1
0,00:00,000 --> 0,00:24,000
Cut the tie to anything holding you back from success. Welcome to the Cut the Tie podcast. Hi. I'm your host, Thomas Helfrich. And in each episode, we bring you real entrepreneurs that really overcame challenges on their journey to become successful. We look at the impact, the moment, how it affected everything in their lives. Follow us on Apple, Spotify, and YouTube. Now let's meet our guest on Cut the Tie podcast.
2
0,00:24,000 --> 0,02:52,000
Hey. Welcome to Never Been Promoted. Hi. I'm Thomas Helfrich, your delicious host. Why delicious? Because here's a good thing. You if someone asks you how you're doing, reply with delicious. It will define if someone's actually listening to your reply or not because there's no way they don't go, what? Excuse me? Really effective in a grocery store because it kinda makes sense you're shopping for food. Try it out. It's a tip to make engagement, but make sure you smell well because you can't smell bad and be delicious. I don't think so. There's debate around that. We're not gonna have that today. Today, we're gonna help some, entrepreneurs cut the tie to things holding them back, and and and that metaphor really holds true because there's things there's things in your life. There's, entitlements you feel like you should have, like, a better car. I should, you know, have this or should've gotten that or should've, could've, would've. You know, lead the what ifs for the future, not for the past. And, you know, if there's people in your life you need to kinda figure out, do that. If you have a a knowledge you don't have, go get it. And it it these just become excuses and things that you just get, you know, create a fear around, and then you just get in the cycle. So if you can learn from our guest today, not only kind of how, you know, he built his, you know, Kurt Johnston Johnston is the, is is our person today. We we, to be fair, there's a t in his last name, and and while he's talking today, I'm gonna go through and edit, the title of this because I think it's wrong. But, anyway, Kurt Johnson, a Pallet analytics. We're gonna talk about his journey, and what he does for the world today. So if you can learn how to use analytics a little better in your beta business, learn a little bit from how he built it, we're good. Now I have one simple, call to action. Just follow the podcast. Apple, Spotify, if you're listening, just take a moment. Look at your phone. If you're driving, just do it quickly so you don't get a wreck, and hit that follow button. If you if you're on the YouTube side of it, subscribe, of course. But follow, follow, follow because that way you can get the latest stuff. Alright. Enough shameless plugs. Kurt, how are you? I just noticed that that the title has the misspelling your name, which is good because it's a good marketing trick for me to completely repeat your name over and over and over this Johnstown. Totally agree. Totally agree. Yeah. I mean, I plus, also, some people may prefer the British spelling, so who knows? We'll go with that. I don't I don't it it's more of, you know, it's funny when I we were talking off camera. It's it's very obvious to see where technology is automated and because where you've entered it and where, the human Yeah. The human human jumps in. So No. Anyway,
3
0,02:52,000 --> 0,03:08,000
sorry about that. I like that. There was a time whenever my wife was on the phone with somebody in in telecom, and, they were sitting there going, are you sure there's a t in your last name? Oh my word. And it's like check. Yeah. Yep.
4
0,03:08,000 --> 0,03:26,000
That's fine. You know, you you have a I mean, it it throughout your life, I'm sure you're like, oh, I've had to explain this over and over and over. And, you know, it it I have a last name of Helfrich, and I think that's the way you pronounce it. Unless you mean a German, it sounds way sexier. It's like, Helfrich is it? Helfrich is something.
5
0,03:26,000 --> 0,03:27,000
Yeah.
6
0,03:27,000 --> 0,04:17,000
I'm not a linguist, so I I don't probably and people are like, how do you say it? I'm like, I actually don't know. But you choose with it. But but it's funny. The thing that trips people up is, like, do you go by Thomas? And I'm like, that's why I wrote it that's why I wrote it down. Yeah. Right. Right. Yeah. Generally, you'd write it as Tom if you wanted to be called that. Okay. I will I know this has nothing we're talking about today, but as a data point, if you sign your email, let's say David, that may have been the guy's name, and I reply back, hello, David. Sorry for this or apologies for that, whatever the thing was, to be more formal. And you reply back again as the person goes, don't call me by David. That's only my mom calls me that. And you got, like, 80 probably at this point. Right? And I'm thinking you don't sign your name that way because I don't know because it seems more informal and you were complaining.
7
0,04:17,000 --> 0,04:19,000
Yep. Exactly.
8
0,04:19,000 --> 0,04:32,000
You're like, I was just following your lead, man. Yeah. Right. Now if I reply back Kurt with a c and no t, you can you can call that out. Like, that is a % wrong. Okay. No. No. No. I I'll definitely keep an eye on that. Yeah.
9
0,04:32,000 --> 0,04:40,000
Okay. Oh, look at that. The title changed. Look at that. Amazing thing. Oh, I like it. Yeah. That's really good. That's really good.
10
0,04:40,000 --> 0,04:45,000
Where where do you where do you reside at home? Your exact address, please.
11
0,04:45,000 --> 0,04:53,000
Oh, no. Exact address. Somewhere in Georgia. No. Actually, I live in Alpharetta, not far from where you do, apparently.
12
0,04:53,000 --> 0,05:02,000
Yeah. So we're, we're we're we're neighbors. I'm off the Exit 11. So as far as close to foresight as possible without getting the benefit of their taxing. That's, like Right.
13
0,05:02,000 --> 0,05:08,000
I'm yeah. That's pretty much true for me as well. Yeah. You just go as far far as possible.
14
0,05:08,000 --> 0,05:30,000
Right. Just one step away. If we had just gotten one more step, we're getting better at promised land. But Right. I just decided you know, we we decided to pick as, you know, Alfred, if you guys don't know, it's a suburb of Atlanta. It may be a suburb of a suburb. I kid it's like a double bubble, if not a triple bubble. It's just my neighborhood. I don't we don't have a gated community, but if we did, it would be a triple bubble for sure. Yeah.
15
0,05:30,000 --> 0,05:31,000
I agree.
16
0,05:31,000 --> 0,05:44,000
Okay. If by the way, for those out there, visit Downtown Alberta. Really cool. Very beautiful. This is really Very nice. Great restaurants. I can't afford them, but, like, if you could afford to go to restaurants.
17
0,05:44,000 --> 0,05:56,000
Yeah. No. There's there's great stuff there. And and especially around the holidays when they have their tree and stuff like that, and they have that play area for the kids and stuff like that in the lawn. It's really, really nice. It is nice. It's very nice.
18
0,05:56,000 --> 0,06:11,000
You're, you're you're in the analytics game. Do you wanna kinda start with, you know, like, I would say, start about three fourths into the movie of what you do today, what you do it for, and then back up and tell me tell us your story a little bit of how you got there.
19
0,06:11,000 --> 0,08:09,000
Sure. So, I am, the founder and CEO of Pallet Analytics. We typically just go by Pallet, and it is a retail assortment optimization system. And basically, to translate what that means, it essentially helps retailers decide what products they should have, how many products they should have in their category, as well as how they should price it effectively, to be as successful as possible. And Pallet does this from their pre season period when they're start first planning kind of what they're gonna be doing all the way through their during the active life of a product and category all the way out to the clearance period. So that's kind of what it ends up doing, and it does that through a combination of some AI algorithms, that are custom designed specifically for this use case. And then in terms of where that started, my first job coming out of the MBA was with Home Depot, and I happened to join at a very fortuitous time when they were doing lots of optimization work around different aspects of their business. And I got to be on the front row seat of a lot of that stuff. I got to be engaged in a decent amount of it. And so that sort of set the entire direction of my career since. And then from then, I moved on to working at smaller retailers, which kind of changed the perspective of the problem statements of retail, you know, from a massive retailer, it kinda knows what it's doing, to smaller ones, which are, yes, they still know what they're doing, but it's a little bit more, hodgepodgey. And then you have to kinda, like, figure out, okay. Well, how do I fit into this kind of different space, and how does that change the dimension of my work? And so I learned a lot through that process. And then ultimately, I kept going from one retailer to the next, figuring out that there was the same problems over and over again that I kept having to solve. And a lot of my friends were, like, going, you should create your company. You're you found consistent problem statements that appear again and again and again. Right. You just have your own business. It took me a couple of years to finally pull the trigger, but here we are.
20
0,08:09,000 --> 0,08:26,000
When you, made the move from the W 2 land Yeah. To to, oh god. Did you already have a client or clients set up, or did you kinda go cold turkey, we'll figure it out? Cold turkey figure it out. That is that will shrink a bubble right there. That will Yeah.
21
0,08:26,000 --> 0,09:06,000
Yeah. No. No. No. We we like, the thing is is I had some connections that were nice. But, like, no. I had no clients just, like, waiting around for me to do something. It really was just a matter of saying, look. I think there's some real big missing gaps in the in the market for this kind of work. And especially as it related to, like, retail, fashion, anything fashion adjacent, there was a lot of issues with the existing systems out there. And so it was like, how do I solve that? And, you know, it took some time and a lot of effort in terms of the data analytics perspective, but at the same time, it felt like it was worth doing, and so it was a little bit of a leap of faith.
22
0,09:06,000 --> 0,10:11,000
Well, I mean, it's a it's a lot of it. Right? Too because there's I mean, there are so many data analytics companies that claim to do lots of things, and some do, some don't. I'm not trying to be Right. Beaten up industry. But Oh, I will. Kinda like kind of like the and kind of like the name of our title here, seeing the story in your data. You can see any story in data. If if you look for what you wanna look for as opposed to what's being presented, You can make up anything, and and I think that's sometimes the rep that analytics companies have gotten is that, oh, we'll help you do this, this, and this. And they're like, well, but I'll you know, that same data tells a different story. So maybe take that just from, like, a a first step of, that premise of how maybe you built your company differently to because because I know that was a problem, and I know that's a buying challenge for Yeah. Data officers or, probably in this case, for middle sized, more CEOs or, CIOs of how how we're gonna get the right or take the right data or marketing or whatever it is would be. Right? How did you guys approach that by an objection? Or is it was there a different one that you said, no. We have to do it slightly different to to make it believable? I know you do an algorithm too, so that helps. But
23
0,10:11,000 --> 0,11:37,000
Yeah. So so, I mean, as it relates to seeing a story in data, like, every story is open to interpretation. Right? Like, you can have two people read the same poem or the same story and come away with different interpretations of what that means. It doesn't mean that any of the language changed just like in analytics. It doesn't mean that the data changed. It doesn't even mean that what people are make assumptions people are bringing into it are wrong. It simply means that when you're doing it with a little bit of an outsider perspective, you're probably gonna miss some important parts. And one of the things that I kept running into, when it came to working with optimization firms within the retail space because I was I was the client before I was an actual purveyor. So, basically, part of that was the fact that I kept seeing why do they keep doing it the way that they're doing it. And part of that came from the fact that a lot of the time whenever you would talk to them, you would kind of learn that they came from predominantly a statistics and mathematical background, which in no way, shape, or form would I ever question or say that's bad. It's just the fact that if you don't have domain expertise also, it it like, that's the reason why I get a little bit antsy regarding some of the new data companies that keep popping up. Because a lot of times, it's like, why are you listening to a company that's never done your job tell you how to do your job? Well, because you because because I think what happens is, right, you assign meaning
24
0,11:37,000 --> 0,12:42,000
to something. So so if you strip away the meaning without context Correct. A different story versus I'll give you an example. I used to listen to Dave Matthews a lot Yep. Right, or even Pearl Jam for that matter. Mhmm. And and you'd hear songs and lyrics, or or Mumford and Sons is another one example. And then maybe in the last three years or so, I've kinda been on a little more of a faith journey. And I hear those songs now, and I hear those lyrics completely differently because they found in scripture some of the words they're saying and and the concepts. And I'm like, oh, man. Like, that that is a whole different impact of what he's talking about or they're talking about in song. And that data point of context of, hey. That person had been on something probably similar or they've they're they're they're speaking that you know, in high school, you wouldn't have gotten that, or whatever. And and I think that's the that's the point of context and and assigning meaning to something, and ten years from now may have a different meaning to me. Right. Anyway, so I think that I mean, I don't know if it's the right analogy, but but that's how it works. Right? If you don't have the right assigned meaning and context, the data could be shaped into mean almost anything.
25
0,12:42,000 --> 0,16:45,000
Right. I mean, the the basic thing is is the fact that so much of data science as a practice is built upon models and assumption and mathematical assumptions that have been in place for many decades. And most of the time, this was all directed initially speaking, a lot of the early models were dictated towards, you know, finance and trying to predict stock prices and things along those lines, which makes total sense. And, obviously, there are lots of systems that do that today. But the problem is is that when you take that exact same type of algorithm and then you apply it to a completely different industry, if you don't know that industry tremendously well, it's not that the results of that algorithm are gonna be wrong. It's just the fact that you're kind of making certain assumptions that there's similarity between these two things. And if you don't know enough to know what to question and say, no, that's in fact not true. Right? It becomes a major issue. And so it's really from the standpoint of something they're saying, the most important thing you have to do with any kind of, like, AI or any kind of, like, data onboarding within your company, whether you're a retailer or otherwise, is you have to have some melding of the minds between the data side and the business. If if you're doing data science without any kind of business acumen or expertise in that particular domain, it's a hobby. If you're doing it if you're doing the op the reverse where it's all business experience and no data, you're just gonna be very, very naive about some of the assumptions you're making that when you make decisions. So it really is bringing these two things together that's most important. And because of my background, because I came in to Home Depot at the time I did, and I got so exposed to all that optimization stuff and all that data science stuff. Because before that, I was not a data scientist on any level. But getting exposed to all of that and then being so entrenched on the business side and the decision making side, I think that's allowed me to kinda straddle between the two worlds a little bit and act as sort of like an in between. And so I think people sometimes misunderstand, and they think that if they wanna create a company, whether it's in data or otherwise, they have to be, like, the most expert person ever about this one technical piece or they have to be the most skilled person in this other piece. And it's like, no. You just have to have a sort of a shared melded perspective that gives you some ideas to how all these things fit together. Because fitting together is the heart's part. It's not doing You gotta be 51%. You gotta know it 1% more than your client. Right. I mean, the base the basic thing is the fact that, like, if we actually look at historical information, right, like, most data analytics projects, whether it's AI, GenAI, or otherwise, that tries to get embedded into an organization, they tend to only succeed around 10% of the time. The reason why the reason why that's happening is because the people who are making the business decisions don't understand the data science, and the data scientists don't understand the business. And so it's like these people don't know where to fit it in. They just see it as another way to improve their reporting, which is not really fundamentally what it's trying to do. And the data scientists don't understand the business enough to know how to fit into any of the process or the goals. So it's just you have this weird sort of, like, two sides of the same situation, but you have it to where both are naive about the other one's capabilities. And because they don't understand this and they don't have anybody to kind of, like, lead them the right direction or how to figure out how to bridge the two, A lot of things just outright fails simply because it this is the basic problem. This is like most AI, Gen AI, and all this kind of stuff. It operates on a height cycle. It's interesting. It can do absolutely everything. It it can in isolation, But the moment you're trying to figure out how to fit it into a specific puzzle, the piece has to fit. You you can't just say, I can just fit it anywhere. It it doesn't work that way. You have to bring that business experience into the question of where do we need it to fit and how what is it really good to fit for?
26
0,16:45,000 --> 0,17:23,000
Well, so, I'm gonna come back to that that, where do you use data? Because I think some people don't know how to what what that means. And and I'm gonna come back to that. But but is there an moment when you're first starting as you got you know, talk about your first client, and what's any moments you had to to to help you win the next client and or or or lose the current one or whatever it was. So because that happens. Like, oh, you set off, you know, the Mike Tyson rule. Like, everyone's got a plan till they're punched in the face. Right? Talk about your, your your journey of, you know, I got my first client this way, and and here was my moment that set off the next chain reactions.
27
0,17:23,000 --> 0,19:06,000
Yeah. So so my first client was actually pretty interesting because it was, it wasn't even a traditional retailer. It was me trying to figure out how small of a client I could actually take on with the new algorithms that we had designed. And so it was actually a TikTok social seller of all things, which is very data wilderness wild west. It it's pretty sparse as well. So it was like, does this actually work within this environment? And so first of all, it does, and that was fantastic. But what we basically learned from that whole procedure is just the fact that, like, I think we kind of assumed that even though we felt like the algorithms were good, we also kind of assumed that there is a point where you just simply have too little data to do anything with. And I think a lot of people will labor under this assumption that because they have too little, they can't do anything with it. And basically, what we kinda come came around on that was just and they're saying, no. Any anything is better than nothing. You you you if you're if you're able to use any data for any purpose whatsoever, that's great because we were able to take something extremely sparse and extremely weird and strange and nothing that any data spec could have prepared us for. And yet, we were still able to make something actually really happen for that client. And it just kinda gave us a lot of confidence, not only in our algorithms, but also gave us confidence in the under underlying assumption that being able to take data and just do something with it at all instead of just worrying about, oh, maybe it's not clean enough or maybe it's not enough, period. There's no excuse to not using what you have.
28
0,19:06,000 --> 0,19:46,000
You should always use And you can assign meaning to it and say, hey. This data whatever is these maybe it's client acquisition. Like, we we I I love selling these because it's easy to deliver, has good profit, whatever it is. But we haven't sold that many. And you're like, cool. Let's extract synthetic data or create synthetic data that replicates that at the same percentages. And what I'm giving a very broad, you know but the point is you can create data to help make some type of a prediction or direction of which then the synthetic data can exit as you put in real data behind it. Correct. And you also can weight it. I mean, you can say, hey. Real data is weighted at, you know, 90% or synthetic's 10. Yep.
29
0,19:46,000 --> 0,19:49,000
Yep. And you're There's so many
30
0,19:49,000 --> 0,20:00,000
algorithms around that. Right? So of how to do that. Talk about that a little bit because that's like a whole different world as not being a, you know, a mathematician or a data scientist. You've you've created patterns
31
0,20:00,000 --> 0,22:11,000
that that accomplished that didn't exist before. Right? So so it's part of your business. That's part of your model. Like, you have, IP. You have an intellectual property on. How did you Yeah. We yeah, the the way that we actually operate is actually pretty interesting because of the fact that the algorithms that we built were based less on having it to where so we don't just sit there and look at it from a longitudinal perspective. We look at it from an overall, like, village perspective is how we refer to it. Essentially, the idea of each product is its own little part of a village. And, essentially, these relationships between these products and how they behave when they're interacting with each other when one is shifting its price or one is being promoted and things along those lines. How does that influence everything else? And this is a very well understood concept in retail called cannibalization. But a lot of times, it's kind of treated as sort of like a secondary step in the process to where you're like, okay. Well, we're gonna look at cannibalization afterwards and just see if our underlying assumptions would cause problems with this as a overall, practice. But, like, when we were looking at it, we were like going, no. Sometimes we don't get enough information to just sit there and say, you know, you didn't give us enough of this, change in your business or change in your product's pricing or how you were behaving around it. So, therefore, we can't do anything for you. It was more from the standpoint of saying no. You can learn from the overall, offering as a category and as each of the individual members and how they influence each other. And you can generate a lot of assumptions to create new data around that information. So there's historical information for the entire category in terms of this very summary level information around history. But then you can take that and take these little interactions between these products and the influence they have on each other, and then you can apply that to those different historical trends. And that allows you to build in a lot of additional data that wouldn't be present otherwise so you can run it through algorithms properly. The ones that we built are a little bit less data hungry than some of the other traditional ones, but at the same time, it still is necessary to do that sometimes. You have to kinda, like, build out enough of a set for it to get any kind of assumptions through. Does,
32
0,22:11,000 --> 0,22:17,000
do your clients or new clients go, why can't I just use GPT for this? Does that question ever come up?
33
0,22:17,000 --> 0,22:48,000
I you know, I I've I've seen I haven't heard it from any of my clients. I I have seen it on LinkedIn and otherwise, though, and I'm kinda like, guys, just keep in mind, it's it's a large language model. It's it's really good for summarizing things and describing stuff. That's really good. But, the the way that I always try to explain it to people when they bring it up is I'm like going, look. If you had asked JET GBT a few years ago, what the next big smartphone should look like, it would never have described an iPhone.
34
0,22:48,000 --> 0,22:50,000
Maybe it's right.
35
0,22:50,000 --> 0,22:51,000
I
36
0,22:51,000 --> 0,23:48,000
just saying. I I I It's one of those things. You look back in history, like, politics assigned. Yeah. There are those who view certain candidates and certain people as nuts, crazy. Like like, oh my god. But yet they've won. And so you may look back in twenty years, like, holy crap. Yeah. Right? We were we were crazy. Anybody didn't believe we're crazy. And then you might look back like, man, nope. Legitimately nuts. Like, yeah. And you don't you don't know until, like, the effects are downstream from where it is, and also your assigned meanings to those those things. And and Yeah. Somebody who's starting, let's say, a data company or they're thinking about it Yeah. First of all, frame their mind right. Are Are you really a data company, or are you a revenue generation company that leverages data? So so is there a framing of of where you position relative to marketing versus who, you know, who buys it versus what your actual values? Like, how do you actually think of yourself as an organization, I guess, is probably the question.
37
0,23:48,000 --> 0,24:48,000
I I think of I think of Pallet as a revenue generation company and a demand generation company, that uses data to serve that purpose. I I think I get a little like, I I've talked to a lot of other people who are trying to found their own company or thinking about doing it. And the one thing that always gets me a little nervous is when they start off the sentence by saying, so we're going to use generative AI too. And I'm like going, okay. That's a little strange. Your priority shouldn't be the tool you use. The priority should be what are you solving? What are you fixing? How are you helping people? If you have it's kinda like if you went to a mechanic and he came sit there and said, hey. I'm going to use a wrench to fix your car. The question would be, why only your wrench? Why are, like, why why are you so focused on that one tool? It's a tool in your toolbox. It should not be the whole thing. That's not the focus. Why do you have such an obsession you shouldn't be obsessed with generative AI. You should be obsessed with helping people.
38
0,24:48,000 --> 0,26:13,000
Well, in the problem. Fall in love with the problem. Yeah. Fall in love with the problem. Technology is probably likely gonna be there to solve a lot of problems. Yeah. That doesn't sometimes help you sell it. That doesn't help you manage the implementations, which are you know, you need trust, access, contracts, things like that to get like, that's the hard part of building a business like this is that you're gonna be asking for very intimate details and knowledge, which means you have to get on the MSA and and have a bunch of insurance. So just just know that it's like, oh, we got this cool product and that's great. You're not big enough. You don't have revenue. You don't you know? So when you're thinking of your building your company, did you you know, Home Depot. Right? You get on that MSA. Holy cow. Like, that's a big deal. It's a huge company. It's hard to do, and they're gonna beat the hell out of you in pricing because of it. Walmart, same thing. On these midsize companies, the MSA requirements might be just you having a lunch, but, like, yeah, I'll get you access. Like, it could be a c Yeah. Yeah. Was that an intentional strategy? And, like, sorry, MSA, master service agreement. This is the legal agreement between you and a a provider and a and a and a and a, you know, client basically that allows you to operate. And and so the bigger the company, the harder it is to get on those lists because everybody wants on them. So they manage it down. Mid sized companies, though, sometimes it's just more relationship based to get it. Correct. I I and and, basically, the other thing is also the fact that the more I started looking at building the company because I took about
39
0,26:13,000 --> 0,27:57,000
it took me a good five or six months to kinda, like, figure out what was the positioning that I wanted to target and what was the overall strategy and who was I really looking at from a customer base. So the thing is is the fact that the more I researched this particular, like, field that I'm in, tangentially at least, around price optimization or retail optimization, the more I kinda figured out was, like, for the most part, there's, like, there's a number of vendors who would be considered my peers. And if you actually add up all of their clients between them, it's, like, less than a 200 people 200 companies. And it's like, the retail is ginormous. And, like, for the most part, they're all chasing after the same people. Like, they're all trying to get the Abercrombie's of the world. They're all trying to get into Dick's Sporting Goods or Tractor Supply and all that stuff. I mean, the companies that are doing really well right now. And from my point of view, because I had had experience in big retail and also had experience in small retail, It made me very sensitive to the fact that it was like, okay. Based on the cost structure they built through fund investments, through just how much people they had in their companies, they couldn't even look at some of these smaller firms at all. They they they had to them because it just wouldn't make sense for them. And so that was one of the intentional things was how do I help companies of much smaller sizes? And that allows me to avoid so much competition and RFP and all that kind of stuff for request for proposals for people who aren't used to that lingo. But around trying to get somebody to do this job for them or companies who simply thought no one was even gonna try to talk to to them about doing this job for them. Well, absolutely. Here we got a question from, LinkedIn here from another Kurt. Proper I'm gonna say proper spelling with a k.
40
0,27:57,000 --> 0,28:11,000
How do you see the balance between data driven decisions and a human element in leadership? So I I think that's a that that's an often disconnect of, hey. The data's telling you this. And Right. Leadership is that's great. But no.
41
0,28:11,000 --> 0,29:54,000
Yeah. So so the thing is, again, it kinda goes down to the fact that you end up having to have this melding of the minds because of the fact that data is never gonna be 100 complete across the board. There's always gonna be certain assumptions that you're having to make with any kind of data that you're taking in. And the big issue is is that a lot of times, the people who are most aware of data and most understanding of data don't necessarily have enough holistic view of the business to be able to figure out where it fits in really well and where are things that they have to embed certain filters or certain perspectives into when they're trying to come up with some sort of a recommendation or how do you frame a decision. And that's where you have to have those human elements, those people with experience, to sit there and say, okay. Here is where we have missing gaps. Here is where we have a challenge or a problem statement that we have just not been able to fully crack. And part of that means you have to embed the data people into your normal business operations. You can't just let them go into a cave and do coding all day long. You have to get them involved in the actual business. They have to go to the meetings that have nothing to do with coding. They have nothing to do with programming. They have nothing to do with data science. You have to get them acclimated to what does the environment look like so that I can take the data that exists today and understand this team was trying to solve this question. This one was trying to solve this. Otherwise, you have absolutely no idea what kind of target you're trying to hit. So it's really that combination of framing the problem as well as understanding what assumptions need to be there, what requirements need to be there to make something actually actionable versus just interesting.
42
0,29:54,000 --> 0,30:26,000
We we have another question. I can't post it for some reason. My guess is because one was posted by an AI, one was posted by a human. I really don't know if that happens. But it was it's actually it's still an interesting question. You know, have you considered expanding to an AI integration? So what I would think by that meaning is integrating with AI tools to make them smarter, more impactful. Is that part of sometimes a strategy of here's data and you provide that as, like, a rest service or some kind of service technically to other technologies? Are you or do you keep it in its own box?
43
0,30:26,000 --> 0,31:09,000
I mean, in our particular case, because our algorithms were built very specifically for the use cases that we were outlining, I don't think we would wanna go any kind of, like, more broad approach in terms of overall AI. It's really very specific to how do we solve these specific problems, especially from the standpoint of, like, I would say fashion retail. Mhmm. Because that's a different animal than the kinds of solutions that were being solved in the past, because their their stuff doesn't live long enough. So a lot of the algorithms that existed before that needed a little bit longer lives to the products, didn't really apply well to those. So trying to build out something specific for them means it's extremely targeted. So
44
0,31:09,000 --> 0,32:08,000
Yeah. I I I look at, like, a any midsize retail. Like, I don't know how you compete with a, Home Depot being Sure. Loads just copies them. It's just like that's their strategy. Whatever Home Depot does, we do. Totally fine with that because it creates a secondary market, like our secondary, you know, competitor. Yeah. Then you have, like then this true value who sits smaller store, more located. Alright. So you can repeat that model from any industry, but I think data could tell you here's some areas that are gapped that seem to have demand. And I don't know how you find this data to do this, but that's what some really good data analysis and data science can provide of, hey. If you're gonna have this retail to be competitive or at least to crack through so you could become one of them, you have to find some tips of spears. That's one of that's one of the use cases you can do, right, is to help them position their product or service to what is kind of missing or smaller on the or not really fulfilled by the other competitors.
45
0,32:08,000 --> 0,33:36,000
Yeah. Yeah. I mean, especially with our business, it was really something they were saying because I had experienced so much of the optimization space by the time I actually created the company that I kinda knew what were the big pain points and what were kind of the missing pieces that nobody was talking about. So one of those things was the fact that a lot of it is very front loaded, meaning that in a lot of optimization systems, you'll do be doing really, really well during the regular life of products. You'll be doing really well during the promotional life of products. The moment you take everything to clearance, things fall off a cliff. And basically, what their the cause of that is kind of hard to interpret unless you're actually have worked in that space. And then you can kinda sit there and say, but if you do that mathematically sound thing, aren't you constantly causing this particular buildup an issue in inventory? And if you're not specifically looking at it from that point, if you're just doing what is the most mathematically sound thing to do, in the long run, it actually ends up hurting you dramatically. And that's one of the big issues that you run into with some of the traditional optimization, especially as it relates to apparel. So for me, it was like, how do we build this very explicitly to solve for those missing pieces, those missing assumptions, and kind of, like, interpretation of the information to make it to where it's very uniquely posed to help people in this particular space. You know, I've
46
0,33:36,000 --> 0,34:09,000
I, if we have a date, it's usually at Marshall's. Just say it ends at Marshall's. Just leave that at that. Marshall's, TJ Max, HomeGoods, the whole route. Right? That's what you do. It's like, hey. Where you wanna go tonight? Well, we could just you could drive me around. Yeah. But here's the thing. I go shop there. And I find that if you the bigger the clearance rack is, why would I ever go look at the full the the full price rack? Correct. But you know, by the way, if something's on clearance at Marshalls, it is really not wanted. Because they've already bought all the stuff that's not wanted. And now they're saying we don't even want this stuff. That's where I go.
47
0,34:09,000 --> 0,35:43,000
Well, to that point, though, I think it's also important. A lot of times whenever we see things on clearance, whether it's in the store that actually created it or whether it's on a reseller like a a a Marshalls or a TJ Maxx or Ross or whatever, I I think the assumption that we make as consumers is this product really wasn't wanted. But in all honestly, a lot of that product, a lot of the stuff that ends up on clearance, especially in large quantities, it isn't because the product is necessarily bad. It's because what ended up happening was that the company did not understand how their assortment should be built in the first place. So how that typically plays out is that they'll assume that in their assortment, they need to have 35 different products. Sometimes this is pure guesstimation. A lot of times, they don't really have any analytics telling them how many problem products they should really have. They just go with 35, and they just keep going at 35. But in truth, one of the things that we look at with Pallet is, do you really need the 35? Can you do with 30? And the reason why that's important is because every other additional product that every choice you're creating for your assortment, they're having to hit a certain bulk minimum purchase for their inventory to be able to get the cost structure down enough to where they can sell it at a price that they think is reasonable. So that inflates their inventory, and they just simply can't move that much product. So inevitably, they have to get rid of that stuff through resellers of some kind. So it's not even necessarily that the product is bad. It's just the fact that it's just if they just had too much and they couldn't get out of enough of it. Yeah.
48
0,35:43,000 --> 0,36:58,000
I mean and there's a lot of other factors going in. Like, some of this is relationship based. Hey. Listen. If you buy this lot off of me so I can make my quota, I'll make I'm gonna the next one, you get an extra whatever. You know? That's right. There's that has to happen. There has to be on the buyer side. I need you to move this lot. I know it's more than you need. Take it. Put it to your stores in this other, you know, depressed area, and and put it all in clearance, and you'll sell all of it right away because that's what you're asking. And there's gotta be a strategy for that. I mean, I would think that it's just more like, oh, listen. If you don't buy you haven't bought enough to fulfill your contract, they're gonna buy the clearance rack so they fulfill some piece of the contract as well. That's correct. That's part of what you can solve is, hey. Listen. You have to fulfill your contract. These are terrible relative prices, whatever it is. Minimum, minimum, minimum. At some point, we already know that they're gonna have whatever, a giant sale on whatever you're you're buying. Like, that's when you're gonna go fulfill it because right now, they're just they're pushing price to try to get their profits. And you can show that based on historical data of, like, just wait to buy till this month or this sale or whatever in a sense. So I'm I'm I'm being like, I'm trying to No. No. No. You're using your data to counteract their data science. That's a good deal. Something Yeah. No. No. No. I think fundamentally, though, what you end up finding is is that a lot of people in these businesses, because there's not really
49
0,36:58,000 --> 0,38:02,000
an easy method or or very super obvious method for figuring out how many products you actually need and then how to price them, it it's a real big issue because at that point, they become much more susceptible to those kinds of arrangements that end up putting them in this really weird space where they have all this inventory. There are people in their stores are being, like, flooded with it. And, of course, we've seen how many stores have pulled back on their actual labor. So it's like it it it gets really, really problematic from the standpoint of, like, you just overweight your entire organization with inventory, and then you put it on clearance. And when people walk into a store and see a lot of clearance, they assume the same way that you or I would as consumers, which is, wow. This is just a bad product when in fact, it's just it's a lot of it's actually on the buy side and not on the sell side. That's the problem. Yeah. I like it because I'm trying to help with. I'm a little taller or whatever, and I can find some stuff that's typically no like, you know, it's not the just didn't sell. I'm like, alright. Yeah. Yeah. Great. I'll take that. Right. But it's but you're but it's good that they at least have the option for you. That's nice. Right. Exactly.
50
0,38:02,000 --> 0,38:27,000
Tall and fat. Like, that's the that's the that's not they call it big and tall. I just said, why don't you just call it as as fat and tall? Like, fat fat and bigger. Big and big and bigger. I got another one here. Someone who's anonymous. Here we go. What do you think about that statement? That data is only gonna get more important and critical to success of organizations. Is it is it the new oil for organizations?
51
0,38:27,000 --> 0,39:13,000
I think it's always been the oil. It's just the fact that a lot of people have figured didn't figure out how to burn it properly. I mean, like like, the companies that figured out how to how to do it and how to use it effectively are the ones that are gonna be winning more and more frequently. Like, Home Depot, having been there, I I know they had a lot of data sophistication. There were but there but at the same time, it's not like they had everything figured out. Even the Walmart's of the world and the Amazons of the world don't have everything figured out. So I think the underlying thing I would say is that just because, you know, just because you're small, it doesn't mean you can't do something really interesting with data to learn how to be special and how to stand out. So don't be investment you like for them. I mean, like so so I hear that. I'm like, oh, you know,
52
0,39:13,000 --> 0,40:04,000
this is expensive. And, you know, like so could you talk about for small companies to leverage your types of services? Because I think a lot of people say that sounds great. There's a presumption it's expensive because all the big companies do it. So So and and not saying that you don't get your fair price for what you do, but I don't think it's quite, price expensive because of the return speed. So I mean, like, the ROI is there. Could you wanna talk about, like, a like, a typical business case that you get brought in on from the perspective of ROI? Because I would have to I'm guessing here, but but I'm gonna guess the object the number one objection is what's it cost? Like, what's the investment? Do we Correct. And you're my ROI. It's the CFO saying I'm not funding this unless you have data. Yeah. Yeah. So talk about that a little bit. Because if you're listening and you're like, alright. I'm gonna call this guy because I never knew that you could do it like that. What would your answer kind of be to that CEO?
53
0,40:04,000 --> 0,41:49,000
What I was in there and say is because of the way that we built our algorithms and the way that we approached our back end to our company and the fact that we haven't taken on a lot of investment that hasn't blown up our cost structure, in fact that we haven't built out our company and some massive amounts of people, the fact that we've been trying to be very conservative about how much we put it into that and really focused on the technology side, which is a lot easier to scale. One of the benefits of that is the fact that we're generally far less expensive than the traditional option. So if you've run into issues where you talk to them and it became really scary, that's something to keep in mind. We're definitely not urging you an exorbitant amount, and we're not even gonna start with an exorbitant amount and force you to negotiate down. We've come up with what we think is a very fair approach that allows us to scale based on the size of your business. So, like, for example, I said the very first client I ever had was a TikTok social seller. That's about as small as you can go in retail. And we were able to make it work because it's just a small tiny fraction of a percentage of their revenue, and then we're growing their business by anywhere between 12 to 18%. It's like when that's what you're talking about, the ROI pretty much is phenomenal regardless, and it also allows us to be a very effective business. The other piece is also the fact that because our algorithms were built understanding that not everyone's data is gonna be the same and not putting all of the onus on our clients to get give us the data, we actually work with them during implementation to make it as fast as possible. So if you're looking at trying to get into price optimization or assortment optimization and trying to figure out what kind of products you should have, that's one thing to reach out reach out to us. And we've gone been able to get up and running a test with companies in less than two weeks.
54
0,41:49,000 --> 0,42:02,000
That's bad. Which unheard of in the space. So Which means you're just you're immediately getting value back. You're merely learning something. Correct. And you're, you know, if you're pricing improves by half a percent every month.
55
0,42:02,000 --> 0,43:02,000
Yeah. Your year That's okay. You've just killed your year. I mean, like, it's awful. To give you an idea, the very first time I ever implemented another price optimization solution back when I was on the retail side of things, like, we were nine months into implementation, and we were only about halfway there. So we had gotten literally no value whatsoever. We were just spinning our wheels trying to get them the data that they had requested. And all it would have taken to get it up and running faster is them embedding people into our company for a short period of time and taking some of the responsibility for building out the assets. And because a lot of traditional systems don't work that way, it ends up meaning that you're spending anywhere from nine months to two years before you even get any value back or any recommendations. And that was one of the things when we were building out the system and our overall structure, we were like, that can't be the case. Like, small companies can't give me an implementation fee
56
0,43:02,000 --> 0,43:12,000
of any value and then go two years without getting any help. Like, that's not They just can't take that. They'll never make that investment ever. Correct. That that in sales or a digital inventory.
57
0,43:12,000 --> 0,43:17,000
In our case, it's like, look. We can get you a test up and running in two weeks. After that,
58
0,43:17,000 --> 0,44:42,000
once you do the tests and you're satisfied with the result, we can get you implemented within a couple of months. It's it's very fast. We we Part of that refined use case too. Right? Part of it's your upfront consulting of what's the use case we're gonna go tackle to get the highest value. Then you can knock use case by use case, which creates projects. And then as you create value, then they start budgeting for those projects of, hey. Yep. It's cost. You know, we know it costs about that. It's gonna fund this. Eight, nine weeks, do it because we know it'll pay off in, you know, your your fifteen months from now or whatever it'll be. Right? In our case, the payoff's gonna happen within three to four months after you actually agree. Like, it's really not So that's what I say. So you get an ROI in the first in the in the fiscal year, which is super important for buying services. If I can pay for your services, they will buy more and more. Correct. That also positions you for the two year, three year investment. They can't like, there's no physical way to repay it, but you're like, we can do that. We know it'll work. It's a higher investment, but the payoff should be a hundred x. It's just gonna take five years to do it. Yeah. So you could do that, but you cannot typically position yourself right away into that or your and you have to staff with that. So are you even chasing those kind of deals? Are you looking for the quick use case stuff that's doing this right? How do your how does your business position that? Because that's that that is a different delivery model. It it kind of it kind of works both ways. So so, for example, we do go for typically shorter term type of things. Like, for example, we might go for a contract that's, like, a year. But then again,
59
0,44:42,000 --> 0,45:50,000
we don't count the year including the implementation. The implement in other words, we don't count if we're just doing if you're just doing work trying to get it up and running, we don't count that towards your subscription. We sit there and say no. Until we're actually delivering something to you, you're not paying that. That's not your thing. Right? We don't have to worry about that part. Once we get you implemented, you get a the year contract. The only thing that we do is we have levers in place from a from a negotiation perspective that we provide to clients since they're gonna say, hey. If you agreed to two years, then the amount that we charge goes down. If you do three years, it goes down at the end. There's other different agreements that we specifically built in because we didn't wanna have negotiations be about a conflict where it's like, oh, well, you came in really high, but there's no levers provided to me to get it down. So I just have to argue with you. I didn't want people to argue with me. I was like going, look. I already made the base price pretty low, especially comparative to my market. So it's already low. I wanna get you in as a client. I wanna make sure I'm helping your company. But at the same time, I even give you specific levers that you can pull to get even lower, and that's fine with me.
60
0,45:50,000 --> 0,45:58,000
That's right. Just conscious of time. So if you if you, you know, give me the sixty second thing you would say to yourself if you're starting over again.
61
0,45:58,000 --> 0,46:00,000
Oh, god.
62
0,46:00,000 --> 0,46:08,000
Not now because you you can make that claim and start over. I'm saying you go back. What what would you tell yourself?
63
0,46:08,000 --> 0,46:38,000
I I would probably there there were a number so I I think it would come down to try to get someone interested before you actually, like, quit your job. Like, I I know that Shark Tank teaches us, are you fully invested? Are you a % in, or do you still have a day job? You know, honestly, it's really, stressful on your family to start something without something already kind of in the wings. When you're starting something completely from scratch, it's extraordinarily
64
0,46:38,000 --> 0,47:01,000
stressful and anxiety. You gotta have some because you have to remove the safety net to be all in, because you can never accomplish what you can really do without it. But I agree with you. Get at least one signed up Yeah. Working on it, take away your Netflix on the weekends, work through it, whatever you gotta do. Then it's a lot easier to make that move, or or get laid off or or ask to leave like I did and just
65
0,47:01,000 --> 0,47:41,000
not be able to get rehired and now you're in it. Yeah. And the thing is and the thing is, though, for the most part, like, I would say and say again, yes, you have to be all in, but at the same time, it's like, you know, your your calculus changes around decision making when you're a parent, when you have a family to support and all that kind of stuff. Like, your calculus has to be a little different. So I always find it strange when I hear people who are like like Shark Tank people who are like, well, you know, you should be all in. You should have quit your job. You should do all stuff. And I'm like, like, dude, you created your company when you were in college. Like, you you didn't have these kinds of issues, these kinds of things that you had to take care of. Like, I didn't actually create my company even though I had friends telling me for years I should
66
0,47:41,000 --> 0,48:41,000
until my student loans were paid off, until a lot of things got paid off. I I'll I'll leave people with this. Like, right? First of all, Shark Tank isn't real. It's Yeah. It's a show. It's produced. Correct. It's made for drama. Yeah. You know, it's you know, stop and I you you know, stop stop trying to raise money and just go out there and raise health. Go sell stuff. Yeah. Like and and stop doing that. And I I think, you know, I was on the blocks, and that's one of the things I've like like, just bring the cash register. That'll attract more people than anybody. And and I think when you plan your company, whatever else, I don't subscribe. I love the stuff Alex Hormozi does. I like the Shark Tank from a drama standpoint. I don't know what Gary Vee does. I really am confused by his brand. But, the but but I don't think everyone should have the mindset we're gonna get rich. The truth is replace what you may you could have made in a w two, and you're killing it. Yeah. Yeah. Absolutely. You're absolutely murdering it. If you made a hundred and 50 a year and and you're like, I can make a hundred and 50 a year on my own, you're killing it. Just be just keep that perspective in mind. Yeah. You don't because you don't report to anyone but yourself.
67
0,48:41,000 --> 0,49:55,000
Yeah. Absolutely. And the other thing is also the fact that you're allowing yourself to democratize what you know. Like, the basic thing is when you're working for a company, the only thing you're adding value to is the stockholders of that company. You're not actually helping anybody outside of that. When you are actually on your own and you're able to work for different companies or different clients or different people, the benefit of that is huge because you're able to take whatever it is you know that's unique to you and important and being able to apply it to a large a much larger base, and you're able to help for more people. Like, I'm not I didn't create Palette because I wanted to become rich. I've created Palette because I kept seeing things falling through the cracks. I kept seeing things in the traditional market that I just didn't agree with, And I kept seeing situations where people would get excited and then everything would collapse after at the end. And I'm just like, I I don't want that for people. I don't want people to go through those really painful stretches trying to get something up and running that they need for their business and being scared of that. I'm like, I'd rush hour, get you up and running and get you a benefit because Right. That keeps you in business, that keeps your people in payroll, that keeps all of that stuff going. I've seen too many people get churned in the machine. I don't want to be
68
0,49:55,000 --> 0,50:52,000
the person responsible for that. I wanna be the person who tries to fight against that regardless of who asked me to help them. Yep. I love them. Thank you so much for coming on today. I appreciate it. Of course. Of course. Thank you for having me. Get a hold of you at Palette, p a l e t t e, analytics dot net on the screen. If you're listening, check it out. Kurt, thank you. As a fellow Alpharettian is it Alpharettian? Alpharettian. I think it's Alastorettian. I don't know. I don't know. It's gonna be warm here next week. It's all I know. It's in January. We're filming this. Thank you so much. I would put you in the periwinkle room. I'll be right back with you. Let me say goodbye, and I'll, I'll grab you real quick. Okay? Thank you. For everyone who made it this part this far in the show, thank you all once again for listening. Just take a moment, hit the follow button, that that you get the latest episodes. That way, it's really important for the guests, to get their missions out because the more people listen, the more people the more times Apple and Spotify recommend it. And it helps us move our mission forward to help entrepreneurs cut the tie to everything holding them back. Thank you for listening. Until next time.
69
0,50:52,000 --> 0:50:54,000
Thank you for joining us on this episode of Cut the Tie. Let's stay connected. Please hit that follow button on Apple, Spotify, or YouTube. And if you're ready to advance your entrepreneurial journey even further, join our free community at facebook.com/groups/cutthetie. Cut the tie to everything holding you back from success.