in the people who will answer it, which we’ve just found to be true. And when we started to get to five, six, seven, eight questions, the person who answered seven and eight didn’t look like the ones who did it, right?
So we managed for that very closely and very carefully, but also, two, we’ve just over time been able to validate our data on two empirical measures. One is that our universe of respondents matched very well to what we know to be ground truths, right?
So for example, we know, what? I think 12% to 13% of Americans smoke cigarettes. That’s one of the gold standard questions we ask simply to be able to measure how accurate a representative 100 million people are to the full US population. And when we see things out of whack, we can adjust for that by going to find more publishers, right? And so that’s one part.
And then the other part I think is what our customers would tell you, many of whom who have worked with us now for five, six, seven years, is that when they use our stuff to measure something or predict something, it turns out to be pretty damn accurate, right? They feel very confident that what we’re producing are a reliable read or measure of the present and the future.
It is definitely a radical or a controversial methodology, the way we do it. But what matters at the end of the day is to the customers, does it work? And yeah, it works really well.
Lenny Murphy: OK. So now that leads to the news that recently came out on a new application of your data. So why don’t you talk about that, and then we’ll talk a little bit more about the kind of broader trends?
John Dick: Yeah, about a week and a half ago, we announced the launch of a new business with some new strategic partners. I should be clear that a couple of our key CivicScience partners include – The NPD Group is one of our majority investors, and so they were a partner in this new business. But we also had a couple of new investors. Jeff Wilke, who was the recently departed kind of number two at Amazon who left in March, and Thomas Tull who’s a very well-known investor in tech and AI, but also a pretty famous movie producer.
So we brought these folks together to launch a new business to apply this data asset that we had built to the AdTech space, right? And we’d resisted that for a really long time, even though we had massive amounts of this data.
The general way that AdTech had worked in the past didn’t really care that much about privacy, right? And that’s something that we were pretty keen on. We’re not following any PII on these people. It’s all anonymized. We didn’t want to just source that data into the open web and lose sort of the control we had on it. Sorry, I knew I would hear my dog barking through the door, but I hope we’re all used to that on these meetings by now.
Lenny Murphy: Yep, absolutely.
John Dick: Yeah. But then the world started changing, right? Two things began to happen. We started to see the deprecation of the third-party cookie. The last straw that’s yet to fall as Google sort of pushed that out. But almost half the web between Safari and others isn’t supporting third-party cookies anyway. But also just a significant rise in privacy compliance and privacy regulations.
What’s interesting about the data we collect is a couple of things. One is our polls are first-party implementations inside of the content of these publishers. It’s not an ad unit, right? This means within a given context of a publisher, we can use a first-party identifier against a respondent in our system to build that same profile that we were doing in the past. Now we’re not following a user across properties, but we can do very rich data analysis within a given context of a single site.
Two, it’s all permission, right? So each question a person answers or each poll they answer, we’re checking the boxes of CCPA and GDPR, and so forth. And so with this coming new world or part of the world that’s already here – and we think we’re at the beginning of the privacy reckoning, not the end of it – we have a really unique opportunity.
We have a massive first-party privacy-compliant audience data set that’s consistent across hundreds of premium publishers. And so in a world where we move away from very heavily sort of PII-based targeting of individuals who we track across the web and stitch all this shady information to, we’re pretty well-positioned.
We’re also pretty well-positioned primarily, or additionally, because we’re actually surveying people within the constructs of the media that our advertisers want to buy against. So we are actually asking and building audiences within polls that are answered on their personal computers –
Lenny Murphy: Your site, yeah.
John Dick: – or Univision, you know? And so it’s a significant advantage we have. We kicked and screamed, Lenny, as you know. We’ve sort of resisted the AdTech lure for a long time.
Lenny Murphy: Yeah, I tried to convince you, what, last year to –
John Dick: Yeah, you were maybe a little bit ahead of me. But you were right. We ended up having to go and work with one of the bulge bracket consulting firms before our board and investors were comfortable making the big leap that we made because we are making a substantial financial investment into this new business.
We had to go do our homework, and it took a good six months of consulting fees to validate, A, where we believe the market is going is where it’s going. B, that we have a unique competitive advantage or set of competitive advantages. And C, the prize is really big. And so, once they did that, they helped us develop the business plan for it, and here we are.
Lenny Murphy: Yep. Very cool. Now so as I mentioned, obviously, and those of you who don’t know, I’ve been entrenched in this idea for a little while. And I don’t mean that in a competitive way. Smart people have been thinking for a while, this was where the world was going to go, right? And John, you’re one of the smartest people I know, so no surprise you got into this place for all the reasons you mentioned.
So I think it’s incredibly interesting. When lots of smart people come up with relatively the same idea around the same time, that’s something I pay attention to, because that means there’s something there. I don’t know if we call it the zeitgeist, the collective unconscious. Whatever it is, right? But it seems to happen fairly often, that a lot of bright folks think, hmm. There’s something to this. So I love the idea.
And within the last month, we’ve seen two to three other launches along similar lines. Dentsu just launched a product that is leveraging research data to drive that targeting, right? That’s the core. Almost all the panel companies have been chasing this for a while. Of course, Veriglif, my startup, we’re chasing the same idea.
And all variations on a theme with the same idea. How do you produce a privacy-compliant, consumer-centric, new advertising model driven off of research data versus off of the data exhaust that has been driving us for so long? So I think you’re absolutely on the right track.
Again, I reached out to you, what, about a year ago, saying, hey, come play with Veriglif, because I knew how valuable that your data set is for a variety of applications. So I’m thrilled that you’ve gotten to that place and that you did your homework and decided, hell yeah. Let’s do this.
I think that you are going to be one of the companies that will help define what this future looks like over the course of the next few years and be the next big AdTech company. So it’s fantastic. Hats off.
John Dick: Well, I appreciate all that. Let me just say one thing. And I really appreciate the kind words about me being smart or whatever, but I’m also smart enough to know where I’m dumb. And this is an area where I’m not an expert.
So we hired a new CEO, a guy by the name of Doug Lauretano from Media.net who started July 1, but we announced it a couple of weeks ago. He will build an entire sort of standalone management team to build and run this business because he’s forgotten more about AdTech in the last few months than I’ve learned in the last year and a half, right?
So we’re going to get out of the way and let really smart people who have a long history and experience in this space build the next big thing. So I will be a very interested observer and support Doug and the team as much as I can. But by and large, I’m still super excited about the research business that we have running.
So we created sort of this central structure with subsidiaries and brought our friend, Zach Nippert, who I know you know is now running our research operation. He’s the president of that business. So yeah, it’s all my long quest to do as little work as I have to do.
Lenny Murphy: [LAUGHS]
John Dick: But I appreciate the kind words. And yeah, we’re very optimistic. We think we have some very unique barriers around this business by virtue of our direct integration with the publishers themselves. And then, of course, 100 million people already have answered our polls and growing. So it’s a scale you can’t get to a lot of other ways that we’ve been able to build.
Lenny Murphy: Pretty great.
John Dick: Thanks for having us, Lenny. Always love to shine a spotlight on the business, and always good to see you and talk to you. And happy to help you any way we can.
Lenny Murphy: Now, well, I appreciate that, John. I want to give a little more to the audience. This isn’t just Lenny and John with their kind of rattling off ideas. There’s some information that I have that I can probably share now. For instance, when we were starting Veriglif, IBM was one of the companies we were partnering with because they saw the vision of this with their Mediaocean business. So Mediaocean’s one of the largest AdTech companies out there. LiveRamp has been chasing variations of this for quite some time.
There’s a ton of money going towards this idea. I would say from what I can see within the market, though, you’re one of the first to launch with a credible solution, with a data set already in place that could help drive results.
So for the audience, I wouldn’t even describe it as the tip of the spear. So you’ve got the first-mover advantage. You’re one of the early companies that have a credible solution that could help scale this. And there is a ton of money and lots of big companies that are chasing this as well, which is going to put you in a very good position in a few years where you won’t have to work very much, I suspect. So – [LAUGHS].
John Dick: Let’s hope.
Lenny Murphy: But the interesting thing, I love that you’ve put a bit of a wall between the two businesses. There’s the research business. There’s the AdTech business. I think that’s always been one of the barriers that our industry has struggled with is the twain shall never meet. But yet the synergies were absolutely there. So has that been part of the thinking in creating that separation, John, to make sure that your research customers feel comfortable with the research business while your marketing customers feel comfortable with the marketing business?
John Dick: Yeah. I mean, we don’t – well, we see that less and less, right? Because I should say they’re independent somewhat structurally, but there’s a tremendous amount of overlap, right? They’re obviously leveraging the same data asset. They’ll ultimately be selling to the same customer, right? I mean, generally within our market intelligence business, we are selling up into senior marketing or parts of the senior marketing organization, and that’s generally where the media budget tends to live.
By far, the most common use of our research platform is what – and I hate this word, but I’m going to use it anyway because people understand it – is what we would define as segmentation. So it’s analyzing these 250,000 different attributes we gather and the relationships between those things across people to help form media strategy and media planning, right?
But that’s still done in a relatively analog manner. We’re saying today, the Hispanic millennial mom who’s in the market to buy a stroller looks like this compared to what she looked like last week, right? And so it’s very agile to do that.
But then what the marketing organization is doing is taking that, putting it into a PowerPoint, sharing it with the marketing folks, and then they’re going to their agency or whoever’s building and buying the media using it as a roadmap. All we’re saying to that same person is –
Lenny Murphy: John, lost you just for a minute. Well, all right, guys. I think the gods of the internet just took their favor away. Sounds like we may have lost John, but I think we were done anyway. This happens occasionally in interviews, so welcome to the new normal.
So we’re not going to edit this out. We’ll let it fly as is. And this gives us an excuse to bring John back for another conversation to follow through on whatever he was about to say, what wisdom he was going to bestow upon us.
So we’ll go ahead and cut now since we lost John. But again, those who are listening, pay attention. This is a big deal. The companies that are going after this are a big deal. There’s a lot. Particularly from the AdTech side, there’s a lot of money flowing into this idea. I’ve been chasing it for a while. John obviously has pulled together great solutions. This is the future of the industry, so it’s cool to be a part of such a major transformation.
John sounds like you’re trying to come back. I’m going to go ahead and stop the recording, though, and we will bring you back for another conversation about this soon. Thanks a lot, everybody. Bye-bye.