In the age of data-driven enterprises, many organizations are drowning in their data. But being data-driven may not be all that it is cracked up to be. What are the symptoms of being data-driven? Is there a better way? In this inaugural episode of The Insights Factory, Host Ian Cook sits down with fellow leaders at Seek, Founder and CEO Erik Mitchell, and Industry Principal for Retail and CPG, Sean Cline, to discuss.
In the age of data-driven enterprises, many organizations are drowning in their data. But being data-driven may not be all that it is cracked up to be. What are the symptoms of being data-driven? Is there a better way?
In this inaugural episode of The Insights Factory, Host Ian Cook sits down with his fellow leaders at Seek, Founder and CEO Erik Mitchell, and Industry Principal for Retail and CPG, Sean Cline. They discuss the importance of insights, what they are and how they drive businesses forward. They share case studies across retail verticals to demonstrate, not only what differentiates an insight from mere data, but also the tangible ways retail leaders can use insights to impact the bottom line.
“You have the ability to not only see the information, but then tell the story around it. I think systems and tools and insight functions need to be able to do that, be able to tell that story and an insight does that.” – Sean Cline
“The difference between, let's say a dashboard and or a report and an insight, is the report's going to tell you that you lost people. An insight is gonna tell you why, and it's gonna tell you, now what do I do?” – Erik Mitchell
“I always like to think about somebody who's a master craftsman at something. Whether that's creating things out of wood or distilling whiskey and bourbon, things like that. These companies are all masters at their individual craft, but, their craft is not analytics.” – Erik Mitchell
“There are constraints and rules around how that data's presented as well. It's a challenge and to automate that function, or systemize that function, if you will, is key to having success in this speed to insight type conversation.” – Sean Cline
---------
(00:33) Episode backstory
(02:23) Erik and Sean’s comments on the backstory
(6:51) What is a challenger brand?
(7:11) What is an insight?
(10:48) Erik and Sean’s experience finding insights
(17:30) The data supply chain and the last mile of analytics
(21:26) Hurricane data example
(24:24) Product weight and shopper demographic example
(26:17) Can you scale insights across companies?
(29:25) Advice for business leaders starting to look for insights
--------
This podcast is proudly sponsored by Seek, the leader in cloud-based creation and delivery of industry-focused insights.
--------
Connect with Host, Ian Cook, on LinkedIn
Connect with Erik Mitchell on LinkedIn
Connect with Sean Cline on LinkedIn
Follow Seek on LinkedIn
Ian Cook: [00:00:00] Some days it feels like we're drowning in data, but data isn't knowledge. Data without context or the expertise to understand it is just something eating up storage space in your warehouse. That's where being insights driven comes in. Insights driven organizations focus on generating actionable insights rather than just collecting and munging data.
Insights drive better decisions. Welcome to the Insights Factory.
On today's episode, Tell me if this sounds familiar. You're out at a new restaurant. The one everyone has been raving about. You greet your friends, order a drink, place your orders from a list of truly stunning dishes. A few minutes go by and the waiters arrive. With all the fanfare of a magician's flourish at the end of a new trick, they show you the plates.
Covered in only raw ingredients. Uncut vegetables, sides of meat, fish straight from the water, oils, herbs with dirt still clinging to their [00:01:00] roots. Waiters look at you as if they expect you to start clapping. No? Not familiar? I didn't think so. Now, ignore the restaurant and think about one of the most vital assets of a modern company.
This is exactly what happens when we buy new data. We're promised truly breathtaking results, but we're given only the most raw inputs, and expected to assemble the end result without even so much as IKEA drawings of sad blobs getting their fingers pitched. Worse yet, we're charged even for the pieces we didn't use, and each piece of the same ingredient might have a different cost.
But there's a better way. And we're here to talk about it.
I'm your host, Ian Cook, Chief Technology Officer at SEEK. I've spent my career working with, in, on, and around data across industries, ranging from energy markets to healthcare, and most recently, consumer packaged goods. I am joined today by my fellow Seek teammates, our leader and CEO, Eric Mitchell, and [00:02:00] Sean Klein, Industry Principal for Retail and CPG at Seek.
Eric founded Seek Data after a career working for some of the biggest names in CPG and retail, and Sean had an only slightly longer career in CPG, having worked in nearly all facets of the CPG world. Alright, let's get into it, gentlemen. Eric, you wrote the piece that I took the introduction from. Is this something you see routinely?
Is this a pattern that you've seen happen often in your career?
Erik Mitchell: Yeah, I think, and this is something that I personally experienced and we see across a ton of our clients today and frankly, pretty much every company in any industry working with data deals with this today, as you mentioned, with the analogy of the restaurant, you know, one of the things that business teams struggle with in particular is just the ability to get value out of the data that they're buying.
And over the last few years, as organizations were striving to become more data driven. The only way that they knew how was to just buy more data. And so if you think about this from a restaurant perspective, you know, it'd be like opening up a [00:03:00] restaurant and doing nothing but buying ingredients. You just, you know, that people want food.
And so you just continue to buy more and more and more ingredients to become a food driven restaurant. Right. And so what ends up happening is the customers at a restaurant end up being frustrated with the output, which is just ingredients, as you mentioned earlier, and business teams end up frustrated in the same way.
They're just getting the data in a new format from their I. T. Teams are from their data teams, etcetera. They're not actually getting the results that they set out to achieve.
Ian Cook: Why would they be buying the data if they weren't sure what they were going to get out of it?
Erik Mitchell: So that's a good question. And I think, you know, I'd be curious to hear Sean's perspective on this too.
I think business teams for the most part know what they're trying to accomplish, which is the most interesting part of this particular equation, right? So, you know, it's like we go to buy data because we have a problem we're trying to solve. But when we buy that data, we don't actually solve that problem.
And so there's a gap that exists there that every organization faces. It's not a problem of understanding what I'm trying to accomplish with that data. [00:04:00] It's just a problem of actually getting the data to solve it. So, you know, Sean, I'd love to hear your perspective on that as well.
Sean Cline: Yeah, that's a great, great insight.
Also, if you think about even free data, free data, lots of it means that you might find something that, that allows you to sell more stuff or to provide a different insight back in the day, there were a lot of what I would call button pushers in this function. Analysts who were really good at harvesting data from some of the different systems, particularly in retail CPG, some of those syndicated, some of those directly retailer POS data.
They're really good at button pushing and really good at pulling reports. You can't see the air quotes and what they lacked in was being able to take that data or those mountains of data and make that into. Well, what we're calling an insight, but ultimately a story and then be able to convey that story in a way that provided guidance to, in this case, a retailer partner on what to do, bring more inventory and bring less [00:05:00] inventory and move this item to these stores, things like that, there were a lot of button pushers and I sort of have a story that that'll sort of relate into this.
So in an early, early position that I held, I was a marketing coordinator. So I did a lot of. Scouring of syndicated data sets, and this literally was by paper every month. We'd get our big giant report. It was, you know, four inches thick, big stack of paper. We had an administrative assistant who harvested our POS data from a very large retailer.
And at that time, there was a dedicated computer within the office. And she was really good at harvesting that data. Again, button pusher, no offense to her whatsoever. It was the role at the time. And she was going to take maternity leave. So they said, Sean, you're up. You get to man this computer and pull reports for us.
And on a whim, I literally put a weekly POS number into a [00:06:00] line graph. And it was amazing, the reaction that internally I got from our production planning team, they were like, Oh my goodness, this is great. Now we can sort of visualize what is happening from a sales standpoint. That led to getting pulled into a conversation around demand planning.
We had a feature coming up in this retailer and they wanted to know how much they were going to order. And just with a couple quick calculations, that data. Became information, which became an insight and it became a story that we leveraged first internally. And then we were able to take that to our partner at the retailer and say, this is what we think, and this is why we think that.
And they were very intrigued to see how we got there. And they were very excited. And the interesting thing, the company I was with at the time, challenger brand. By far and away, got a lot of extra time with the merchant and a lot of extra space on the shelf because we were able to provide some of these insights in a way that was mutually beneficial.
Ian Cook: [00:07:00] Sean, real quickly, what is a challenger brand?
Sean Cline: That's a great question. I would define a challenger brand is if you're standing in front of a category and you see the national brand, the brands that you know that are household names, and I don't want to throw any out there, but you, you would know, right?
Those that you don't know, or maybe you know of.
Ian Cook: So one of the things we've talked about a bunch and we've named the. Podcast after this is we use the phrase insight pretty repeatedly. Sean, Eric, my question to both of you is, can you describe a little bit better what you think of as an insight over and above just say dashboard or having the data?
Sean Cline: So to me, it's sort of like. Leans into that mutually beneficial piece of information that changes the trajectory in whatever it is we're talking about, whether that be replenishment, sales, distribution, pricing, those are sort of the basic foundational functions. And so in order to package that up, and there are constraints and rules, but you got to be able to package that up and [00:08:00] convey what it is you're seeing and why that's beneficial.
One of the things when I was. Young and, you know, long hair and earrings and an analyst. One of the things that somebody told me once was you have the ability to not only see the information, but then tell the story around it. And I think systems and tools and insight type functions need to be able to do that, be able to tell that story.
And, and insight does that insight is more than just a little tidbit of data. It's not a number.
Erik Mitchell: Yeah, I think overall an insight is something that frankly drives action. And so if you think about it, I can create a report that tells you that sales are up and down or in other industries. Right? I can tell you that insurance claims are up and down, or I could tell you that, you know, in health care, the costs are up or down, things like that.
Right? But those don't necessarily, although they are information and they are helpful, they're not changing the behavior. In and of themselves, [00:09:00] and so I think as we think about insights, I think one of the challenges that many organizations have this goes back to again. This whole restaurant concept is the insight is the meal.
The insight is the curated meal. And so the challenges people can't get to that because they're either not using the right ingredients. Or they don't really know how to cook. When you think about creating an insight, if I'm trying to drive action, then it's, how do I use this information to grow my business?
The difference between let's say a dashboard and or a report and an insight is the reports going to tell you that you lost people in insight is going to tell you. Why? And it's going to tell you now, what do I do? So if we think about examples, we were actually chatting about this earlier today on a call with a client is, you know, they're thinking about how do we use things like weather data.
And as we're looking at weather data, how do we think about using a 2 week out forecast to say at the [00:10:00] end of the winter, I'm going to have stores that are stocked with cold weather gear. And my first instinct would be just to mark them all down because I'm now in spring, right? Well, when I mark that down, I lose, let's say, 30, 40, 50 points of margin.
So I'm losing 30% on this stuff. But what if there are stores that are still going to be cold for the next two weeks? I don't have to mark that stuff down 50%, right? But I wouldn't know that. Unless I go through the process of finding the insight to say I have this question, which is I want to use weather data and I want to use something else to determine what stores are still going to be cold two weeks from now, a report doesn't tell me that an insight.
If I get to that piece of information, I can change my behavior and say, keep those stores at regular price for the next two weeks because it's still relevant.
Ian Cook: Eric, Sean has talked a bit about learning that process of going from Button pushing to having true insight about what the data is telling you.
Did you go through much the same thing in your career of trying to figure out [00:11:00] what the data is good for? And how did you get a sense of what that could be?
Erik Mitchell: Yeah. So it's, it's funny. I tell everybody my analytics career actually started as an accident. I went to college just like I was told I was supposed to got a finance degree and ended up working for a fortune 500 organization.
In a role that focused on pricing and promotion first job out of college, I had a friend of mine that worked in a different department and he created a presentation one time and it had a really pretty chart on it. And I just said to him, hey, what tool did you use to create that chart? So he told me the tool.
The tool was called Tableau. This was probably in 2014 ish. So tablet was relatively new. Yeah. So I called the help desk and said, Hey, I would love to get a trial of Tableau. And about an hour later, It's on my computer, and I'm thinking, this is great. It's a two week trial. And it's free. I'm going to spend as much time as I can learn how to do this about a week later.
My director comes over to me. And at the time again, I'm [00:12:00] a new analyst. I was scared to death of my director. So my director comes over to me and says, Hey, I got a charge on my cost center for something called Tableau. And he said, and it was about 2, 000 and in my head, I thought I was gonna get fired from that moment on.
I was like, I'm done here. I'm getting fired. The last words he said to me of that conversation were, you better use it. And then he walked away. So from there, I was like, oh, crap, I'm going to get fired. I need to prove that I can generate value from this thing. So. Started playing with Tableau and really understood.
Okay, now I can work with bigger data sets. And when I'm working with bigger data sets, I can solve bigger problems and I can think about problems in a new and different way. And so I started to do at the time what I thought was just more complex analysis. What I was doing to Sean's point earlier is creating insights.
From that data, right? I'm no longer just creating reports. I'm creating insights. And so all of a sudden you start to talk about in particular in the [00:13:00] pricing and promotion world, they study things like, do I run a two for four on a product or do I run a promotion at one 99? Theoretically same price, completely different consumer behavior.
When you run this and it's simply because the illusion of multiples, right? So it's like all of a sudden I'm learning insights about that.
Ian Cook: And you have access to that consumer data.
Erik Mitchell: Exactly. Yeah. So through partners that we work with, again, a lot of that data is coming through these partners and the challenge that a lot of companies have today is they're getting that data, right?
It's part of their ecosystem. They're just not using the tools or using the capabilities to get to insights. They're just building reports. And so that's the whole foundation of the insights concept.
Ian Cook: Excellent. Thank you. And I wanted to toss this over to Sean here really quickly is I grew up in a time when people talked about big data and everything's worried about big data.
That feels pretty much over. People just have data. They have a lot of it. The system that you worked with Sean provided a lot of data [00:14:00] that ostensibly was valuable. But the question was, how do you make those kinds of insights as Eric mentioned out of it? Did you routinely have those kinds of challenges with the people consuming the data you provided?
Sean Cline: We did, and it started really as kind of a tech solution, a technology solution to really a business issue when I came in, we started to look at it as a business solution with a side of tech, if you will. So we really started to build. For lack of better terms, kind of our reporting around, how do you solve for specific use cases, things like replenishment, things like on shelf availability, you know, even kind of guiding reps to what stores and what tasks to do when they're in those stores.
And then we, of course. There, there were a lot of scorecard type things, whether you were up or down percent of total kind of things, but, but really kind of moving towards this business solution, as opposed to a technology solution, [00:15:00] probably the biggest thing though, where that type of organization comes into play is the data and the data acquisition.
If you think about it, there was value in providing clean and I'll call it hygienic data to a company that makes candy. They're not really kind of tech people, right? Although they're investing in those types of folks, BI people, data scientists, things like that. They're really good at making chocolate, for example, but they're not great yet at creating kind of these business solutions, using these data networks and data
Ian Cook: providers.
That definitely speaks to my experience. I, one of the things I spend a great deal of time doing is data cleanliness, data standardization, data normalization. When we think about buying data is, Oh, this is going to be clean. It's going to match exactly what I want. And people don't quite realize just how much effort has to go into that.
So you spend your time rather than utilizing the data, trying to make the data into something that you can just put to use as quickly as possible. So this is one [00:16:00] of those promises that we're given when we try to order the data off that
Erik Mitchell: menu. Yeah, I think one thing to add to that, which is super interesting, and Sean can speak to this from his experience as well.
There are companies out there that are also fantastic at selling that menu. Any technology company, any data company has generally thought about these types of solutions, right? They've said, hey, you can use this for things like replenishment. And then the business users are like, oh, this is great. The person told me that I could use this for replenishment.
And then they get the data. And then that provider is like, well, we don't actually do that. But you could if somebody knew how to do that. To Sean's point, making chocolate is a great example. I always like to think about somebody who's a master craftsman at something, right? Whether that's like creating things out of wood or distilling whiskey and bourbon, things like that.
These companies are all masters at their individual craft, but their craft is not analytics. And so what's interesting is you end up with, again, a lot of people that are hungry for these things. So they go [00:17:00] into the restaurant, they're starving. The waiter brings them a menu of everything that they want to see, and then they pick what they want.
And again, to the conversation earlier, the plates come out and it's just all the ingredients. The question is, how do we think about that? How do we solve that particular problem? One of the ways that we like to look at that is as a supply chain and as what we would call a data Supply chain.
Ian Cook: Okay. That's interesting.
So you think of the data and is existing on this chain that we often think about in terms of products. There's a supply chain for a product. Somebody's got to build it, ship it, get it to a store, get it to your house. You're saying the same thing exists for data.
Erik Mitchell: Yeah, let's talk about a product supply chain here for a minute if I were to create a new candy, right?
So I create a new version of a chocolate bar. I might be a master at producing that and I can produce it at scale. So a couple things have happened already in that supply chain. I've come up with that product idea. [00:18:00] I've manufactured that particular product. Next step is I have to put it in my own warehouse, right?
I have to store that product. That's relatively easy. I can find warehouses. I can store that product. The next challenge is, okay, how do I get that product from my warehouse to Sean's warehouse as a customer of mine? And then how does Sean get that product from his warehouse either into a store or Or in today's world to Ian's house, what you see across any product supply chain is that the transition from Sean's warehouse to Ian's house, that last mile is the hardest and most expensive part for any organization today, right?
Everything before that is relatively streamlined. When we think about a data supply chain, it works the same way. And again, Sean can speak to this from his experience working for one of these providers. It's relatively streamlined at this point to curate data. Yeah. Stored in my own warehouse, get it into a customer's [00:19:00] warehouse.
The challenge becomes that last mile of analytics, which a lot of people will categorize as just making dashboards. That's what we call a symptom of being data driven is you just have a lot of data and you have a lot of dashboards, but you're not actually doing anything with them. Sean, do you have examples of those types of things from your experience?
Sean Cline: Well, I have a couple of thoughts around that. So if you think about getting to the point of an insight, then what, what do you do with it? I've uncovered some little tidbit of information. That's going to help my retailer partner do something, optimize inventory, sell more stuff, find a new store to put it in, whatever it is, all things that number one need to be mutually beneficial.
Otherwise you have no story to tell and you have no message to share. It has to be mutually beneficial. So check that box. Now, what, how do you get it to them? There are very specific, I'll call them rules or constraints around how you can present this type of information. For example, there are minimum order quantities, and this is kind of a.
granular example, [00:20:00] but an example, nonetheless, you have to be in case pack. You have to be in pallet quantities. You have to be in full truckloads. You have to ship from a certain ship point that carries that product into specific distribution centers on the retailer side that have enough. Distribution more than one store or a number of stores to be able to meet a lot of those minimum order quantities.
So if you're presenting information to that retailer, for example, I want to expand my distribution in item a into these stores. If there's no distribution where you're suggesting it goes there. You can't necessarily present that information and get it done lead to a much longer conversation. So to Eric's point, there are constraints and rules around how that data is presented as well.
It's a challenge and to automate that function or systemize that function, if you will, is key to having success in the speed to insight type conversation.
Ian Cook: And I think this is really interesting because we're kind of hitting on a similar thing in both [00:21:00] cases. There is a way to deal with the product that you have and delivering it to a specific person that requires you understand quite a bit, not about just the problem they have, but the information you have to provide it.
What I'd love to hear is an example of when you've created an insight in your career and said, this is something that I have picked up and that you saw. And to Eric's point, action taken based on the insight that you created.
Erik Mitchell: Yeah. I mean, I hate to throw another weather one at you, but I, but it's fascinating.
So, so think about hurricanes. What's interesting is you think about preparatory items for hurricanes. The challenge that one of our retail partners had at one point was, Hey, we know that we're not serving our customers better in an emergency situation for any retail or for any company. Frankly, the most important thing is you're serving your consumers.
At their largest times of need, hurricane situations can obviously be devastating to people helping, helping people prep for those in many different areas is impactful, how to categories like peanut butter [00:22:00] or coffee or pet food or any of those react or behave, I should say, in advance of and On the tail end of a hurricane, there are cool ways to create insights about what happens beforehand and things like that.
They don't become actionable until you start to talk about the actual items themselves. So one of the questions from the retailer was, I know that sales spike. I just don't know what products and I don't know where. And my assumption is the different products spike different ways. in different stores, but they didn't have any way to solve that.
So what we did was we came up with a process that allowed us to do that. Again, Sean's point at the time, a little bit manual, but you know, the unique insight here was completely different brands spiking, let's say in Tampa versus Miami. You think about the different cultures in those areas and maybe how that influences different types of brands or preferences or sizes.
That was a super fascinating one. Another one that we had was helping a retailer [00:23:00] understand. Their pet category, and we used a store clustering exercise to say, you know, Hey, where should you add space or take space away from things like giant bags of dog food or wet cat food? We ran our store clustering algorithm and immediately this group of 50 stores shows up.
And that group of 50 stores, all of a sudden you look at a map, you're like, wow, those are all really urban stores. And so you start to look at the behavior of those stores, and they're really heavy wet cat food stores. So tiny cans of wet cat food that sell really quickly, they sell a ton of the units per week.
And it's interesting because we went to the retailer and said, hey, we think these, there's something different about these stores. Right. This insight about the fact that they behave this much different tells us that consumer is truly different. And so you should make these changes, et cetera, et cetera.
What's interesting is if you think about it, it's common sense, but you don't recognize the common sense until data helps you do that. Right. Which is these are highly urban stores. People have cats, they don't have giant [00:24:00] dogs. And some of the feedback that they got was, I can't carry a 50 pound bag of dog food on the subway.
So I'm not going to own a dog, I'm going to own a cat. And by the way, I get wet cat food because I also don't want to carry a 50 pound bag of cat food. So, you know, it's things like that that are insights that then that retailer can say, Wow, I'm going to change my set. I'm going to change the shelf to reflect that.
Which, in their case, drove the business up 15% year over year. That's a
Sean Cline: really interesting insight and I've got sort of something to piggyback on the back of that. So talking about weight and shopper demographic or shopper makeup. So in the category that I spent many years in, if you think about heavy product, the shopper of this particular, it's a CAT product, is a, a traditionally a female, you know, call it 34 to to to 52, I think was the age group, something like that.
32 to 54 I think was the group. And typically speaking, Based on the information that we were being fed by way of our agency partners and things like that, [00:25:00] that shopper is, you know, Five foot four, I think was the average and, you know, kind of a smaller build person and to be, you can't see me on the video, but I am of that same stature.
So I was a good example of eye level within that category. And at that time also environmental sustainability was a really big thing. So how many trucks can you take off the road? How much less diesel fuel are you using? How much oil or plastic can you reduce by? So. We came up with a product that actually lightweighted of otherwise very, very heavy product.
And like Eric was talking before, this weight thing comes into play, especially when you know who your shopper is and how they operate. So we took basically 50% of the weight out of this product, thus allowing our traditional shopper an easier experience to carry a much lighter product. But we also had an environmental sustainability play as well.
We took a number of trucks off the road. And, you know, without much fanfare, we really kind of changed the category, those [00:26:00] items, those products exist by all of the large suppliers within that category today. Uh, and we were just a little challenger brand who came up with this idea. So we
Ian Cook: have great examples here of insights that you two have come up with and people will find through the process.
One of the things I'd find about the claim about dashboards is to get information to a lot of people very quickly, i. e. scale is the insight, something that you can scale, or does it have to come from the individual like Eric being, you know, a little detective and saying, you know what? I think there's something going on here.
And now I can say, great, I went through this process and we can take action. Or can we get these things to be more frequent be across companies. How do you view that as a potential issue for becoming what you've called Eric, insight driven.
Erik Mitchell: I think overall, it's definitely something you can scale. Now, again, it are, is every problem the same for every customer?
No, but I think the goal here again is speed. [00:27:00] And so as you think about driving speed to insights, right, I think what we can do is we can say, okay, the tools that we have available to us are things like large data sets, much more advanced cloud data warehouses, application based. Programs and tools so we can leverage and then front end tools that can scale.
And so as we think about being able to scale this insight world, I think it's more or less teaching the technology about the process that then will generate the insight. So if you think about it, it's using things like machine learning and AI to analyze. Store item level behavior and give you exception reports or call out specific things.
It's not going to catch everything, but it speeds you up. And so I think generally teaching the technology to use these types of patterns at scale can generate insights at scale where you know that every organization cares about store and item level insights. And [00:28:00] let's say anomalies, that's the store item level.
Every organization cares about anomalies at store item level. And we can teach the technology to find those. And that at least gets organizations way closer in a way, faster manner than they've ever been able to before.
Sean Cline: And to that point, with the availability of a more advanced cloud structure, it allows you then to standardize what are seemingly disparate data sets like different retailers.
For example, you can standardize the way that data is. Collected stored and then reported out on, and I use that term loosely, but basically presented in these insights where you can say there are trends happening at multiple retailers. And we're able to now see this at scale, you know, with flavor sizes, package type, pricing strategies, things like that, that technology was not available even five years ago at scale.
That was approachable and reachable by the masses.
Ian Cook: So you two [00:29:00] both know me, you know, by new to the consumer packaged goods world. So I'm learning a lot of these things. Talk to me about where I would want to go. If I were starting out in this and wanted to be more focused on insights. I've got a bunch of data.
I'm just swimming around in it. I'm not entirely sure how to get moving forward because I know there are things there that I want to work on. If this idea of the final meal sounds like a great option for me and I want to get there, what kind of things can I do right now to start down that road? Sean, why don't you go first?
Sean Cline: I would think of the basics first. Start with the basics. Data acquisition, data hygiene. We talked a second ago about cloud structures and cloud storage and things like that. Obvious choice. So more from a business standpoint, same kind of approach, focus in on the basics, bring good products to market that have true market fit, understand where and when that would be appropriate, but also to manage that business, you know, focusing on the foundational things like on time and in full [00:30:00] and on shelf availability and things like that.
All of those functions require a decent amount of not just. Data, but then insights to manage appropriately and manage well,
Erik Mitchell: I think to add to that, I would start with the basics and start with them in an evenly distributed fashion. And what I mean by that is a lot of organizations, again, in the data driven world.
Invested all of their resources in data and then starved the organization for insights. If they had a hundred bucks, they spent 90 of it on data and then had to build an entire team, a capability by a tools, et cetera, with 10 bucks. And so what I would do if I was starting out is leverage an approach to say, okay, I'm solving one problem, not 10.
So I want to solve one problem and I want to solve it as quickly as possible. With the most relevant resources for that problem. And so I'm going to go buy the one data set and maybe I only need that data set. One time, right? Instead of buying a five year subscription to an annual database, [00:31:00] I can buy that data one time.
And then what I can do is maybe I spent 20 on that data feed. Now I have 80 left. So what do I do now? I need to buy a new data visualization tool, right? I need to buy some sort of cloud computing algorithm.
Or a, what we call insight that can solve this problem for me with the goal, simply being solving that one problem as fast as I can in the most efficient way possible and efficient does not mean cheap, by the way. So, and the efficient just means using the budget that I have in the smartest way to solve that one problem.
And so, you know, and if it were me, I would say, let's learn from what organizations. I've done over the last five years, which is overspent on data and underspent on the ability to analyze and turn that data into insights. And I would say, let's distribute that 100 that you might have as your budget evenly across things like data [00:32:00] tools.
People, et cetera.
Ian Cook: It's interesting. I really like what you're both saying. One of the things I might add to that, that definitely was something I had to learn in my career is I often showed up as the data science machine learning AI kind of person and would spend a great deal of time just sifting through data, cleaning data, running algorithms on data, building products with data.
The. Time I would have spent or think I could have spent better is talking with someone like a Sean or with an Eric to say, I need to understand the problem better. I need to know the insight that, you know, their customer is doing something specific. The fact that this is an urban store rather than a rural store, and we want to split that way, even to just have In the report, urban versus rural split is not something that I can just think up.
Well, I know 400 ways to pivot this data, but I'm not entirely sure what's going to matter the most to the business. So having someone like this to be able to say, I [00:33:00] know what kind of problem you're facing here. The types of ways to look at it very easily is also something I think would be probably great to add in.
And one of the things that has been fascinating working with all of you is again, I can bring all of my. Background in technology, but when we get to sit across from a large company and they say, well, this is the problem I've got, I was like, that's great. I just don't have the experience to know exactly what the nuances there are.
So I think it's really interesting that you focus on how to get the right data and figure out a problem and. Of course, you know, me, the idea of focus is something I hammer on constantly, but then I come in the other direction and see you all and see like, no, this is the expertise I need to turn my data into something useful.
Cause I know I can do what I want with the data. I just don't know why I'm
Sean Cline: doing it. Yeah. And, and to that point, data scientists didn't necessarily get introduced into this world, this industry vertical, if you will, until about four, five years ago, you know, and even then it was only the super large top 10, top 20, top 30 brands who had budget to be able to do [00:34:00] that, but it's definitely becoming more and more prolific.
Ian Cook: All right. I want to thank you both for being here and having this conversation today. We're actually going to hear a lot more from both of you as the podcast grows. Before we head out though, can you tell me one thing that you'd think the audience should take away from today's discussion? Eric, go ahead.
Erik Mitchell: I would say my one key takeaway would be focus on. Generating action from your data instead of just collecting data. I think as we go on through the podcast, we'll talk more and more about what that means and how you do that and how we can help you learn from our mistakes and from our experience getting there in a
Sean Cline: faster way.
Yep. And my key takeaway would be to take that data, create insights from that data. Data to information to insights to story and story is the big one being able to do something with it Like Eric said earlier being able to take action on change trajectories make something happen with that story It's massively important.
Ian Cook: Great. Well, thank you very much[00:35:00]
This podcast is proudly sponsored by seek the leader in cloud based creation and delivery of industry focused insights Thanks for listening. If you like this episode, feel free to rate and subscribe wherever you get your podcasts.