In this CEO Series interview, Leonard Murphy sits down with Bob Fawson, founder of the Data Quality Co-op, to explore his extensive career in market research and the innovative mission of his new venture. From building telephone research panels at Western Watts to leadership roles at SSI and Numerator, Fawson’s career reflects a deep commitment to improving data quality. With the Data Quality Co-op, he aims to bridge the gap between buyers and sellers in the research industry by offering tools for transparency and better decision-making. By addressing challenges like fraud detection and attention quality, the platform ensures that data inputs are reliable and fit for purpose, creating a more trustworthy ecosystem.
Fawson also shares valuable insights for industry professionals, emphasizing the importance of balancing humility, confidence, and adaptability in navigating rapid technological shifts, including the rise of AI. He highlights the critical need to evaluate data inputs rigorously, especially as AI transforms research practices. Fawson’s advice to newcomers and seasoned professionals alike is to continually build skills, leverage connections, and embrace curiosity to thrive in this ever-changing landscape. With initiatives like the Data Quality Co-op, the future of market research is poised to prioritize quality and innovation.
Transcript
Leonard Murphy: Hello everybody, it's Lenny Murphy with another edition of our CEO series of interviews. Thank you for joining myself and my guest. And today I am joined by Bob Fawson, the founder of Data Quality Co-op, but of course you've been around for a long time with lots of other titles we could throw in there, but welcome Bob. How are you? I'm looking forward to it. love getting into these type of conversations. So, why don't you talk about Data Quality Co-op and a little bit of your bio? Give us your origin story so everybody can kind of level set on why you're here, why of all people get to sit in that coveted chair of a conversation with me. Yeah.
Bob Fawson: Perfect. it's thanks for giving me a chance to talk about myself, Lenny. It's human nature, so we all like to do it. I've been in the industry for over 20 years. started at a company called Western ts based in Utah. Watts stand for wide area telephone service. So everything I know about telephone research I learned by building a panel at a telephone research business and we called that panel pinion outpost. Merged the company with SSI in 2011 and then I spent over a decade with SSI and subsequently DA playing product and strategy and operational executive roles and learned a lot about this business. my interests were always in academic research until I found market research and more particularly found the sample space. I think it's so interesting to think about the inputs that drive so many outputs and so many decisions across the world. After DA spent a little bit of time at Numerator helping them build their survey research business. Learned a tremendous amount first time in my career working primarily with brands and working primarily with behavioral data. And now we founded the data quality co-op. That's right. Yeah. So,…
Leonard Murphy: Tell us about that.
Bob Fawson: I maybe kind of tying back to the introduction, thinking about the most fun I've had in my career, some of those moments were really the early days at Western Watts, figuring out how to build a panel and how to do it in a way that got our clients excited about the quality of the data they were receiving. And after a couple of decades doing that I had a lot of deep thoughts about what works well in our space and what's broken. And with the data quality co-op it's a chance to think about how to help both buyers and sellers solve their biggest kind of day-to-day dilemmas. Buyers who struggle to know really the quality of the data they're buying before they make a purchase or before they use it. and sellers who are struggling to differentiate and make the business case to invest in quality. And so what we're trying to do is build a business that could provide transparency to quality and help both buyers and sellers make better decisions. And that really animates me. Yeah.
Leonard Murphy: Or a kind of fraud detection software which shade on those companies they do a great job. What's the angle that you guys are taking that adds a whole other level of value towards all of the other existing ecosystem around fraud detection? Yeah.
Bob Fawson: So I think there are two broad quality challenges to solve whether it's fraud or whether it's attention. the first is very much a technical challenge and I'm with you Lonnie. shade on anyone who's engaged in the kind of cisiphian task of managing fraud. and I feel really good about all the investment that's going into these technical data quality management platform solutions. I think there's a second challenge that maybe goes under-appreciated and we'll dive into that. You did a podcast with, D recently…
Leonard Murphy: Yeah. Mhm.
Bob Fawson: Where you talked about that. it struck a few chords with me but I think it's a coordination challenge. So every time somebody uses a research defender or empirium or runs a speeding check or uses roundt to evaluate an open ban or red, that goes into a data table and nobody ever looks at that again. And so how some futures markets operate or credit reporting agencies, we're collecting those signals from our clients. We're giving them BI tools to make more holistic decisions about their procurement and then we're also aggregating and normalizing those to put quality outcomes in context. Are you buying from a source that consistently overdelivers on your quality expectations? because that source should be able to grow their business and grow their share and that buyer should introduce less risk into the recommendations. That agency who's buying should have less risk in kind of making recommendations to the brands who are their clients. And so that's kind of the idea is through a clearing house and through providing more transparency that everybody at the front end of the research process can make more informed decisions about the quality of what they're buying such that the outputs are better or at least more fit for purpose.
Leonard Murphy: So, chatting beforehand, you brought up, Verisk, as an example, but, I also hear Equifax, Experian, TransUnion, all of those companies. but the difference is are you applying that at the individual level or at the supplier level or both? Okay.
Bob Fawson: Yeah, I think it's both and of course, any kind of data about an agency or a supplier is really just the aggregate of the performance of the individuals. over time, regardless of where those quality signals are being generated, whether it's in a survey or through some sort of outside of survey technical tool.
Leonard Murphy: and today your clients are primarily buyers, agencies as kind of the tip of the spear, the primary buyers of research. Totally get that use case. But I would imagine that if you are a sample company, especially a marketplace that you would both look at this as, wow, this is awesome. And I know. I don't even know if I want to know what's in my sausage. I mean unfortunately so what does that use case look like for you in the future of how all of the key stakeholders in the industry can leverage the platform and by the way listeners this is not a sales pitch for data quality co-op this is as Bob mentioned you check out the greenbook podcast with that recently did and lots of actually podcasts I've done on this topic this is a fundamental issue for the industry. It's existential in my view. so I do want to hear about this, Bob, because Lord knows we need better solutions or else we're going to be in real trouble. So, yeah.
Bob Fawson: the vast majority of my career on the supplier side of the business. So just personally where do I have the most empathy it's for the dilemmas of suppliers and my intention is to provide suppliers with tools to differentiate whether that's differentiate their company and brand or wouldn't it be great if you received a proposal here's the quality we deliver on average here's 2x at a different price point and here's.5x at a different price point and depending on the question you're trying to answer any three of those bundles might be the right choice. and so, I'm engaged with suppliers across the ecosystem as design partners and, making sure that we understand their concerns, their fears, but ultimately what I think a supplier gets out of this is they get visibility into how their sample performs inside the surveys at the agency buyers. they may get a reconciliation request today, but they don't know that, for publisher X where they're spending $100,000 a year recruiting consumers, they don't know that, over-indexes on low quality open ads or speeding or,…
Leonard Murphy: What? Yep.
Bob Fawson: Technical quality fails. And so, if I'm at a supplier, I'm going to spend my incremental dollar on a source that's going to not just produce a join form, but produce kind of a stream of clean data that will build trust with my clients and build my brand over time. And that's how they grow. I think trust is something in the decision kind of procurement decision- making process that's becoming more important. Yeah. Perfect.
Leonard Murphy: Yeah, that's really cool. So, the Good Housekeeping seal of approval, so to speak, of when we stop recording, hang on a minute, because there's a couple thoughts I had on that, but I don't want to get into one on this. that is cool and exciting. but along came AI right and lots of conversations about synthetic sample on not asking as many questions as we build up these repositories of information that we can mine. So what's the future look like to you right when we're coming in that obviously data is fundamental but how we use it and the business models that we currently operate within are likely going to change at least in my thinking. So what do you think
Bob Fawson: Yeah. Yeah. It's interesting. I mean, I guess we have some smaller scale parallels. People are comfortable making big decisions off of smaller sample sizes through qualitative when they feel like the individuals are talking to have been validated in some kind of extraordinary way. And I view AI really similarly. if you're going to model that data over and over and over again and you're going to use it to inform more decisions than a kind of traditional survey data set, you should be more and more interested in the quality of the inputs. And so I think a healthy shift toward AI will be equal part kind of development of models and of evaluating the training data sets and making sure that the data we collect and feed those models with is data we feel comfortable about amplifying using AI. And so I'm very optimistic about that future. I'm very excited to see it and the kind of shift towards digital qual and kind of qual at scale I think is a precursor of how that might look as AI grows.
Leonard Murphy: Yeah, I agree. It will be as the need to my supposition is every question asked will become more important going forward. And if that is the case,…
Bob Fawson: Yeah. and…
Leonard Murphy: Then knowing that we're getting the right information from the right people is equally important.
Bob Fawson: I think that presents some challenges for the way that we manage quality in the industry today. as AI generated data becomes more and more looks more and more like survey data. The quality discussion I think should shift less from evaluating outputs and more towards evaluating the inputs. Did the training data set come from people who we've seen before, have been thoughtful before, have a history with, are real people? Because ultimately we want to make sure that the inputs we're putting into those models are high quality because the outputs are going to look very natural to us and they'll become I think increasingly difficult to distinguish from traditional survey data. Yeah.
Leonard Murphy: Yeah, I agree. I've already seen examples of that, there are platforms in sales right now that are both qualitative and quantitative that if I did not know what was happening I would think that those were real live respondents in the traditional paradigm and they are not.
Bob Fawson: Yeah, very much.
Leonard Murphy: Yeah, interesting times from Yeah.
Bob Fawson: It's interesting, too. I don't think this is all that unique. and this is why I never get in invited, on other podcasts, Lenny, because, I was having a beer and, checking out the Chicago Mercantile Exchange website a couple, weekends ago, thinking about we talk about sample or data being a commodity, how do other commodities markets work? And I was shocked by the variety of just wheat. If you dig into all their information about I don't know there's 20 types of wheat and there's 20 gradations of that wheat and all of that is graded and there are standards maybe a little different than sample but if you're going to ship that wheat it's got to be low moisture content if you're going to make organic bread it's got to be something different and one of the things that I would like to see in our space is thinking about sample whether we call it a commodity or not or thinking about data and saying what am I trying to build with that input? I'll end with the wheat analogy, but do I want the organic kind of low moisture content, high protein content, or am I trying to confirm something that, I'm not quite sure about? Maybe we've all used the term cheap and cheerful. there's a world for that.
Leonard Murphy: There is.
Bob Fawson: And I think the biggest problem is the lack of ability to differentiate. because then nobody trusts each other. They don't quite know what they're getting.
Leonard Murphy: Yep. Yep. Agreed ory. I was actually a guest on a podcast right before this conversation with the AYTM and it went to the same place look there is a role for cheap fast surveys which is clients know they're throwing away 20 30%. Already that's acceptable for certain things it's not acceptable for other things. So it's…
Bob Fawson: Yeah. Yeah.
Leonard Murphy: what is the consequence of the decision being made and what is the use case and same thing with synthetic example I've been playing with GP01 and to explore personas and just kind of ideulate and throw some things out there it's great right now but also I asked do competitive analysis and it hallucinated right so it's like there's just limits on doing this and We'll just have to adapt to it. But to your point is garbage out, one way or the other. or weed out.
Bob Fawson: Yeah. Yeah, that's right.
Leonard Murphy: Separate the wheat from the chaff. So. All right. You and I could go on for a long time. and we're trying to keep this relatively short. so you've been around the industry for a few decades now, like me, right? We have to own that. We're old, and you would assume that we have some experience, some lessons to learn, some wisdom. So lay it on us Bob. What are some of the key lessons that you have learned that we can impart to our audience?
Bob Fawson: first I feel like I need to defend myself. I may be old, but I still feel young, Lenny. It's important to say I don't know with each passing year I realize…
Leonard Murphy: Sorry, I was projecting. I probably shouldn't have done that. So apologies, Go ahead. You're young. I'm old.
Bob Fawson: how little I know and I wish I would have learned to have a better mix of humility and confidence earlier in my career because as I look back humility, confidence and kind of patience. So, I think at times when my career has been rewarding or when I've been growing professionally, I've been equal parts frustrated and equal parts open to learning and equal part just working as hard and as well as I know how. And so, this is kind of a fun industry because I think at its best, you meet people who are intellectually curious and who are interested in informing the world with better data. And so, I feel very fortunate to have worked in this environment. And I guess that the only lesson worth repeating is, your work is one input, but kind of being open to learning from everyone else in this space is the second one. And it maybe took me longer than it should have to figure that out.
Leonard Murphy: that is great feedback. You don't often hear about humility, people talking about that combination I think characterized humility plus confidence. that I think of my own career and…
Bob Fawson: Yeah. Yeah.
Leonard Murphy: I had imposter syndrome for a long time because I over-indexed on the confidence part and under-indexed on the humility and somewhere down the line through that balanced out through learning from other folks now
Bob Fawson: Yeah, it is kind of tough though,…isn't it? even at this point I'm first- time founder and lots of years working in larger companies so it's almost inevitable that there's always a little bit of that imposter syndrome where you kind of become experienced in something new but I think that's good too and I always aspire to be 100% confident but totally open to the evidence and very quick to change my mind when I realize I was wrong. Easier said than done.
Leonard Murphy: I always say myself, I'm first one to admit that I'm just never wrong. at least that's what my wife says about me. those are great lessons. So, last bit as we wrap up. So, those were life lessons. Any specific advice that you would offer to folks that are listening that are growing to you're the first- time entrepreneur? after decades in the industry. so any advice that you would offer
Bob Fawson: Period of rapid change and AI is shephering us towards a new one. I think if you're new in the industry, don't be scared to acquire new skills. I feel like early in my career, I went and took some data science and database management classes at the local university in my spare time. And that opened up so many worlds just in terms of being able to communicate with people across the enterprise about different topics and understand things. And I think if you're, maybe a little bit later in your career in the industry, don't underestimate the connections you've made and the knowledge you've built. I'm continually surprised by people who've been willing to reach out and help me on my entrepreneurial journey and much of our professional discourse is now on LinkedIn. That's great. Love it. I learn a lot there. But, don't be scared to pick up the phone and ask because there's a wealth of support and knowledge out there. And, I think a lot of us under underlever that.
Leonard Murphy: Agreed wholeheartedly. so I'm blessed to do this for this part of my living because I have an excuse to talk to people. So yeah,…
Bob Fawson: That is nice.
Leonard Murphy: it's a fringe benefit for All thank you, it's been a great conversation. Best of luck on data quality co-op. I think we'll hear more and more about this. At least I hope that we do. And where can folks find you?
Bob Fawson: … you can find me at Bob Fawson dataqualitycoop.com. reach out anytime. And Lenny, thanks for the opportunity. This is fun. it's been just wonderful to reconnect.
Leonard Murphy: No, I'm glad and I feel the same So ditto. It won't be the last time, that's for All right, so we're going to wrap it up there. thank you to our producer, Bridget. to our Big Bad Audio. Thank you to sponsors and most importantly, thank you to you, our listeners, because you gave an excuse for Bob and I to connect and have this conversation. But that's it for this edition of the CEO series. We'll have another one real soon. Take care. Bye-bye.
Comments are moderated to ensure respect towards the author and to prevent spam or self-promotion. Your comment may be edited, rejected, or approved based on these criteria. By commenting, you accept these terms and take responsibility for your contributions.
The views, opinions, data, and methodologies expressed above are those of the contributor(s) and do not necessarily reflect or represent the official policies, positions, or beliefs of Greenbook.
