Anticipating the Future: Dave Ritter on AI, Analytics, and Retail Insights at Walmart

Description

AI is reshaping how businesses gather, analyze, and democratize consumer insights, while maintaining the critical role of human expertise and intervention. In this episode of The Curiosity Current, hosts Stephanie and Matt speak with Dave Ritter, Director of Insights and Consulting at Walmart, exploring how AI is transforming market research, consumer segmentation, and customer insights while discussing the balance between technological advancement and human intuition.

‍

Transcript

‍Dave - 00:00:00:

‍

I think there'll be levels of intervention with all of AIs. To me, AI is the new calculator. It's the cheat code. I think you're going to need to be comfortable asking a lot of questions. Obviously, Curiosity, podcast, right? So ask questions about things you don't know. Ask questions above your client. Everybody's there to help you. I think that's critical.

‍

Stephanie - 00:00:23:

‍

Hello, fellow insight seekers. Welcome to The Curiosity Current, a podcast that's all about navigating the exciting world of market research. I'm Stephanie Vance.

‍

Matt - 00:00:33:

‍

And I'm Matt Mahan. Join us as we explore the ever-shifting landscape of consumer behavior and what it means for brands like yours.

‍

Stephanie - 00:00:40:

‍

Each episode will get swept up in the trends and challenges facing researchers today, riding the current of curiosity towards new discoveries and deeper understanding.

‍

Matt - 00:00:50:

‍

Along the way, we'll tap into the brains of industry leaders, decode real-world data, and explore the tech that's shaping the future of research.

‍

Stephanie - 00:00:58:

‍

So whether you're a seasoned pro or just getting your feet wet, we're excited to have you on board.

‍

Matt - 00:01:04:

‍

So with that, let's jump right in.

‍

Stephanie - 00:01:07:

‍

Today on The Curiosity Current, we are joined by Dave Ritter, Director of Insights and Consulting at Walmart, where he's leading the charge in harnessing consumer insights and advanced analytics to shape the future of retail.

‍

Matt - 00:01:20:

‍

With over 15 years of experience in business strategy, customer experience, and data-driven decision-making, Dave has played a pivotal role in repositioning Walmart's Great Value Brand and democratizing customer insight across the company. His work spans from leveraging AI to streamline NPS programs to innovating within consumer segmentation and shopper experience.

‍

Stephanie - 00:01:40:

‍

In today's conversation, we'll explore how AI is revolutionizing primary research, the balance between listening to consumers versus asking them directly, and how technology is reshaping the way businesses understand their customers.

‍

Matt - 00:01:54:

‍

Dave, welcome to The Curiosity Current. Thank you so much for joining us. We are excited to have you here today.

‍

Dave - 00:02:00:

‍

Thank you guys for having me. I've listened to all of the episodes. It's great stuff. Keep it up. Honored to be here.

‍

Matt - 00:02:05:

‍

Well, you appreciate it. So let's just jump right into it. We always like to understand a little bit more about... Where our guests are at in their journey and how they found themselves in the career in which they sit. So you've had quite an evolution in your career. You were deeply involved in primary research and now you find yourself in an insights and consulting leadership role at Walmart. Can you share a bit about that journey? Tell us, you know, how your role has shifted from researcher to consultant?

‍

Dave - 00:02:32:

‍

Yeah, sure. I started out in marketing and just to give a friend's reference, I kind of did the reverse Chandler Bing. So I was in sort of a marketing creative space and I was, oh, I'm going to make ads. I'm going to be the idea guy, right? And I was at the University of Florida in the advertising program, their master's program. And I previously worked in marketing during that time. I did a internship at an advertising and PR firm and I started doing IDIs. I started helping with positioning statements, started getting into survey development. I'm like, oh, okay. I like this. I like this. So I started steering away from that creative process and that started the journey in research. And I ended up with a company called Kelton Research out in Los Angeles, California. And they're a boutique firm and they work with global clientele and they take you and they drop you in the deep end of the pool and you find out very quickly if you can swim. And their differentiator is they do tons of methodologies and they go really fast. I don't know what speed they're running at, but back in the day, they would do a segmentation in four weeks, turnaround time, like crazy fast. So, I did that for a year. And then my brother had his first child in Iowa, and it was time to go back to the Midwest. So I started working at State Farm. And State Farm was really that client-side trial-by-fire, you know, scoping projects, assessing solutions, budgets, communicating findings with recommendations. Worked there for a long time, and then Walmart came and approached me. And that's really where the pivot started to happen, because you had broader responsibilities outside of research. With Walmart, there was no shortage of data. We have plenty of what. So, and we're also an EDLC company, so it's everyday low cost. So before we even have budget to talk about research, it's what do we already know? And that is a big what. It's a big question, right? And then the other thing that people say is speed of retail. Don't like when people say that because it's usually a symptom of foreplanning, but literally things are crazy and we're going super fast. And a lot of times we're in a state of perpetual reaction and constant prioritization, but differentiation really becomes anticipating. So earlier in my career, it was curiosity. It was all of these things, but now I've kind of pivoted to this consultant role where it's so much more than primary research. It's analytics, it's social listening, it's everything. And it's what's next. What's, where's the puck going? We always like to use that Gretzky quote of go where the puck is going, go or not, don't go where it is. Right. So. That's kind of the pivot in the journey in a nutshell.

‍

Matt - 00:05:02:

‍

So. I love it. So that's super interesting. So the change was created in part by growth in your responsibilities, but also by, it sounds like, some pressures that were new to you, at least in this different position. It sounds like the pace was very different. And I also really like the call out on the enterprise that you're stepping into at Walmart. Very different in that you're sitting on a wealth of information already, and the company culture is such that you want to make use of the knowledge that you have, before going out and conducting primary research. It sounds like that was kind of the change that happened.

‍

Dave - 00:05:35:

‍

It was. And interestingly, the cultures are very similar. State Farm's very nice. Walmart's very nice. One thing I would give as advice, if anybody tells you you can't work in a different industry when you're in marketing research, that's not true. You can learn any industry. I had recruiters telling me, oh, you need to work in insurance the rest of your life. That is not true at all. So the pivot can be made.

‍

Matt - 00:05:56:

‍

That's fascinating. Yeah, I was told that I have to be an automotive researcher for the rest of my life very early on in my career, and that has not been the case.

‍

Dave - 00:06:03:

‍

There you go.

‍

Stephanie - 00:06:03:

‍

Or the supplier side, client side. That's another one that I would just suggest is a false dichotomy. I think that, yeah, one makes you, yeah, better at the other, like just by nature of having seen the other side. So.

‍

Matt - 00:06:16:

‍

So you've also been at the forefront of really utilizing AI to transform primary research and consumer insights at Walmart. Super interested if you could start by giving us a high-level view of how AI is being used in market research and how you're using it to gather insights and understand consumer behavior.

‍

Dave - 00:06:36:

‍

Certainly. I think right now we're in a test and learn phase. I have a wealth of opinions about the future, but right now we're applying it in our day-to-day and things like end-to-end. So you think about survey writing, analysis planning, automated reporting, insight summaries and storytelling. Those are kind of the things we're testing and learning right now, and we're finding about 60% of the way they're kind of hit and miss, right? And I've heard that from others, right? More tactically, I'm leaning in more with coding open ends, video transcriptions, coding like that, larger literature reviews, because I mentioned earlier the what. It helps to summarize a lot of content. And then the last one that I think is super important is research on research. So before we go out and do anything, we ask AI what it says, and then we'll go ahead and do the research and see if it coincides. I think more importantly to understand, is it giving us the same result? If so, that may be some things we can move over and automate. If not, then that's definitely something we need to tell the organization, because that's a huge problem. So that's sort of something that we're telling our teams across to do and start to document.

‍

Matt - 00:07:43:

‍

That's really interesting. So it sounds like you're doing like an internal validation of AI in the field, as it were, really kind of comparing it to the traditional methods and seeing if it's telling you the same results.

‍

Dave - 00:07:56:

‍

That's the plan, but we'll see how it goes over time, right? It's a moving target, so that's the tricky part of all of this, right?

‍

Stephanie - 00:08:02:

‍

It is, yeah, for sure. Which leads us into our next question, where my lead-in is literally with AI rapidly evolving. But it's just in another sort of context. I feel like there's a growing debate about the value of listening to consumers through passive data collection versus asking them directly via traditional sort of market research methods. I don't think that's a particularly new tension point, but I'm curious, are the trade-offs changing between these two approaches in the context of AI?

‍

Dave - 00:08:34:

‍

So it depends on the business problem that you're trying to solve, right? So I would say, so let me take a step back. The quality conversation you guys had previously, the quant data piece, I think it was Roddy Knowles, is that his name?

‍

Matt - 00:08:47:

‍

Roddy Knowles, yep.

‍

Dave - 00:08:48:

‍

He said 30 to 40% of the data is not great. We can validate that. So I would say, if you're not having conversations with your panel provider, your research vendor, your research team, you probably have some bad data in there. The other piece of this is the social listening component. We don't know how much bad data is out there from a social listening perspective. Twitter, the estimates are 15 to 68%. So that means we've got bots saying things out there, advocating for brands, misinformation. You saw it in the election. So it's going to depend on what research question you're trying to answer and the validity of what you're looking at. So I want to make sure that that's out there because that's really important in terms of what we're talking about. But I digress all to say that it depends on your business outcome, right?

‍

Stephanie - 00:09:32:

‍

So... But that's such an interesting point you're making, which is really that the same problems that exist with primary research exist with listening data. Right. That we're that it's the same.

‍

Stephanie - 00:09:43:

‍

Particularly social media.

‍

Stephanie - 00:09:45:

‍

Yeah. Yeah. It's the same problem.

‍

Dave - 00:09:46:

‍

It is. And not to get dystopian or anything, but at some point I worry that the Internet is just going to be AI talking to AI. So, absolutely.

‍

Matt - 00:09:54:

‍

Yeah, yeah, yeah.

‍

Dave - 00:09:55:

‍

Gathering it and trying to make sense of it, right?

‍

Matt - 00:09:58:

‍

The bots are having a conversation. We're on the sidelines at this point. We're the spectators. Yeah. So you mentioned in your past democratizing customer insights. That's something that we hear a lot in the industry, but it's really something that you've kind of made core to the work that you've been trying to do at Walmart and elsewhere. How do you envision AI playing a role in that goal? You know, is it kind of enabling this real time access to consumer insights or what can AI do for us when we're trying to make our research more accessible to other teams?

‍

Dave - 00:10:30:

‍

Yeah, I think at Walmart, our insights team is very small, but mighty, very skilled, right? We're often in a position where we have to scale and democratize tons of data. So think about dashboards. And I know my clients are probably saying, I need to know the dashboard. Like, I need a hole in my head. Like, we don't need more data. We don't need more dashboards. I think what I'm most excited about is when AI gets to a place where it is solving the what, and it reconciles the what. Because we have so many different pieces of data that are saying similar things, but there's nuance and chaos. If you think about Point of Sale data, you think about market share data, if you think about panel, all the different things, we have to reconcile that, and they have to reconcile it. And ultimately, if we could get to what problem they're trying to solve and one source of truth, it makes our job so much easier as researchers to help them understand the why and the who and the how. So that would be the biggest unlock and my hope for the future.

‍

Stephanie - 00:11:24:

‍

Interesting.

‍

Stephanie - 00:11:26:

‍

To switch topics to segmentation for a little bit, and I know, I mean, it sounds like earlier in your career, you were churning them out in four weeks, the old-fashioned way. So good on you. Probably don't even need AI. But we have a question in here that, you know, AI's ability to sort of analyze vast amounts of data has really opened up new possibilities in consumer segmentation. And I'm curious if, in your experience or your perspective, how does AI-driven segmentation compare to, and I think in particular, like a traditional cluster-based analysis? So not an a priori segmentation where we're just doing demo cuts, but like one where we're saying, hey, let's do cluster analysis and see what's in here. You know, maybe we're doing an append so we have extra data. But how does it compare when we talk about these sort of AI leveraging models? Is it something where we're just creating deeper understanding because we can leverage so much more information? Or is it allowing for a different type of segmentation where we can get hyper-personal and do things that we could never do in a traditional way?

‍

Dave - 00:12:25:

‍

I think it's the latter. I think you nailed it. I think in the short term, it's going to create speed and efficiency and scale. So most organizations, they want this massive segmentation that everybody buys into and they're very political. And then some SBUs have specific segmentations. But I think this allows businesses to get super granular in their business, find the variance, find that customer, that white space they're missing. Longer term, I think it does hit on some level of personalization. If I want to be provocative, maybe segmentation goes away and it gets to a level of individual personalization where we don't even need segmentation anymore.

‍

Matt - 00:13:00:

‍

Right. It's wild.

‍

Matt - 00:13:02:

‍

Yeah. I mean, that's a super interesting idea. I wanted to ask about this aspect to the great AI and market research debate, the tension between qualitative and quantitative data. I know when we first started thinking about how AI can be utilized in market research, a lot of thought immediately went to the qualitative side of things because it's a language, you know, think about from the LLM's perspective. It's a large language model. It's great understanding language. You know, it's great for qualitative data. Now we see, you know, people are building machine learning applications to really dive into the quantitative data as well, like sales and transaction data. You mentioned Point of Sale. I'm curious, how do you see it influencing the way businesses integrate the two sides of the research house, so to speak? Is AI going to be able to help us bridge the gap between qual and quant and really pull it all together in that, you know, synthesized view that we're looking for?

‍

Dave - 00:13:54:

‍

I don't have the answer to that holistically. I can tell you what I'm doing. My career has been predominantly 75% quant, 25% qual. What I'm starting to do is make this shift over to qualitative because of the ability to scale qual in a quant way that does numerous things for me. So what I'm trying to do is embed more open ends, more opportunities to code audio and video to get depth at scale in a quantitative way. And that's super powerful. I can leverage that in so many different ways. I can get quotes. I can get sizzle reels. I can start to get representative sample from quant. Now I know we had communities, but this is a new level of speed and agility and size that we haven't had access to before. And the other piece of this is the quality piece. If I see you on video, I know you're a real person. Now let's wait six months. We may get deep fake stuff and we have a different conversation. But right now it makes me feel a little more comfortable and have reassurance that I have quality sample because I can actually see the person.

‍

Matt- 00:14:57:

‍

Yeah, that's a great point. I mean, it can be thought of as solving two problems at once, right? You're verifying that the individual you're speaking to is a real individual while achieving that quality scale. I think we see a lot of energy going in both those directions in the industry right now.

‍

Stephanie - 00:15:14:

‍

For sure. Yeah, I would say it's fundamentally changing the way that we do research even here at AYTM. Just being able to do quality scale completely changes the kinds of questions that you can ask. It's pretty remarkable stuff. Okay, to switch gears a little bit again with you, Dave, it seems like in your career you've driven initiatives to improve customer experience through customer-led insights segmentation. Surprise, this is still another AI question. How do you see AI impacting not just the data gathering phase, but also that decision-making phase? And I think you kind of hinted at it before a little bit, too, where you were like, sometimes we ask the AI and then we're running the study and we're seeing, are these the same? So it sounds like an initiative you guys are at least starting, like wading into the water. Is that fair?

‍

Dave - 00:15:58:

‍

We are. And just to take a step back, when I was at State Farm, we had a process. It was inform, shape, monitor, measure. And almost all aspects of research fit into those buckets. And I feel like that's a component of AI. If I think about things, maybe five years ago, we had drone delivery. Drone delivery goes in form, right? Nobody knows what's going on. We had to do some research to inform. And then it kind of goes around the cyclical process. With that, I think there'll be levels of intervention with all of AI. So there'll be no intervention where the thing about no intervention, where a client comes in, gets an answer, comes back. The thing I'm excited most about that is a research ROI. That is the piece that I don't think anyone's ever unpacked. And what I see is we can look back and see who is making the query and what inputs went into that query. Because you think about all the different tracking studies you have and all the studies that are informing. You can go back and see how many times that tracking study actually informed a decision and tie it back and calculate an ROI. So that's what I'm excited about, the no intervention. Because you're always going to need people to inform the AI model. And some of that can be automated and efficient. But you're still going to need that piece on the back of the house. Now, the middle one is moderate intervention. I feel like that's sort of where we are now a little bit. Where it's going to get you almost there. It's going to get you the what in a really specific question in a really specific business case. And then the last one is full engagement. And that is where we have these seismic shifts in what's happening. Maybe it's tariffs. Maybe it's the iPhone. Things that just shake the foundation of what we know. And AI just has no idea what's going on. And you will need human intervention to help the organization navigate. So I think with all of that, it's verify, verify, verify. And then you have to have a complete comprehension of the inputs and the outputs with all this.

‍

Matt - 00:17:50:

‍

That's a great framework.

‍

Matt - 00:17:52:

‍

I know. I really liked that.

‍

Matt - 00:17:53:

‍

That's such great shorthand. I'm going to have to write that down. I think it's a really smart way to at least have some sort of systematic mechanism or heuristic you're using to evaluate how and to what extent you're going to utilize AI in a particular business problem. I think that's a really smart way to build intentionality into your process.

‍

Dave - 00:18:13:

‍

The thing I was thinking about this morning prior to this conversation, it's going to be interesting to listen to this conversation in six months or 12 months.

‍

Matt - 00:18:20:

‍

Totally.

‍

Dave - 00:18:21:

‍

Because things may be completely different. It might not age well. They are going so fast. That's right. But it'll be fun to listen to. You know, we had some good ideas, right?

‍

Stephanie - 00:18:29:

‍

That's right.

‍

Matt - 00:18:29:

‍

I love it.

‍

Matt - 00:18:30:

‍

So do you see AI predicting consumer behavior or preference that human researchers might not be able to identify? How would that change the way the business responds?

‍

Dave - 00:18:42:

‍

Candidly, I think you have to have the right inputs and resources. I think certainly human beings could uncover the same ground, if not more predictive with insights. But all of that is a dependency rate. However, AI is going to expedite things, enable thoughtful iteration. And sometimes it may be met with apprehension. And I'll give you an example. I'll go back to 2008, the mortgage crisis, Michael Burry. He was looking for data points that no one else was looking at. And he was surfacing those things. And, you know, I think things like that, you're going to have to have human intervention to validate, advocate, and influence. AI is not going to do that. It's only as good as the person who's ingesting the information. And it can be easily dismissed. So, I didn't really answer your question. I kind of just danced around. But did you like that?

‍

Matt - 00:19:29:

‍

No, that's great. I think that's great perspective. I mean, it's so hard to tell where things are going to be, to your point, even six months from now. It's just important to always keep that human in the loop. It makes a lot of sense.

‍

Stephanie - 00:19:44:

‍

I'd also just watched, I think it's The Big Short, right? Recently. So very fresh for me. I was like, hey, I know what you're talking about. Yeah.

‍

Matt - 00:19:53:

‍

It's an interesting anecdote. So last AI question from my perspective. What role do you see then for human intuition and expertise in the field of primary research in the future? Or do we feel like we already got that one?

‍

Dave - 00:20:08:

‍

Well, I've got answers for all these. I know there's a little bit of duplication, but I can answer it if you want.

‍

Stephanie - 00:20:13:

‍

Yeah, go for it.

‍

Dave - 00:20:15:

‍

To me, AI is the new calculator. It's the cheat code. And I think that that doesn't mean it's not important to understand the principles of math. I think researchers are still going to have to understand how all of this works. And I think the most valuable researchers in the future of AI are going to be the ones that aren't afraid to push the boundaries of what's possible with AI and have the presence of mind to know when to intervene. There's always going to be, like I said, there's always going to be disruptive moments like COVID and things like that where we're going to have to step in.

‍

Matt - 00:20:42:

‍

Yeah, that's great perspective. Yeah, it's like, you know, when I get that question sometimes, you know, like, hey, Matt, what do you think about AI? Is it going to take your job? I'm like, I don't really look at it that way. I mean, as a researcher, I talk a lot about how, you know, I'm meant to be curious in my role. You know, curiosity is important. Like, you should be curious about AI. Like, learn about what it can do. Learn about how it can, you know, upskill you as a professional. I think that more and more folks are starting to take on maybe a slightly more optimistic tone towards it in our industry than there was maybe a few months ago when we were in more of a doom and gloom phase. That fear might be ebbing. But maybe I'm just projecting, but that's my thought.

‍

Stephanie - 00:21:23:

‍

I think it's so true, too. Like, and sorry, I just jumped in like I'm the guest here. But I really think our industry is pretty, like, embracing of AI. And I really say that in, like, contrast to maybe the arts. And I understand why there's a difference. I really do. Like, I don't think this is an apples to apples situation. But it always surprises me when I encounter people who are not pro-AI because, I feel like we work in an industry where it's very much embraced to the extent that it can be.

‍

Dave - 00:21:54:

‍

And I don't understand why it's so divisive. Just my take. But, you know, you see college professors saying don't use AI. Or I posted on LinkedIn recently. If you use AI in an interview process, you'll be eliminated. That's the point. Like, I want somebody that is a guru using AI. And if you're using processes that can be gamed with AI, change your processes. That's a you problem. So, I'm excited about it. And I think when we get into leadership positions, it's important to be the old person in the room saying, I used to manually code open ends. And I used to do text analytics. And this is way better. Right? So.

‍

Matt - 00:22:31:

‍

Yeah. That's such a great point. You don't see programmers saying, I really wish we still had punch cards. Like, I don't see anyone saying that.

‍

Stephanie - 00:22:39:

‍

No, we have a duty to inform. You're right. I love it. Well, Dave, to wrap this up, we always like to ask a couple of questions. The first one, especially after one of the last things you said, I feel like I might have an idea, but maybe you'll surprise me. What is one piece of advice that you would offer to someone who's just coming into the field of, you know, consumer insights?

‍

Dave - 00:23:01:

‍

Can I give a couple?

‍

Stephanie - 00:23:02:

‍

Of course. Yeah.

‍

Dave - 00:23:04:

‍

Yeah, I think you're going to need to be comfortable asking a lot of questions. Obviously, Curiosity podcast, right? So ask questions about things you don't know. Ask questions about your client. Everybody's there to help you. I think that's critical. The second part is be prepared and share a point of view. So if you're sharing insights, be ready to have a point of view in anything you're talking about. The next one would be think like a customer and your client. I think where those two meet is where you can start to anticipate their needs. So with everything happening with tariffs, you know, think about what questions your client's going to ask. Think about what's happening in the market and how that's going to affect the customer. And then you can get in front of research plans and insights that are going to inform your client to help the customer. And then lastly, allocate some time to be proactive. Test and learn. Conduct research on research. We're not doing our jobs if we're not trying to break something. So that would be my takeaway to a new researcher.

‍

Matt - 00:24:02:

‍

I love that.

‍

Matt - 00:24:03:

‍

Skate to where the puck is going to your point. And my last question is always, you know, what's one shift that you think we'll see over the next few years and how brands are using consumer insights to drive strategy?

‍

Dave - 00:24:14:

‍

I think we'll see much faster threat and opportunity detection. At State Farm, it was kind of primitive, but we called it event management. So it was kind of a situation where it's like, if this happens, then this. So think about like a claim, then it would execute ABCD. I think from a brand perspective, this will sort of not replace brand management, but you're going to have all these things in place to manage crises, but also opportunities. Think about Wendy's from a social media perspective. Think about recalls, all the things you have to manage. AI can step in and kind of run point and if not do things whenever, you know, that trust is granted. I spoke a lot about this earlier, but some of this could entail some facets of Research Planning in terms of that shape and form monitor measure. So if you get in a brand situation like, hey, we need to do a brand refresh. Here's what you need to do, that kind of thing. So.

‍

Matt - 00:25:06:

‍

I love that as it thinking about AI developing into this sort of like tactical awareness power up. I have not heard that perspective yet. That's interesting.

‍

Stephanie - 00:25:14:

‍

Yeah, same. Well, Dave, we appreciate your time so much today. This has been a fascinating conversation. Can't wait for it to come out so others can hear it. Anything else from your side?

‍

Dave - 00:25:25:

‍

No, thank you guys. I appreciate the time. Keep doing this. It's great. We only get better if we talk to each other and share ideas. So thank you very much.

‍

Stephanie - 00:25:34:

‍

So true. Thank you.

‍

Stephanie - 00:25:36:

‍

Thank you.

‍

Stephanie - 00:25:38:

‍

The Curiosity Current is brought to you by AYTM.