Drew Smith is Vice President of Global Data & Analytics for Little Caesar Enterprises and Ilitch Companies. Working for the the third largest pizza company in the world, that also owns a major league baseball team and a national hockey team, and having spent seventeen years at IKEA, Drew has deep experience in applying data and analytics to drive the business forward to deliver results, build sustainably, and change how people work more effectively.
Welcome to the Soda Podcast. We're talking about the Data Dream Team with Jesse Anderson. There's a new approach needed to align how the organization, the team, the people are structured and organized around data. New roles, shifted accountability, breaking silos, and forging new channels of collaboration. The lineup of guests is fantastic. We're excited for everyone to listen, learn, and like. Without further ado, here's your host, Jesse Anderson.
Hello, and welcome to the Data Dream Team Podcast. My name is Jesse Anderson. My guest today is Drew Smith. Drew Smith has done all kinds of things. He is currently the Vice President of Global Data and Analytics at Little Caesars Enterprises. Before that, he was at IIA, International Institute for Analytics, and even before that, at IKEA. Welcome to the show, Drew. Would you mind introducing yourself a little bit more?
Sure. No, thanks for having me, Jesse. I think what you said is mostly accurate. I've done a lot of things. Some of them have been good. Some have been less good, but I've done a lot of things. Happy to also add to my title, I am not only responsible for global data analytics at Little Caesars, which is the very famous pizza chain, but I'm also responsible for that role at what's called Ilitch Holdings.
So that's the family that all owns Little Caesars, but they also own the Tigers which is a baseball team, and the Red Wings which is a hockey team. And then we own about 500 acres of mixed use development down here, including big arenas and stadiums. So it's a really diverse business, which is super energizing from a data geek point of view. The pizza business, of course, is fascinating, but it gets even more so when you add those other companies. So, happy to be here.
Well, thank you so much. I am excited to dive into that, because I didn't realize that other part. So I'm curious about those other manifestations. Let's start at the first company that I mentioned: IKEA. Could you talk about some of the lessons that you learned there at IKEA?
Yeah, sure. So, I'll go from the long stuff, long, long ago. I started off at IKEA in the stores after graduate school, actually. I started off with focusing on the supply chain. We were building up new stores, and then I went over to the corporate office, and at the corporate office, I was involved in a bunch of things. And in 2008, when the market collapsed, the first time we had a big, big market change, we had a couple of ideas. And it was either hide under your desk and just use the cash that you have, and ride out the storm. Or the other area was to kind of go on the offensive.
So my boss at the time, who was such a good big thinker - maybe not great on details, but a great, great big thinker - he sort of said like, what if we didn't have static pricing? What if we didn't have this same price all the time, which was then the gospel of IKEA. Yeah, we could do that. There's lots of people that do that. And then we started digging into it. So there was a lot of big data research, and then there was a lot of testing. And that was sort of the first thing I learned, was the power of getting more data through sort of active testing. So we were trying different sales offers - a crazy deep price, and a not so deep price, and a combo price, and all this stuff that we hadn't done before. And for the explicit reason of just learning, in the first sort of three, six months of a year.
And then we were able to really work with a small team and quickly realize and capture that learning and try new experiments. And eventually we got to a point where we had a really quite deep knowledge of which price points had the greatest price elasticity, and for how long we should run any offer. So we got really smart really quick just by trying stuff and by piling a lot of data, but also by having the right people in the room to discuss the results. And that was math. We were doing a lot of statistics on that. There were also commercial people. It was supply chain people who said, I can't get any more of that stuff even if you tried.
So really, I remember the energy, you can kind of hear it a little bit, that revolved around the data. That was really insightful for me, because before that it was very much descriptive analytics, what have we done? Now we were talking about what we could do. And that was really exciting. So, it's called predictive analytics. I don't know that I knew it was called predictive analytics in 2009, to be honest with you. Maybe I should have, but I didn't. But it was really good. It was really fun. And that was really an eye-opening experience for me, that I had a talent to bring people together and to do these things, and that these things had a really big impact on our business. So I was excited for that. And that was my first big foray. And then of course, you know, when you do something good with data in any business environment, they ask you to do more.
Could I ask you a quick question on that? Was that price elasticity, was that just online? Was that in stores too?
No, it was only in stores. I mean, it was just, this is in the bad old days. The iPhone was two years old, you know? So, I mean, this was only in stores. This was, this might even, and I can't remember the details, but this might have even been like, printed promotions in the Sunday newspaper kind of thing. I mean, this was really down and dirty retail, but it was so unusual, 'cause IKEA had never done this. IKEA had always said, okay, we print the price with our big fat catalog - which we used to send out - and that's the price for the whole year, and you can never change it.
And because of the crisis, my boss sort of saw an opportunity in his wily political way, like, hmm, maybe we can break the rules, because everything's kind of going to chaos. And I thought that was really inspiring, too, but that's another story. So no, this was just down and dirty, you get an email that says, hey, 50% off a very famous product for three days, and then we track the difference, and we put a different offer in LA than we put in San Fran, than we put in whatever, and we were able to test things. And so when you do these things, typically, at a big company, people will find another place for you to do them. So then that was like, well, if you can do this with all that data and stuff and all that math that you worked with, what else can you do? And what we could do is focus on the goals of the company, which is sales growth, margin growth, supply chain optimization. It just became like, okay, where can we point this capability?
And we did that for a few years. And then that became an important enough thing on a US basis that they invited me over to Sweden, and I worked in Sweden where I did that again, on a global basis. Really focused again on pricing. Pricing on a global basis is really hard. You have currency transfers and all that stuff, but you have so much richness when you're talking about what is a low price in China versus what is a low price in Norway. Middle income in Norway is seven and a half times middle income in China. So it was really interesting to get involved and bring in external data, stuff like that.
And then, one of the things I did, which in retrospect was really good, is I kind of theoretically stepped out of data for a couple of years. And I owned an entire business category, so I was responsible really from raw material to customer returns for the IKEA product category that is like, bookcases and TV storage and stuff like that. Which included some billion dollar sellers and some really big names. And of course I used data on that. I mean, I used data to first examine the state of that business, and then I used some data to sort of look at different customer behaviors in different countries to find out which products likely had untapped potential.
'Cause it's easier in a product development organization to sell more of the thing you already have than it is to create a whole new thing. So you kind of have to build two engines at once. You've got the sort of revitalization engine, and then you've got the renewal engine. But that was really awesome because I think, for me, I was always focused on the goals that we had set in that particular business. It was about 4 billion US dollars at the time.
And it felt really, when I jumped back into leading analytics, which I did my last role at IKEA, I really had a feel for, okay, you're a business owner, I know what your daily life looks like. I know the pressures that you're under. I know that even with the most convincing statistics I can provide you, that might not be enough. We need to have a deeper conversation. We need to really understand what you're going through. So I think that knowledge that I had, working in that product category, has helped me keep a focus on the outcome of the analytics to drive the business forward.
And I think a lot of analytics leaders have that, and might even have it more than me without that adventure into sort of the business side, for lack of a better way of saying it. But for me, it was invaluable to have that, as I say, adventure on the business side. That was the IKEA journey, and as I say, the last role in IKEA journey was to start up a global enterprise analytics. We had lots of functional analytics, some of which were really good, but on a global basis we really had nothing. And we really had a lot of disparate systems, we had different methodologies, and so when you were looking across countries for best practices, it was really hard. Every country in a sense was its own island. And that was, for me, a wasted opportunity. So, we actually stood up to some global practices and some global ways of working.
What are you most proud about from your time at IKEA?
Oh yeah. That's tough. 17 years. I mean, to be frank, one of the most prideful things is actually one of the product things. So there's a product out there in the IKEA ecosystem that became the fastest ever selling top five product, a product called EKET. And there's a lot of innovation baked into that product. But what I'm really proud about when I think about that product is how much of a team build that product was. So, you know, analytics is a team sport, which we know and we discuss, but so is product development.
So I was most proud of the fact that we just had this team that just, they just clicked. They were really supportive of each other. They were really supportive of each other's, you know, unique competence. They challenged each other, and that's why the product is good. The product is good because of the people involved in the product development. I mean, that sounds obvious, but sometimes that's a little bit easy to forget. And I think it's of course true in analytics as well. An analytics product that is good, you can trace back to the people that were involved in the creation of that. So that's something that I was really proud of, because it's not only a great product, it was a great experience. And it's great learning for me, it's something I carry with me every day.
And then the other thing I have an interesting mini mini thing I'm really proud of, because we pivoted on this analytics product in a way that made me, I remember being nervous, maybe even disappointed at the time when we pivoted, but in retrospect it was super smart to pivot. So if you've ever shopped at an IKEA store on a Saturday, it's quite busy. And the line before you check out can be quite long. And that's frustrating. It's frustrating at any retailer. Could be Walmart, it doesn't really matter, 'cause by the time you get there, you're like, I wanna get outta here. I'm done. Right? I just wanna go home. So we started talking to the store managers and we said, well, how do you deal with this problem?
And I had been a store manager. I remembered a little bit. And they're like, well, you know, we can always kind of shift. We can always respond. We can ask someone to take their lunch break a little bit later. We can ask someone to stay a half hour earlier. If someone's local, we can even call them in earlier. But you know, there's ways we can react. It's just that by the time the problem hits, it's too late. The tsunami has hit and the beach is underwater.
So we said, okay, well we have all these sensors in the building. Not tracking people directly, but just tracking crowds, so, you know, basically heat maps and stuff. And we know that people just follow a relatively established pattern. So we should be able to say, like, if you have a big crowd here, so many minutes later you're gonna have a big crowd at checkout. And they said, yeah, that's true. That's what they experience. And even the human intuition, some of the best store managers will tell you like, yeah, that's why I'll call kitchens or I'll call bedrooms and say like, are you guys super packed at your planning station? Oh yeah, we're crazy. Okay, then I know.
So we took that sort of thing and we used the sensor data and we actually started to build in some other data, like external data, like weather data, traffic data, in the context of London, underground volume, stuff like that, which is available, publicly available on API. And we were able to actually build a prediction engine that would say basically within a certain amount of time, you're gonna get crushed at the cash lanes. And I thought this was super cool, so I started like, my mind started going, oh, we can do little dashboards, little needles. And like, we can predict what's on the cart, because if the cart's really full, that's gonna be a "go to delivery." And I started really geeking out on this. And a couple of us in the team were thinking the same way, 'cause it was super fun to think about this stuff. And a store manager, one of the really good ones was like, yeah, that's not necessary. I'm like, yeah, but look what we could do. And he's like, all I need is for like - and he didn't know what he was saying specifically - he’s like, I just want a text message that says I'm gonna get a big line at the cash lanes in like, 20 minutes. That's where I said I was disappointed, 'cause I was like, oh man, but I could do such cool stuff. But he said it with such conviction that it was obviously the right thing to do, was just a simple text message.
So then the only other question I asked him is, okay, what is the right amount of time? Like how much time do you need to respond? It turned out he needed 35 minutes. 'Cause he tested that over a couple Saturdays. At like, 20 minutes, he couldn't respond. So then we just built a simple engine that basically pinged his phone if he was in the store, or pinged the store's sort of manager phone, which is a special phone that they get, which is connected to all the systems. And it just said, you're gonna have "an unmanageable queue" I think is the term they wanted to use in British English, in 35 minutes. And it worked.
So I'm proud of that, because I know we could've done the wrong thing. I know we could've just ignored the stakeholder and built this really cool dashboard. And I know it would've been cool. It would've looked cool and all that stuff. But at the end of the day, what we gave people they really, really liked and was useful. It helped them do their job better, and it solved a problem - it didn't solve, but it helped address a problem that for 25 years has been the top three complaints at IKEA, is the length of the line. So you're, you're helping people, you're focused on a problem that the customer says is a reason I do or don't come to you. And I don't know, I have a hard time believing you can do better than that.
I love that story. And I'd agree that that would be something that you wanna be proud of. I love that you really focused on what the customer needed, because what you've talked about, what you originally had, that would look better on a resume. Much better on a resume.
You could add a little screenshot and all that stuff. Yeah.
Yeah. What doesn't look as cool on resumes is the person actually used that information that I gave them. So that's what I'm taking out of that, out of your stories. It sounds like your mix of experience in the business was just the beginning; actually talking to the people, communicating, that's really that key. I love that story. Thank you. So looking at your time at International Institute for Analytics, that was where you and I first met. Tell me more about what you did at IIA.
Yeah, no problem. Yeah, it's where we met. You had been an expert there and you had, I think, spoken at a couple conferences and stuff like that. And I ran a group called the Analytics Leadership Consortium, and that's basically a subscription group of analytics leaders from diverse companies, so no competing companies in one group, so that we could have folks like you come and speak to them about challenges that they had, about real time architecture, or data teams, or new new skills in data engineering. And they could ask you questions.
And they could ask each other questions without a concern that trade secrets are gonna leak, because it's just one CPG and one financial services firm. So, no big deal. But the interesting thing of course is that the lessons are so common, right? I mean, you have very, very different products, and some even different work cultures or something like that. But in terms of like, how do you build a team? You know, how do you build a data strategy? How do you transform people's cultural mindset to think about data first? So that type of stuff is what we did at the Analytics Leadership Consortium in the IIA.
What made you leave IKEA to go to IIA?
Yeah, I mean, that was really obviously a very, very tough decision at 17 years. I wasn't one of those persons that thought I would be at a company so long. And in fairness to IKEA, it's not the same company. I was in many, many different types of companies. It always felt a little bit different, but I always had this aching feeling like maybe I don't know enough of the world, to be honest, even though I'd, ironically, lived overseas. So I kind of wanted just different exposure. And I knew in other meetings that I had with other people, at events in Europe that were similar to the ALC, that yeah, we all face the same challenges and there's a lot of institutional knowledge in people. And what you know about data, maybe I know about something else. And what someone else knows about engineering, someone else knows. And I thought that was really powerful. So I was attracted to the concept of the ALC, 'cause I thought it was so powerful.
And I was interested in taking a move to another direction. And, I felt like at IKEA we had switched from a delivery, what I call, "think big, act small." And they had really become enamored with a total sort of digital rebuild, if you will. And I saw the value in that, but it's not really where I wanted to be. I kind of like to build the product that predicts a queue, and then we use that technology elsewhere and we could use that learning elsewhere. But I'm more of a - deliver results, anchor the learning, build sustainably. And so we had a bit of a difference in the way they wanted to approach things. So it's also just, yeah, this isn't the place for me. I couldn't get behind that approach, 'cause I didn't think it was right. It turns out that it worked okay.
So when you have a situation where you're like, in my opinion, either you're in - especially in a leadership position - you’re in or you're not. Like there's no sort of, kind of, maybe. So there are many factors that moved me that way. One was a super keen interest in the concept of shared, two was a belief in the IIA guys and Bill Franks in particular, and three was just a change in direction at IKEA in the digital space that it wasn't for me.
So you mentioned those patterns, that that was really what brought you into that. What was the most interesting problem that you saw in that discussion?
You mean at, at IIA? In the ALC? That's a hard question to answer because of the width of problems that people tried to solve for. And what I mean by that, so I actually kind of really liked the debates, even though it became a bit dorky and a bit semantical around data literacy and data culture, and are these really things? And data driven and these sorts of discussions. And the reason I like them is because at the end of the day, people did agree that in most of their businesses - so these are legacy companies, they aren't digitally native companies by and large, and even if they are digital companies, we had a gaming company, so a video game company you would think is digital. It's not that digital - it did come down to changing the way people did their work. And that is really big. And you need to appreciate the size of that, because you are essentially saying to people, we used to do things this way and we are probably successful. We could be the largest CPG in the world. We could be one of the largest investment companies in the world, but now we need to do things another way to continue that success.
And that is a really hard thing for people to hear. And you have to be careful about how you say it, right? Because essentially, you could be perceived, and I heard people understand that they were perceived as the "hey I'm here now, so I'm gonna fix all your stupid decisions," right? Like nobody really means that, but you can understand if people perceive that. So I thought that was really interesting because that's a people problem, and those are way more complex than any technology you can imagine. Right?
So I thought that was really interesting, the people challenged the nature of it. I also found it really interesting and I think it was a bit of a red herring that people were really obsessed with organizational models. Centralized, hybrid, decentralized, like what's the best organizational model? And I often found when people were asking that, they really weren't addressing the root cause that they didn't have their data in good shape. They didn't have the right talent in the team. Like, they were looking for the shape of the people to make the business work when that's something you add or something you adapt to as you go. So that was interesting to me that people were so obsessed about that, because a few of us - it wasn't just me, a couple other folks said things, they're like, I don't worry about that. I don't don't know how to describe my organization. Here's what I do. Here's how people set up. Here's how we deliver projects. What do you call that? And you could say, oh, I know that's exactly called a hybrid model. They're like, I don't care. But so there was a really interesting dichotomy there.
And then finally, I'm very, very interested in technology, even though I'm least capable in that area. And the reason being is I think it's really hard for people to pick the technology that's right for them, because it can become a big sort of ego thing, because they don't know. Because it's really hard to wrap your head around. So you have to find people that you trust. You have to talk to those people. I always enjoyed having you come in, we had a couple other experts come in, because it's like, I don't really have any smarts that could outsmart you on the technology side or the engineering side. So I just rely on asking good questions or asking the team, the group to ask good questions, to sort of provoke your deepest thinking. That's the thing you can do when you don't know something, is ask good questions to provoke someone's deepest thinking, and share very deeply your situation.
Because normally, when you have a technology, the current understanding of your current situation is almost more important than some sort of future dream world. The future dream world is already out there, it's architected by people who've done this before, but no one really understands your current landscape as well as you do. And it's this bridge between the current to the future that's the hardest to build. And I don't think it's 'cause the future rock is very stable. It is very stable. It's 'cause the current rock is like, very hard to get people's hands around. So the technology thing was interesting to me because it seemed to be so important, and people seemed to be so worked up about it.
You know, Drew, as you were talking, I am reminded of every single reason why I wanted you to be a guest and so those of you who are listening, go back and re-listen to what Drew just said. Re-listen to it several different times. Those are the same things I see. This is exactly why I wanted Drew on, because these are common problems, and we spend way, way too much time on them. There are other better problems, there are other fundamental problems.
The shape of your team, get your team, the members of the team right, and then you can shape it, for example. The technology, if you don't have the right people on that team, the technology team, it doesn't matter what technology I do. So I can actually share that. So some of the, some of the meetings that I had with some of the IIA people, the problem was not the technology. It was, yeah, your team's really not going to be able to use this, unfortunately. Because they don't have the skills. There's a lot of reasons. And so that's usually that step you want to take. So thank you so much for sharing there, Drew.
Absolutely.
Related to that, what was the most clever solution you saw?
So that's a super good question. So let's think. I wanna think about this a little bit. Yeah, I mean, we weren't really solution oriented, so I'm having trouble coming out with one exact example. But I'm trying to think about people, 'cause one of the other things we did at the ALC is oftentimes people would present what they have done at their company, and one company I think - I always feel like you can say the company's name when you're gonna say nice thing, so I feel okay - one company that I thought was just really good at focusing on any given problem and focusing and getting themselves in a good spot - not the perfect spot - for that, for fixing that problem was steel case, which is a commercial furniture manufacturer actually based in Michigan. They would be that way.
I felt like when they had a problem, for example, they had a problem with getting the engineering talent that they wanted, and I think all managers put their brains on that, and they recruited and they trained and they created an ecosystem that made it easy for data engineers to come into. I felt like they were just really good at focus. I know that sounds like an evasive answer to your question, but when I think about them, I think about how they solved three or four problems that are common. Going from the whole data scientist who does everything, and then you realize that doesn't work, 'cause some of them are good at data science and some are good at engineering, and that doesn't mean they like either one, to really recognizing as soon as they recognize the problem, the whole management team would kind of just put their focus on that. They would bring that into our conversations, they would ask for additional conversations with peers in the network, they would make phone calls with IIA experts, they would do whatever they could. And they would really focus.
And I really admired that, and they're smaller. They're like, I don't know what they are, they're public, so I think they're around 4 billion, but I still think that that focus and they solve the problem of teams immigrating higher levels of engineering capability through that focus. Because it didn't become like a maybe and a delegated out. It was like, okay, we're all gonna stay in this. We're gonna make sure we have the right structure, we've got the right job profiles, we've got the right things they're working on, we're listening to what they need from a technology point of view, we're moving in that direction. It was like all focus in on that. So I think they were really good at fixing things extremely well by staying focused.
I love that, because you might know this, but Steelcase was a client of ours, too. So I know Jorge listens to this podcast. Hi, Jorge. I loved working with Steelcase, and it was for that same exact reason. And it wasn't just, you saw that because the upper management was focused. So I worked with the individual contributors. We helped them set up the team and get those people with the right skills in the right places. And yes, their focus was incredible. And that's something I preach in my consulting as well. Focus on a few things, get them right, do them, execute them. And then once you've done that, you can move on to the next thing. But if you do 10 things all at once, guess what? You're not gonna be able to do 10 things all at once. And they, yeah, they focus so well. So I was really proud of the progress that they were able to make. It sounds like you got that sense of pride too, working at IIA, too.
Yeah. I mean, I loved my clients at IIA. I found them really, it was funny because when I was recruited over, I was talking to the woman I worked with, a woman named Lise Massey, who's awesome. She's now at Nike. And she said, oh, you guys said, you know, what, how do you like the job? And you know, it was sort of a peer interview. I think I had already had the job. And she's like, I really like the people that are in our ALC. I was like, oh, that's like some soft Portland hippie crap. That can't be true. They can't all be that nice. And of course they weren't all that nice all the time. But by and large, I loved the fact that people came to those meetings energetic to learn, energetic to contribute, open to other people's ideas, you know, just motivated. And they were just good people in that way.
And I never had a situation where somebody would say, hey, so and so at this company or so and so at that company said something really smart. Do you think I could reach out to them? I was like, yeah, of course you can reach out to them. And every time it was like, oh, thanks, that really helped. I had a conversation with them in like, 20 minutes. So I really like the people, and I like the setup. I think the setup sort of attracted those people. Sort of people who were in that way. And that's my mindset too, so that was super fun. And I think what I was proud about is that we were really responsive to what people wanted to focus on.
There are trends that happen. There are evolutions that happen. As I mentioned, I was a little skeptical about this data culture thing, but, you know, it kept coming up and kept coming up, kept coming up. So, you know, the team at IIA was Mark Demers and staff, did some research and it was really great stuff. And then I was ready, and so we really listened to what they wanted to work on, and we really focused on the problem they wanted to solve. And I was really happy because most times we got good positive feedback. Sometimes we failed. A speaker was off or I misunderstood what people wanted or what have you. But I was always proud of the feedback, 'cause the only way we knew if we were doing well from that much arms distance really, was if people kept coming back, if people kept renewing, if people kept saying, oh, that was a really good session, if people were energetic and suggested other sessions. So I think I was proud of the fact that we seemed to help people. And that's the whole purpose of that, was to help people.
And listeners, if you aren't hearing a pattern in Drew, it's always listening to feedback. You do an incredible job of that, Drew. So I'm thinking of all the things I need to start listening to better feedback.
You wouldn't be where you were if you weren't, but I think also, but it's true Jesse, but I think also I'm a weird person 'cause I actually like people. So like, I'm really interested. It's like one of the things I like about my current job is I meet people in so many different fields, but I think at the heart of it, analytics is there to solve problems. Right? And it's just, the thing that's forgotten is that there's a comma at the end of that for people, right? So we solve problems for people. And if that solution doesn't really meet their need - either it doesn't work, which is always possible, or it's not particularly understandable, which is oftentimes the case, or it pushes them into a discomfort zone that you need to help them with, then you still have to deal with that "for people" part of that sentence. So I think for me, it's also just a, maybe it's a natural inclination, but it's also the way that you are effective in analytics is by making sure that people can use things. You're there to solve problems for people. So you have to express an interest in those people, and generally speaking, the best way to do that is to listen to what their challenges are.
Cool. So speaking of your current position, I was completely surprised when you went to Little Caesars, but now as you talk about the position more and more, I'm thinking, oh wow, that would be pretty interesting. So tell me more about this position and why you chose it.
Yeah, sure. So, I'll also answer, I'll say that the reason I left IIA is, as much as I love my clients and I love knowing that we got good feedback, I really kind of ached for that like, okay, you actually put the thing, the learning into practice, you know what I mean? So, somewhat jealous of Jorge and those guys at Steelcase, or Craig at McDonald's or whatever. Just jealous that they got to kind of then go after the meeting and really try it and see what happens. So that I could never get past, so eventually I had to say goodbye to IIA.
Yeah, the reason I was interested in Little Caesars was actually two, two main things, to be honest. Maybe three. So the first of which is that like at IKEA, I am the first enterprise data and analytics leader they've had at Little Caesars and what's called Ilitch Holdings. And when I interviewed with the CEO, I said to him, pretty straightforward,I was like, hey, you know, was it 2021? Yeah. 2021. I was like, you're a little late. You know that, right? Like we've got work to do to play catch up, There's this competitor down in Ann Arbor, in Ypsilanti, that would be Domino's. That's the number one pizza business, and that's always the people you chase. And he's like, yeah, I get it. But what I liked about it was he took accountability. He said, we identified this need maybe 2014, 2015.
And then, you know, we had this change in the business and that change in the business and we built Little Caesars arena and that takes time. So I love that he was sort of like, yeah, it is what it is. It's fine. We are behind, but you're gonna help us get up to speed and deliver world class analytics out to our whole company, including our franchisees. So I was like, okay, I like the size of that. I like the honesty within the CEO. And so I liked the fact that it was new, it was challenging, it was a ground up kind of thing. And the other thing that I said I liked is, we not only own the third largest pizza company in the world, but we own a major league baseball team and a national hockey team, which are quite big enterprises.
And we own the stadiums and the arenas in which they play, and all sorts of different properties that the family was a big supporter of this town that I live in, which is Detroit. So they really wanted to revitalize the town. They spent some of their own money and worked out to be a good business as well. So I was really interested in that because - and what's funny is people were like, oh, you must be a big sports fan. I'm like, nah, not really. I'm as interested in, like, how do you optimize parking? How do you, we own the parking arenas and like, well, what should you charge for parking if it's Billy Joel, which is a bunch of like 65 year olds with high income versus if it's Dua Lipa, which is more younger people who are perfectly willing to walk past the bar. I'm really interested in those problems. So I think I was interested because it created more problems than we have at the pizza company, which are sizable and which we can do a lot with. But these - the diversity of problems was super interesting to me.
Again, learning a little bit from IIA, I know what we learned about predicting pizza sales and optimizing pizza sales we can apply to predicting and optimizing parking. So, very different things, but I think it's possible. So that was number two. And then number three was sort of the company culture. It's a family company, and it's sort of a really heart driven company.
I think the founder, like many of early 20th century founders, founded in 1959, was just one of those people who would just take amazing risks and just try things. So, the way we sell pizza is different from others. You know, he bought the hockey, the baseball team in 1982 because it was gonna go under. He moved the company back into downtown Detroit when everyone else was leaving. So this sort of like in it with a heart, and he's a sports guy, so he was competitive. In it to win it. That kind of culture I really like, too. So those three things really motivated me to say yes, when they reached out.
Another reason why I wanted you on the podcast was that deep experience. So now that you've switched and you have this experience at IIA, you had this experience at IKEA, what lessons did you take there?
Yeah. I took a lesson that I'm having trouble keeping to, which is, this takes time. So I've been now with Little Caesars for about nine months, and there was a bit of strategy creation in the first bit. And then we've finally got some good first hires, building out the data team. And I'm like, that was maybe in April, and now I'm, like, already anxious to get products out. So, I took the lesson of being patient, and all of my friends at IIA said, hey, this is a couple years. It takes two, three years to get this really up and running when you've got a big company, which we are. But man, I'm like, I'm trying to learn that lesson.
So I have that lesson, I have to remind myself of that lesson on a daily basis when I'm meeting with the guys building analytics products, and "guys" being a gender neutral term. So patience and planning is one of the other ones. But as I said, my reason for leaving IKEA is I do want to deliver smaller things that enable larger things. I think that's important. I think it's important because we are in an organization that doesn't really know how data can help them at the way maybe you and I know, or my friends and IIA and all the ALC colleagues know. And I don't really believe any amount of me doing a presentation or a tour is gonna help. But I think when we put in their hands the version of that thing we put in the hand of the manager at the IKEA store, the checkout, the queue predictor, then the light bulbs will just start firing.
I know this from experience, that it's far easier to give people an example of the thing that you're talking about that data can do. It can help you add math, computing, and prediction to something you already do, which is try to slow checkout lanes or try to make sure that pizzas are hot and ready. When you show people it, then they will get it more than you think. I remember we did a big thing with the supply chain at IKEA, but it was narrowly focused on old merchandise that we were trying to sell through. So, you know, you stop selling merchandise, so you can bring in fresh merchandise.
And, you know, it had a lot of math behind it, had an ensemble model and all that fancy stuff, but at the end of the day, what it was was a forecasting and a pricing module. And this dude who must have been like all of 23 years old came to me and said, I didn't understand it all what you were doing, but now I see that you could probably do this for regular prices and forecasting of news, if you could forecast olds. And of course he's a hundred percent right. But there was, I couldn't communicate that until we showed him something else. So I'm trying to be patient, but I do know the value of getting tangible, relatable products out to users and getting their feedback. Making it better, but also helping them expand their field of vision.
So they go, oh, this is really interesting. And this looks like a solution to a problem I think I have somewhere else. I can't predict coworker churn or I can't predict the effectiveness of a given limited time offer. If you can predict this, can't you predict that? Yeah, but let's talk it through. What are we really trying to solve for? Is this a big problem, a small problem, a national problem, a local problem? So I also know that I'm just anxious to get those things out, because I'm anxious to have the energy go from what happens in analytics - which is a push, here's, what we can do for you - to a pull. Here's what we need from you coming from the business side. And when you make that switch where it's a pull from the business side saying, here's what we need from you, and you organize that properly, it's just awesome. It's just so much fun, because you get people energized to be a part of the story. You've got people energized to contribute resources. It just flips the switch a little bit. So, looking forward to that, but I gotta be patient.
Now let's invert that. What did you decide that you didn't want to take to Little Caesars/Ilitch that you were doing at IKEA?
Oh, yeah. That's a good question. Well, I'll tell you one thing I haven't yet decided. Because I don't know that I can answer that well, but what I can tell you is at IKEA, what we wound up doing was having focused areas in different geographic regions. London, Philadelphia, Shanghai, and Malmo, Sweden, which is in the south of Sweden. Focused on different areas of analytics, so marketing, supply chain, stuff like that. And we have a good, big international presence at Little Caesars, and I'm not quite sure whether that model carries forward, in part because I left before it really got off the ground, although I know people who say it is working. I just don't know how much of it is the focus of that group on a core area of the business that's most focused in that area or whether it's sort of adapting to cultural norms, which are easier to do when you have a team sitting in a given location.
So we're extremely big in Mexico, Little Caesars. We're growing in Europe, really good in South and Central America. So I don't really know. So what I have not taken from IKEA is the model of placing different analytics teams in different geographies. And I'm cautious of that, because I don't really, I don't think I know if it's successful, why it's successful. So I don't know how - the international expansion, when we support international expansion more deeply than we do today, I'm not quite sure I know how to do that, if I'm honest. So I'll have to look elsewhere to find some inspiration for that.
Okay. Well, speaking of inspiration and data leadership, you talk about finding your voice as a data leader. What is your voice and how do you find it?
Yeah. I think you spoke a little bit about it before, and it changes a little bit as you, depending upon the type of team you lead and the size of the team you lead and the mandate that they have. I apologize if that came in the mic. Right now, one of the best things I do for my team - and this is something I've done for most teams, but it's an emphasis right now - is I try to be like the Clarifier-in-Chief. Like, what is the business really asking us? What do they really want? What's at the heart of why they're asking this? So, you know, they want something by a certain timeline. Why is it that timeline? Really just sort of like, why, why, why, like a pesky toddler, right?
So because I know that with a young team, the bias is towards answering the question that they think they know is correct. They just want, they wanna deliver. There's an eagerness to it, which is awesome. But at the same time, I know that oftentimes it's easy to misinterpret what people want. They say what they want, but they mean something different. So you really have to get in there and make sure people are super clear.
And the other thing I do is the same thing, maybe on the other side, with the business leaders to say, like, I hear that's a priority, but let me tell you the five other things that are going on. Let's talk about it. If you think it's bigger than this, these five, let's bring the stakeholder for those five in. Let's have a conversation together. Let's really go after this, as opposed to sort of just saying, yes, you're on the backlog or no, we've already established stuff. So I'm also the Clarifier-in-Chief in terms of like, why are we doing what we're doing? And I had a guy the other day sort of make fun of me, if I'm honest, because he said I gave a 15 minute prelude, I think was the term that he said when I did a meeting. But okay. I did. Because I wanted clarity, clarity, clarity is one of the things I'm trying to drive right now.
And then the other thing I think, you know, for me, you hit on it and I reflected on it when it comes to the product development, the physical product development, I am sort of relentless in being like, what is useful? What about this thing that we're developing, this analytics product, is so good that it's useful? And one of the things I try to think about sometimes if it helps people is, you know, would the person pay for this? Could you charge the marketing department for this service? We're not set up that way. We will never do that, but just think it through that. And then people will normally will go, eh, if we're gonna charge for it, I think I might make it a little better looking or, yeah, I think I might invest a bit more time into external data, or it's just a little bit of a poke, but again, the key thing is, you know, are you really emphasizing how useful is this thing? Truly, how useful is this thing? And it helps that I've done that.
And it helps that I can admit to people like the story I told you about, the queue manager, that it's easy to get distracted with cool. But cool is not always useful. Those two are not inherently interchangeable. So, chief clarifying officer, I guess, and, you know, key focus on utility.
So in your LinkedIn profile, you talk about your framework of people, process, and technology. Could you talk about that and how you brought that into Little Caesars?
Yeah, no, definitely. I mean, I think in my profile, what you're referring to, is I consider those things - and whether the data and analytics Illuminati, or whoever created that phrase, I don't know who it is, "people, process, technology," maybe you know, if you're a good historian of these things. Whether intentional or not, to me, those things are total rank order. People, number one. Process, number two. Technology, number three.
And the reason they are rank order is, you could think about them in the negative. Well, what happens if you have great technology and even great people, but it's like, there's no clear way on how you deliver value? That's your process, so then you're messed up, right? You get an awesome process and awesome people, but the best they have is like, access. You know, you will run out of road really quickly. So I think that the people are super important, and they're important across a couple of dimensions. First off, and it's maybe now I might even semi-contradict myself, it's people in the universal plural term. It's more than one person, but it's also persons.
And what I mean by that is, increasingly in my mind, in an analytics environment as a leader, you have to think about the diversity of talent that you have and that you need. I am always looking at the people that work for me and work for my managers, and I think they're awesome. I'm really excited. We have some people who are new. We have some people who are established, and they're just doing great things. But I'm always saying in the back of my mind, like, what are you not that great at that either we can develop in you, or maybe what are you not that great at that you don't wanna develop and that we need to recruit for? I'm always thinking about people in the mass. And because I'm thinking about the team and thinking about what are the skills across the board that we need? And how are we making sure that people can contribute with the skills that they have and how are we recruiting for the skills that we don't have?
So I think the people thing is something as a leader, you have to think about the individual. How are you doing, how are you feeling? You know, how are you contributing? But you also have to think about the group. Do we have the talent that we need in a diverse way at diversity and cognitive diversity and background diversity and in skillset diversity? And then the process is simply as a leader, how do I support the team to do the most amount of work with the least amount of effort? That for me is a process, right?
That's the process. How do you make sure you're working on the right business priorities? How do you make sure that the most common roadblocks that are gonna come up at them - that could be access to tooling? You hear these stories of people like, it takes us six weeks to get a tool. Well, as a leader, you just need to fix that. Like, I'm sorry, that's a fixable problem. You need to find the people in your organization who are being bureaucratic, you need to fix that. So process, how do you enable those people to run as quickly as possible through all of the gate checks that happen in any given analytics product development? And then technology, how do you make sure that when they need something that it's the right thing that they can get?
But I do still say those three are in rank order: people, process and technology. And I think it's easy for people. You know, it's easy for me to think about the people thing. You've heard, I'm sort of a people person. So I also have to be cognizant, that - am I paying enough attention to the technical hurdles? So we have an example right now where I think maybe we need to look at some rearchitecting, because we are expanding the footprint of our PI platform. And I don't know how to do it, that's for darn sure. But I also know my bias is to think I can fix that with people or process, but in reality, it's probably a technical problem. So you also have to be aware, if you think about that paradigm, where your skillset is and where your inherent bias will push you to solve problems out of one of those three.
It's probably true, I would guess, if you're a technologist, and you aren't that comfortable, you know, with assembling teams, then you've gotta think about putting yourself in the tough area of assembling a team. As I have to sometimes bury myself in technical literature that sometimes I don't understand every fifth word.
A lot of what you're describing there reminds me of data mesh's socio-technical approach. Have you looked at that at all?
You know, Data Mesh is one of those things where I read it and every few words I don't understand, and I do understand. But I understand conceptually what it's about. And I understand that my, the thing I'm most unclear on is, or the thing I have a concern about is, how widely capable does the rest of your organization need to be around data for Data Mesh to work? Because it seems to be a highly decentralized approach.
It is a highly decentralized approach. And there's an episode where I interviewed Zhamak, and I asked her that question. Because that's an issue I see in it, too. And so a common theme that you'll see in those companies is the people in the data team, they can do this, but as soon as you expand out, you're going, it doesn't drop a little bit. It drops dramatically. So that could be a big problem.
Well, I'll listen to that episode. 'Cause again, I think sometimes if you think about things that are obvious now, like cloud, right? It would've been also easy to be mystified by cloud maybe 10 years ago. But if you start at least thinking about these things and going back to my patients thing and back to my IIA folks, colleagues and clients saying, hey, this is a three to five year journey, I have to start reading about Data Mesh now, 'cause maybe we are in a place where that could be solved for in two or three years. There maybe Data Mesh as a concept becomes further developed and highly supported. At which point I don't want to start my research about it then. I want to have the small, you know, diesel generated part of my brain that goes real slow just to think on this every once in a while.
Drew, how do you recommend teams be organized to get the most value from data?
I am a fan of teams that are sort of set up like pods, if you will. I think that the majority of your team members should be attacking known business problems in a sort of cross-functional analytics team, which includes all the classics. So the business stakeholder, I think that the time of an analytics translator is gone, I think the business people know enough to talk well. And I think the data people know enough to talk business, but then you need some data science or some data analyst capability, depending upon the level that you're doing. And you need some data engineering capability. You may need some UI or UX depending upon how widely you're spreading what you're spreading. So I'm a big fan of those teams sticking together.
And I recognize that is probably also because again, my lesson and my learning and maybe it becomes my bias, which comes again from actual products being developed, when you have a diversity of talent that you put in a psychologically safe space, so you say, hey guys, you make your objective clear. You give them the resources, you have them work through a defined process, and you give them the mandate, and you let them be challenging to each other while still being respectful, you're gonna get good products. It's just gonna be hard for you as a manager to not always step in. So I'm a big fan of sort of self-governed pods once that is clear.
The only thing I know that we need to do, because we're early stages, is we need to reserve some time for that pod to engage with its functions so that we are building out sort of data science model libraries, and we're building out data engineering capability more broadly, and we're building out data assets more broadly. So I think for us, that pod, unfortunately, can't only focus on the given analytics product they're working on, but they're gonna have to reserve some time for feeding that back into the larger ecosystem, which needs to be kind of transformed.
So, generally speaking, the organization, for me, usually means that in our stage, which is early stage, that the analytics resources are centralized and that the business resources are borrowed, so to speak. Maybe it's the other way around, the analytics resources are lent to the business user. And then that centralization, I've written about before though. You have to be careful, 'cause it can go too long. You can sort of love the fact that you always see everything happening in analytics as a leader, but in reality, they might not be answering the most relevant business problems.
And so at some point, you've gotta adapt your model and put more analytics resources on the function. But early stages. I think it's a centralized model with people working on design, on decided analytics products that are really clearly prioritized, really connected to the business, with some reserve time for, for feeding back in the lessons that they've learned that can be shared across teams is really important.
And was that a change that you started making at Little Caesars or have you not made that change yet?
No, we have that in Little Caesars. The only thing we are starting to try to find out well is where should we... we have data for way more than analytics, right? Our number one important need for data is to sell pizzas and make people happy with pizzas. Like, that's the number one. And then we can analyze that data. So what we haven't understood necessarily is where data engineering for the purpose of operational data needs to focus there, where we might need data engineering, additional data engineering resources for analytics. But otherwise we're fairly centralized on the analytics side. And then we work together with business stakeholders in finance, operations, and other areas.
What are you excited about right now in analytics?
Well, that's a good one. I'm actually ... so on, on a technical side, so there's a few things. So, I also have data governance in mind, and I know that data analytics people don't like to talk about data governance. They love them, but they don't like to talk about 'em. But actually, we are evaluating data governance tools, and those tools have embedded AI and ML. Now, sometimes those are fancy words, but the ability to actually do data discovery and data deduping with digital tools is really exciting to me.
'Cause the last time I dug my hands into data governance was at IKEA, and that was highly manual. That was like literally going around an office in Sweden being like, oh Jesse, you own this system. Could you tell me? I was like, oh my God, I hated that. So I'm really excited about the advancements that really, to be honest, started in other parts of analytics that have come into data governance. I'm excited on a technology point of view that - maybe this is a little bit people won't like this, but oh well - with the three cloud providers, maybe it's two and a half, depending upon Google, with the three cloud providers largely established now, there are so many people that know how to operate in those systems. You know, people like Databricks or Dataiku, or whomever they are, which if you can leverage those capabilities, you can do it with relatively little disruption in your cloud architecture.
So I'm excited that people have, we've got some stability in cloud. And of course the cloud providers will tell you no, but we do the best auto ML and we do our own best. Yeah. I'm not so sure about that. So, the ability of third parties to integrate with those guys is super exciting. The other thing is, at Little Caesars, one of the things I'm excited about, and I say this but I always wonder if I have taken my own salary down a peg when I say it, but, you know, it's 2022, so I'm excited that I don't intend to invent much, to be honest. I've got a brilliant data scientist and he's working on a prediction model, which will come down to within minutes by product, which is amazing, but he's using existing models. He's just super smart at using them.
We're using cloud technology, we're using other embedded services in those clouds. And we're gonna use a data governance setup, which is relatively standard. One that Peter Kapur and I wrote about. So I'm excited that I can do a lot more doing and a lot less sort of contemplative thinking, because it's 2022 and a lot of the stuff that we're doing, people have done already. And I don't mean any disrespect to myself or anybody, but if the first thing you're doing when you're approaching a new analytics role or a new analytics problem is to just sit there and think by yourself, I think that's crazy.
Because there's folks like you out there who are experts, you listen to your podcast, you could go listen to other people's podcasts, you could go read Brian O'Neill's thoughts on product analytics. There's so many great people who talk about this stuff that you can get ideas and you can get maybe 80% down the ideation road just through what's out there already. And I think that if you're not doing that, you should question whether or not you're just trying to create something new for the sake of creating something new. And that's fun, but it's not really value adding to my mind. So I'm excited it's so established. As a field we're becoming a bit more established, and it's becoming a bit more able to kind of get in there and do stuff.
Those of you who aren't watching this video, when Drew was talking about the data governance, he actually curled up into a fetal position telling that story.
Oh man, it was brutal, but you knew it was so important. So you had to do it. And my data governance manager here is awesome. And my MD Emily, she's awesome. But they're like, yeah, we're not doing it. I'm not doing that.
What do you never compromise on?
Yeah, it's gonna sound super duper cheesy, but I never compromise on disagreement with respect. What I mean by this is, I don't have a lot of tolerance, I have no tolerance for passive aggressive behavior. I have no tolerance for not telling the truth. At the same time, I have no tolerance for people who are like, you have to do this because you're just a lowly analytics person. Or analytics people saying, oh, that person doesn't know what they're doing because they're stupid. And they don't have the same graduate degree I had. Like, there's so much that can be done through healthy dialogue, which might feel uncomfortable in the moment. We might really disagree. We might really think the other person is not listening or the other person is not paying enough attention. You might really think that, but there's so much that can be accomplished with sort of healthy, challenging discourse, which in our current world - on this side of the ocean more than yours - is tough, but you have to get in there.
And I think one of the things that I always say is like, at the end of the day, as long as you yourself go in with your heart thinking, this is what's right for the company based on what I know, and you start with that framework, someone who maybe has more tenure in the company or has a higher role or maybe just a different perspective can say, I think you're wrong about what's right with the company, what's right for the company. Okay, let's discuss that, because you might be sharing something I don't know. Maybe there's a history. We tried this approach and it failed or what have you. But as long as you have, I want to do what's right for the company, then your idea as itself can't be wrong. And so wrong that it should just be disregarded.
So, you know, discourse and disagreement, always with respect is the thing I don't compromise on. So if someone's holding back, I will have a one-on-one. and I've had it already with people. Like, in that meeting, you knew the answer and you decided not to share. Let's talk about why that happened. Or I've had a conversation already with a senior leader - said, like, this person just joined the company like a month ago. How are they supposed to know this? And when you talk to them in that way, they're never going to ask you again. That's horrible. So like, you have to also as a leader be willing to have those conversations with people one on one when they violate that thing which you say you don't compromise them.
Oh, that's excellent. I think you're the first guest to not only say that and then hold others accountable. I love that. Well, Drew, thank you so much for coming on the show. I really appreciate it. And I love how much experience you have and how openly you talked about and shared your experience in it.
Well, thanks a lot. It's been a pleasure. I've always enjoyed following your work, and you're one of those people I look to when I try to understand the things in the technology and data teams world that I don't understand. So I've always appreciated being able to find your voice out there on the interweb and in other places.
Another great story, another perspective shared on data, and the tools, technologies, methodologies, and people that use it every day. I loved it. It was informative, refreshing, and just the right dose of inspiration. Remember to check dreamteam.soda.io for additional resources and more great episodes. We’ll meet you back here soon at the Soda Podcast.
