Joel is the CEO & Co-Founder of Alloy (www.alloy.ai). Alloy’s platform aggregates and analyzes information from all sources in the supply chain, providing insight and visibility into consumer demand.
Discover more details here.
Some of the highlights from the episode:
Follow us on:
Instagram: http://bit.ly/2Wba8v7
Twitter: http://bit.ly/2WeulzX
Linkedin: http://bit.ly/2w9YSQX
Facebook: http://bit.ly/2HtryLd
Joel is the CEO & Co-Founder of Alloy (www.alloy.ai). Alloy’s platform aggregates and analyzes information from all sources in the supply chain, providing insight and visibility into consumer demand.
Discover more details here.
Some of the highlights from the episode:
Follow us on:
Instagram: http://bit.ly/2Wba8v7
Twitter: http://bit.ly/2WeulzX
Linkedin: http://bit.ly/2w9YSQX
Facebook: http://bit.ly/2HtryLd
Hello and welcome to the leaders in supply chain podcast. I am your host rather Palomar you and it is my pleasure to have with us today. Joel Joel. He is the CEO and cofounder of alloy. Alloy is a platform that aggregates and analyzes information from all sources in the supply chain providing insight and visibility to consumer demand. They can tell you what when and how much stock you need so you don't have to guess. They basically join companies like Google, IBM, Microsoft and SAP and being named as a best in class provider in the consumer goods technology, magazines, 19th any other words? They will basically top 10 vendor for both the artificial intelligence and supply chain planning business functions. They recently closed the 12 million series a round for their artificial intelligence poets play chain platform, which brought the total funding to 15.3 million. They currently have offices in San Francisco, Vancouver and Berlin and I'm very excited to have Joel with us today to share more about about himself and about the company. Joe, welcome.
Speaker 2:Thank you. Excited to be here. Excellent. So I'd like to start with yourself first, right? Because, uh, and I got this quote and my team got this quote from, uh, from you saying that your career has really been spent building big data platforms in lots of different spaces starting in retail and moving over into financial technology now with alloy. But for you it's all about the challenge of how do you connect, um, things, how do you connect disparate data systems that sits across a lot of different companies. And that's, that's what you're saying, that that's the core of what you tried to do at all, although as well. Um, I'd love to maybe get you to share a little bit with us how you started your career in this, what, what drove your passion for, for data and for, for technology and you know, just a little bit about your background in general.
Speaker 3:Yeah. So I initially came out too to the San Francisco Bay area a little over a decade ago as a graduate student studying economics. And so definitely had, at least from an academic standpoint, a lot of interest in how firms operated, um, how they can do so more efficiently. And, uh, it was doing a lot of applied statistics on data sets and, uh, I eventually transferred over into the private sector. There was just so much exciting stuff happening in technology that I wanted to just felt that there was more opportunity and was able to start working, as you mentioned, with, with large retailers to start with and kind of a data science capacity, helping them understand how they could operate more efficiently with data sets that I would have drilled over a as a graduate student. And it's, um, in, in tackling that problem, I think you just see so much opportunity for companies to operate more efficiently. The data is there a, it's a question of how do you kind of unleash it and help you make better decisions. And I think that's true certainly in supply chain where there's so many companies that are trying to coordinate. And I think the problem is particularly acute here, but it's true in other industries as so, uh, you know, it was able to work in the retail industry for awhile. Switched over as I mentioned in the financial technology, found that there were a lot of similar patterns and I worked with a number of engineers who became my cofounders and we had solved this a couple times and we felt like this was a space that was just right for it.
Speaker 2:Yes. A lot. And it's a, it's fascinating. Um, yeah, so basically you're, you're, you're from a technology data, uh, academia almost saying it's, it's, it's assassinating because we've actually had on the podcast drove, uh, just a couple of podcasts ago and his runs, the company called, she bought and he also has from the academia and then he kind of transitioned into the entrepreneurial world. So it's interesting that the, and this seems to be some commonality and it's, it's also that level of how do we make it more practical, right? Then how do we take this maybe sometimes theoretical problems and really applied in day to day life. So I'm so interesting too to hear your, uh, your start and let, let's, let's go a little bit into allies. So low is, um, you describe it as a digital native company built for today's consumer and the way they are consumers buy and you've basically been inspired and want to take a look at$20 trillion consumer goods market, right? So you aim to equip brands with the modern technology platform to help them become more demand driven, more compete or complete and to grow. Um, now to get some practicality to that, right. Then the, and, and, and make it really tangible. Maybe let's take an example. How are you solving a problem of one of your clients? Let us go in and share with our audience some concrete case study.
Speaker 3:Sure. So let me walk you through an example of something that happened over the holidays. Um, in the consumer products, um, kind of world, um, you know, the November and December, oftentimes, um, at least I guess in certain parts of the world are, are a huge percentage of your sales, right? In certain categories. And, uh, as you mentioned, you know, we're really trying to help companies become more demand driven. What does that mean? It means being able to respond super quickly as demand patterns change and customers are pretty fickle. Um, probably more so than they ever have been before. So, uh, we had recently started working with a large consumer electronics company. Uh, it's one, I'm sure many people on this podcast probably own their products and they were going into the holiday period, uh, and had shipped out a bunch of product to a major retailer. Uh, is this is a new product introduction where you often have the least amount of kind of historical information about how this product's performing, what replenishment patterns look like. Um, and so they had shipped us in and they were getting the last set of orders for that holiday period we're coming in. And what we were able to do. Uh, we were, we knew what the company at this point in time was to take a look at all those borders as they arrived and to make recommendations that it actually looked like they were underestimating the amount of product that they would need. And the key reason they were doing this as we kind of drilled into the detail is that the out of stock rate for this product was actually quite high. Um, and as you, that wasn't being factored into the retailer's own plan for how much product they needed to offer. So they were looking at, you know, the historical pattern of how much was being purchased, not taking into account that, you know, whatever it was, 20, 25% of the locations were already out of stock. And we could project out and say, Hey, if you don't act on this, you're out of stock levels continue to, to increase. And in fact, you know, the, the amount of money that they would have left on the table was, was many millions of dollars. Um, and so that's the kind of insights that we're trying to bring. It's just one small example to our customers of okay, we're tying together. There's greater demand that was expected. It resulted in out-of-stocks. Um, it's gonna result in further out of stocks unless you make a series of adjustments and you take these things into account in your forecasting that ultimately neither the retailer nor us our customer was doing at that point in time. So being able to highlight that it's a win on both sides, both the retailer and our customer win because they have the right amount of product and they're able to recover, you know, many millions of dollars. One of the key things that happened is they made some adjustments. They didn't make all events and you could actually see you aware of those adjustments were made. They did much better where they weren't, uh, they actually saw those out of stock rates jump. So it was kind of a, maybe not the best form of validation. You hope that all the things are solved, but sometimes people can't resolve the issue, but then they get that validation of, wow, you really did forecasters within happen and unfortunately it did.
Speaker 2:They're interested. Then a lot of times the PR, I mean the proof is in the pudding and sometimes you do need to apply for certain areas and then you kind of get convinced and then you want to apply it for the whole pie. Right. Because you realize, holy cow, I could have, I could have made even more money and I guess it works on the other side of the story as well I guess. Right? I mean does it happen on, on the other equation where you overstock, like where you have too much token because you're, you know, again maybe you're following historical data and people actually stop buying at some point product or stopped being so interested in the product, but you are kind of only forecasting and using the historical data you keep producing. Even if your demand has gone down. I mean it would, it would work in that kind of concentration as well, I guess. Right. You probably have cases, case studies like that as well.
Speaker 3:Absolutely. Um, and you know, I think the numbers that I've seen from various research is usually around 8% of revenue is lost from out-of-stocks and overstocks, you know, and it's actually about 50, 50. If you break it down, uh, out of stocks are clearly one that are probably on their face more painful because you're like, hey, we a customer sale, but overstocks to your point or are just as damaging. I mean, it's a bunch of capital that's tied up in products now that you can't sell. You're going to have to discount, um, probably too to get rid of. And I think, you know, when you turn on alloy for that first time, you get a level of visibility that you've never had before. I mean, cause it's looking at every single skew, uh, at every single location. Um, you know, across, you know, the individual stores, the dcs, the fulfillment centers, you know, uh, further upstream your warehouses, you're, you're manufacturing plants. If you were going that far back, um, and you often instantly kind of see like, Oh man, yeah, we're totally stock, but you know, we have 80 weeks of supply, like of this product and man, we're running up here. So we're trying to provide that, you know, kind of that Aha moment of okay, there's a lot of rebalancing that can be done. And oftentimes the high level metrics might be okay. I mean that's the other key thing. Somebody might look at their overall, you know, in stock percentage or weeks of supply that they're holding and that number might feel fine, but you drill into it and you find all the examples where it's often, you know, we're, all these companies are operating at such scale that that's many, many millions of dollars and in so many cases of opportunity. Hmm.
Speaker 2:Yes. Um, and I was wondering, maybe it's also to also give some examples, because I know that you work with, with quite a few, quite a few industries. I mean, I think you'll start getting the consumer consumer goods. So I'm guessing you've just mentioned the consumer electronics. I mentioned, I'm, I'm guessing it's also a big market please hospital in consumer goods. Right. Um, are there other industries where you had a good level of success? I don't know. I mean
Speaker 3:maybe, maybe we can share some, some, uh, some other examples. Yeah, absolutely. So our target customer today is going to be somebody that sells a consumer product across and that can be consumer package goods, you know, consumer electronics, you know, apparel, footwear. I'm kind of fashion it really kind of running the gamut. Um, you know, I think if you look at the top 10 biggest CPGs in the world, we currently work with, three of them are so, so, I mean, we definitely work with a lot of very large, um, you know, multinational companies. We also work with lots of up and coming startups and that's actually pretty exciting for us to be able to see, you know, the kind of the largest companies in the world that have some of the best supply chains and are incredibly efficient and have lots of resources. And then you, you know, the companies that are being started down the street from us that just released some new, you know, consumer electronics product and they're just going to market. So pretty wide range. But yeah, CPG, consumer electronics are our two biggest verticals today. The technology that we've developed is not specific to consumer products. Um, and we do have some companies we work with outside of it that are in chemicals and other areas, but, um, but that's the market that we really started to focus on just from a go to market perspective. Cause it's obviously a huge space and you've got to kind of pick your beach head.
Speaker 2:No. And for sure, um, uh, I think this, this applies to two across industries is just that the low hanging fruits in terms of how fast they move on the shelf, it's, it's definitely they're going to see more industry. It's not as, as often that people run out of cars, stuff like that. Um, and I, I, we go on this question and I think it's something on the mind of, of quite a few of our, our listeners. Um, Eh, the, the question was in terms of the prerequisites that you need, um, in order to make other, to plug in, basically I'll let you right into a company, right? So, um, you need a certain set of data or certain set of quality of data or you know, I mean it needs certain, um, um, I guess quality is the right word, right? Um, or certain maturity of the, the supply chain of the company. Alignment of displeasure. And I'd love to maybe ask you or walk us through, I mean, what's your certain, I mean, let's say that a certain companies are very mature. You mentioned top 10. We are a global consumer goods there. They're very mature from their supply chain, but maybe some of them are not. Right? So what's your prerequisites in terms of, okay, I can work with this client or I can not.
Speaker 3:Yeah. So this might be a little surprising, but, um, we, the level of maturity of your supply chain is at the end of the day, not super important for whether we work with someone or not. And as I was mentioning before, we work with companies that are, you know, just coming to market and maybe they've been selling direct to consumer on their own website. They're just starting to work with their first couple retailers and distributors. Um, you know, so they're very nascent, uh, and really learning this as they go, you know, to the other end of the spectrum companies that have been doing this for many decades and you know, have deep expertise, um, where we tend to thrive the best. Is there some level of complexity in this, right? You're working with multiple partners and one of the strengths of our platform is one of the reasons that you can kind of get it up and running quickly is we work with the, we, we build connectors into those partners, right? So those retailers, those distributors as three pls, we, we already have in many cases connections with them. We could just start pulling data on your behalf. And so the bigger thing is just an attitude and a desire to be more data driven as a company. That's like the number one thing that we're looking for. We don't need a lot of it help to get up and running. It's just, okay, are you, do you have that kind of desire to use all of this external data in many cases that is available to you but often under utilized and you know, are you motivated to kind of put that to work for you? So that's what we're looking for. That's tends to be the companies that adopt are the business units within these companies that adopt earliest are those that kind of see that vision and say okay, I can move beyond the world. I've been doing in some cases for a long time and if I can collect all this data together, I can operate it at a higher, much higher level than I've been doing before.
Speaker 2:Interesting. So basically even if it's, um, uh, and, and you know, I'm trying to get my head around as well. So, even if it's somebody that doesn't have like, I don't know, maybe the power, the operating in a lot of companies to write on in spreadsheets and so on. Non but number one to the most advanced artificial intelligence tool that there is out there, even if that they use, they're using that. So basically you can, you can come, I mean although it can come in and the, and then you become how to, I mean it's almost like you, you, you're their demand and forecasting and demand, the demand, the software is like, it's like you can, you can, um, you can become there. I Dunno. And then it's a combination of that. Then Erp, I mean it's almost like I'm trying to can, I don't know what's the right term to use and what you stand for, but it's, it's, you offer them a suit of, of data driven insights.
Speaker 3:Yeah. And I think one of the key parts of that is because sometimes we'll get this question, how are you different than a European were, were very different from an Erp. You know, that's going to be collecting all that internal data that you have and you know, very operational in nature. We're helping you collect data from all your partners, right? The supply chains and network. There are many companies that you're working with. They are sharing data back with you that you are probably do handling and a pretty manual way. And so we're going to help you go collect all that. We're going to be the tools to be able to analyze it and use it to make you a better partner. Um, so you can operate more efficiently internally, but you can also operate as a better partner and help help those external companies. So oftentimes people get started by just selecting a couple of partners. You don't need to do everything at once. They'll say, hey, you know, we just want to connect the set of partners. Let's see how it works. Let's see how this level's up the way that we operate. And then as we get those proof points, we move into other parts of the business. And I think that works really well with our business model because we can turn these on very quickly. Um, and so you can, you can test, you can evaluate, you don't have to worry about, you know, as super long term, uh, implementation process. You can see that value in you, you can use that to drive kind of, you know, further investment versus making that massive investment up front
Speaker 2:go through. So it is like, okay, I'm proctoring gamble. Just giving an example that I just want to analyze my, how it works with my Walmart sales. Right. The let's say or are for what it was for market in the world.
Speaker 3:Exactly.
Speaker 2:Yeah. And, and basically, okay, let's deploy it for this. I Dunno, just Texas, hey or I don't know, California and let's see how that works out. Then if we, if we're happy then we'll, we'll, we'll go at scale, something like that. Right then, then, um, basically if you succeed then, then the clients give you more and more of the, uh, more and more of their visibility and supply chain, which is, which is interesting actually. Yeah. Cause it gives it to explain a lot of, that's a big bottleneck for a lot of, I don't want to name big software companies, but all this, whenever it comes to large, large implementations, that's a huge problem because people don't want to all of the sudden, and there's been so many failed once, right? Then they don't want to take the risk and they don't want them. They want pilot. So they want smaller types of proof of concepts. Right. So I guess in that respect it makes it very easy for, for a big company to use you guys or for a small company. Um, so that's great actually. And I loved also too to see how does it work with ECOMMERCE, right? With Amazon, with Alibaba, with, I mean all these platforms to sell a online you can, I'm, I'm guessing you can plug in very easily into those as well.
Speaker 3:Yes, absolutely. So that's one of the big value pro positions you offer is we're going to sync all of your ecommerce in your traditional, you know, kind of a retail business. And so you can have one view of all of that. And you can overlay them in really neat ways cause you can get Super Geo targeted and understanding that, you know, these are areas where maybe I don't have a lot of traditional retail presence but I have a lot of ecommerce sales and so hey, I can use that, you know, to maybe drive, I should be in retail stores in that area. Um, so ecommerce is certainly a big part of what we can bring in a, there's a wealth of data available there because that data tends to be more digitized and kind of ready to be consumed. And you can get it in more detail. Um, whether that's being done first party, um, or it's being done through a third party like Amazon.
Speaker 2:And how long does it take, how long does it take to deploy? I mean I know that you have a showcase so does listed on your, on your website. I think this is a big question for a lot of, for a lot of companies, especially the larger they are, the more they would ask you how, you know, how long
Speaker 3:and how complicated this is where I tend to get in trouble cause I get prone to get excited. But, um, you know, we signed a customer earlier today, um, they wanted to just start, um, and they were like, hey, as fast as you can, we want to get live with a major retailer. Uh, they'll have, there'll be up and running and using our platform and two days, uh, you know, in that particular case. So it can be extremely fast. That's not always the case. Um, you know, the bigger companies, they tend to be more pieces of data they want to bring in. Um, it can certainly be more lengthy, but I think one of the big trends I'm seeing happening in enterprise software is that speed of implementation is a bigger area of differentiation, right? It's, uh, and that's something we really pride ourselves in. It doesn't mean that everything is going to be set up and perfect. Uh, there's a lot of change management. That's usually what takes the most time organizations adjusting to the fact that they have these insights and they could do things in a different way than they had before that that probably takes 80% of the work. But you do have to kind of work and start that adjustment, get everything connected. And so, you know, they range from, you know, a week to, uh, it can be a couple months, but, uh, uh, it's, I think very fast, uh, for this, this space and it's something we continue to invest heavily. And our goal is to continue driving that down. I mean, we want to be that, you know, we have this point, I don't know the exact numbers, three, 400 of these connectors into these partners that will continue to grow. I'm sure over time there will be thousands where, uh, uh, prospect can come to us and say, okay, these are, these are the contract manufacturers I work with. These are my three pls, these are my retailers and distributors and my ecommerce platform. And we could say, you know what? We already have all those. We just need to kind of turn it on and the data's flowing and you've got that, that kind of system might not do exactly what you want on day one, but you have that end to end visibility. That's, that's really pretty turnkey, which is certainly possible to do.
Speaker 2:Yeah. I mean the mall at the mall kissed at this client and they just as that you're going to have, the more you're going to be able to do that. Yeah, right. A good question. That'd be another very good question that we go to is, um, is from decia and he was asking what are the pyramids is that you're using is input to forecast the demand because demand and retail is dynamic and is for of a changing. And it's then you have, you know, service levels you have, you need to have some buffers that you'd take into account. Um, you have seasonality, you have, you know, I would have still, uh, um, or, or, uh, overstock. I mean, how do you model some of the, what are some of this pyramid is that you are kind of running through a in the tool. Sure. So
Speaker 3:I think the, the ones that tend to be the biggest, uh, predictors of future kind of performance, certainly it's going to be historical demand patterns are number one. Assuming you have those, um, you know, so looking at historical patterns of ideally that product, if it's a new product introduction, you're going to be looking at similar products. Um, you know, things within that category, um, that you can use to give you that historical pattern of what is seasonality look like, what are kind of longer term trends that may be going on. Um, you're also going to want to bring in all of the a promotional data. Um, so particularly in consumer goods, that's a huge shaper of demand. I need to know about historical promotions, insignificant events. That could be the super bowl or a holiday that's on different weekends. That's all information by the way. We bring it automatically. So all of that kind of, you know, school starts, you know, this week in this region. All that's information we can bring in. You bring us your marketing calendar, we're collecting all the historical demands. Uh, you can also bring in other external factors. They tend to usually be, you know, less powerful drivers. But I think there are things, people often when include that can be macroeconomic factors, can be things like weather. Um, and you know, all of those are basically thrown into, um, you know, our, our platform. It's going to collect all those things dynamically. And, and then, uh, it starts running these bottles. And you know, the key way that I think one of the key things that we always explain what that is, we use a lot of different methods. Some of them are um, have been around for decades. I mean I was talking about kind of coming from the more academic world and these are methods that were discovered 50 years ago by statisticians. Um, you can now do them at a scale that you never could historically with the computing power we have. Um, some of them are completely new and you know, people talk a lot about machine learning and AI. Well, there's a lot of the new methods. Um, and the key thing is you kind of try all of them and then you, you back test, you basically say, okay, I'm going to go see which of these using a, a portion of the data that we have historically have been predictive based on the, the whole data set that we have. And you can do that in an automated way. You can do it at scale. And so you can come out with things that give you a default forecast that's actually pretty accurate by the way, than you do in sambal forecasts that combine them together. So there's a lot of different techniques, but what we want to do is you bring all that together and you try to do it in a simple way. Um, my experiences, you know, companies, especially the larger ones, a mess, a lot in forecasting, oftentimes that doesn't necessarily flow down to a lot of the individual work that's happening day to day. Uh, people are kind of running their own forecast in excel, which means they're usually pretty rudimentary and simple. Uh, there's advantages to simple forecast too. Um, but how can we expose and make that more readily available to the kind of the day to day users and consumers of this information? And that's so we really focus on making it, while they're very complex models, making them very simple in the way that they're applied and so people can understand them and feel comfortable using them. And I wanted it to because a, you know, it's, it's, uh, it's an artificial intelligence driven tool. Um, and unfortunately I think it's unfortunately today I as a term is overused and overutilized and sometimes this, I mean, it is too much of a buzzword nowadays. Right? And so I wanted to Tulsa to ask you, I mean, what do you understand that artificial intelligence, that's another thing that sits and she turned the centric and things. Do I have your mission and how exactly are you using that? There'll harnessing the power of, of, of, uh, artificial intelligence in all that. Yeah, so the, the main way that it's used today is through forecasting. So there are a number of different machine learning techniques, um, where, uh, you know, which is a type of artificial intelligence. If you will, um, that we, we utilize. The key thing is we don't just blindly use those techniques because they're the newest and they have the most sophisticated sounding name, right? It's um, it's you, you try those techniques, you also try more traditional statistical techniques. You use business logic and then you try to figure out which ones actually the most productive because you are, I shouldn't care about which method is used, whether it's a neural network or it's an Arima model. Um, what I just care about is what is the most predictive thing for this products, you know, in, you know, this channel or this account or you know, an aggregate. Uh, and that's I think where, you know, alloy uses machine learning and its tool kits. Um, but we don't just blindly use it. We use it and we validated and we say, well, when it's most, we think it's the best technique we should use it. And when it's not, let's use something else. Makes Sense.
Speaker 2:Yeah. Let's talk a little bit, because you've seen a lot and you get to see a lot, because again, you are working with, you know, cross the field of, you know, or they are large companies to smaller startups and so on. People that are just sitting up. What do you see as common categories of inefficiencies that you come across? I mean, like what, what's the, some of the commonalities, and I think you've touched upon some of them, right? It does. The one silo mentality, maybe disparate type of forecasting depending on it can be on regions or products or, or, or so on. But I was just curious, is there some others or where do you see the patterns of inefficiencies?
Speaker 3:So hopefully this will be, um, an answer that suffices. But one thing I would say that we often find is a shortcoming, and I mentioned this a little bit earlier, is there is so much mmm data out there. There's so much additional data that could be brought in, but probably most people aren't taking advantage of that. Um, really a lot of, you know, I'm sure many of the people that are listening, his job is how do I simplify that into, you know, kind of key KPIs that I can track and understand so that I can, I can understand the health of my business. And, uh, that's a very good thing because we do need to simplify. Um, and uh, you, you don't want to make things overly complicated. One of the real strengths though of computers is that they can process a lot of data at once. They're not always great at knowing exactly what to do with that information, but they can process a huge amount of information and they can help you to pinpoint and be flagged where you can drill down to the deeper level from that KPI. So for example, um, I might want to look at something like on time and in full, right, common metric people are going to be looking at a, and I am tracking that at a high level and that number may feel just fine. Um, but an almost invariably we see as you drill in, you start to see patterns of where you're doing a good job. And within that maybe overall acceptable performance, there are re he'll opportunities improvement. Um, and that's where I would say we often see a lot of the inefficiencies. And I think that Aha moment we love here at alloy is, you know, people have kind of talked with us through their business. They might have an inkling that, hey, we think there's room for improvements, but how do you drill down below that high level number and start to see the detail and then drill down again and drill down again. And ideally, how can you make it so you know, the platforms automatically telling you, okay, this is, you know, if you look across that number you're fine. But these are the routes where your Otip is not so good. And there there's dramatic room for improvement and this is how much money is on the line and available if you can make, um, if you can solve this problem. And so that's the, at least one area that we see, you know, Opportunity from once again, the largest, most sophisticated customers out there and companies, you know, down to the very early stage. Now those opportunities generally people have squeezed out the efficiency as you get bigger and bigger. So they're, they're better at it. But the numbers are also so much larger than that. The actual value of the pie is Israeli massive. Um, and so those are the, that would be kind of an area that I would say we see over and over again.
Speaker 2:I've asked, I've asked it for an example, but I was thinking maybe we can take another one if you have certain, um, a certain very data driven, like okay, we took a, maybe you can share with our audience, look, we took this client and on this particular, you know, we applied it to this particular data set or a retail channel or whatever it is and this is how much they saved or this is how much more money they made. Because I am kind of trying to imagine, imagine the, the, it's, it's almost like, you know, I like what they can do. You know, you go to the dentist and it might, it might look when you look in the mirror that your teeth are okay, right? But then it just looks a little bit deeper. Like, Oh, you have a cavity here. Right? Then it should be fixing it because otherwise it's going to spread. Um, so I'd left us to maybe get another example because I think you must have quite a few of them and see if you have some, you know, some, um, maybe some other client that you can share with us and how he did it and how much it kind of meant for their bottom line or for their cost reduction or whatever it was that it was achieved.
Speaker 3:Sure. So let me share an example that probably follows in that last genre that I just shared with you, but it's very specific, right? To make it feel a little more real. Um, so we had a customer who was looking at, um, in stock levels. Again, you'll see that's a common pattern across our customer base. Um, and they, they were looking at overall levels of inventory, um, at, uh, this particular retailer they had underperformed, uh, over an important period for them. And the challenge they were seeing is it seemed as though that it had sufficient inventory. They looked at the amount of inventory in this retailer. It was like, Hey, it appears we have plenty. Why are we seeing out of star[inaudible] being so high and why you were a sales, not what we expected it to be. And the key thing is being able to break that out. And so you want to see every single distribution center, um, uh, that retailer and the fulfillment centers where they're fulfilling ecommerce. Um, how much inventory is in each one and what are all of the stores that they are supplying and what are their inventory levels and which ones are, are running out of stock. And what they were able to see is that there were some dcs that we're running out of inventory and then couldn't replenish the stores. And there were others that had more than enough. Right? So maybe an aggregate, the numbers seemed fine, but you had way too much inventory in certain locations and not enough in others. So those are the types of things where you can break down, you know, okay, I see there's a problem. The aggregate number looks okay cause I have sufficient inventory and ass aggregate, but it hasn't been correctly allocated. So they were able to go ping that retailers say, hey the replenishment and the way that we're allocating product, it's not correct. We need to reallocate in this way. And they were able to drive down those out of stocks and get the sales numbers to where they expected them to be.
Speaker 2:No, I'm, I'm imagining in my mind another analogy with the zoom in zoom out on the Google maps, right. So it's, it's kind of like giving the client that, that possibility to, to really drill down and see, see exactly what's going on. Um, when do you plan to expand next? Uh, you know, you've raised a, you've raised good, quite some good, the funds, um, your present already in three offices or what's next for[inaudible]
Speaker 3:for Ela? So, I mean, we continue to invest really heavily in product development. There are so many no problems and you know, opportunities in this space. There's, there's more data that we feel like we can bring in to, to have even better predictive models. Um, being able to link more and more of the supply chain. We certainly started on the demand side of the equation, uh, but have, have started to work with customers where we're linking their manufacturing plants with raw material suppliers. Right? So we've, we've gone and done that side of things too. So for us, um, you know, you're, you know, investing in, go to market and growing on that side, but, uh, but really the bulk of the investment is in a product development. Um, you know, so that's, that's for us. Um, you know, we're kind of, the bulk of the investment goes and I think one of the really nice things about venture capital is, is you have investors who are aligned with saying there's a huge opportunity, there's many problems to be solved. Um, and, and they have kind of share that longterm vision for what, what can be possible. Um, and so we're, we're doubling down there.
Speaker 2:And I also like to, um, obviously aside from the podcast in the talent and human resources business and, um, I'm also passionate and curious those in terms of talent and building cultures and especially in, in startups and in younger companies like yourselves. Um, and I know that you were sharing at some point that you are, you, you're trying to maintain and you're building a very open, uh, open and transparent culture in the company and that kind of goes with your broader mission of sharing more information across the supply chain. And that's why you put a lot of emphasis in doing that internally. How does that practically translate? Like how do they, how do, how do you practically, they, they do that in our way and how they also maintain a culture like that, an open culture like that?
Speaker 3:Well, it starts, it starts with me, uh, being open. Um, my thing. So for example, we'll have a board meeting and we'll post the, the deck from the board meeting and notes and we share that with the entire company, right? Kind of unfiltered with all the information, you know, the financials, the, you know, the plan, you name it. Uh, and I think, you know, one of the things you have to do as a leader is, you know, trust that, you know, the team will treat that responsibly. It's sensitive information. Um, and people need to have a maturity to understand there's going to be good things that you're talking about. There's going to be things where you have to improve. Um, and so, you know, as you kind of think about the pros and the cons, well, everybody wants that openness, but you also have to have that, especially in a startup where things are changing all the time. Um, and it's, it's a bit of a roller coaster to, to say, hey, we trust you, um, and we're going to share this information and then a pretty unfiltered way. Um, uh, but it's up to kind of each individual to, you know, kind of be able to, you use that effectively and kind of use it and say, okay, I know there's, you know, nothing's being hidden from me. I can be aware of the business goals and where we need to improve and we're all in this together. So I think it's super powerful thing when, when done correctly, but, but you have to have a high level of trust on both sides about what that means when you're sharing all this information.
Speaker 2:And it was single. So in terms of this I found also fascinating in terms of the type of skillset and and kind of background of the people that you have in the company that you don't have a ton of supply chain experts. Uh, we, you do have people that have worked in supply chain, but primarily you have, you have people that are really, really, really amazing technologists that know how to build scalable architecture and very curious and want to learn. Um, I wanted to ask you this question because sometimes there's a very, very fine balance in this. And, and uh, the question, I'll try to frame it like this. How do you make sure that you're still building a solution that is very anchored in clients' problems? And I'll tell you what I mean. Sometimes, you know, you also see that companies go, they start well, but then they start to lose track in some ways. Or if you don't have that good balance between industry knowledge and technology, they then you end up building a solution that doesn't fit the industry anymore. Right? So how do you, how do you maintain that balance between, you know, having great program is great technology people, but at the same time making sure that it addresses and continues to address any holes around the very significant problem that industry faces.
Speaker 3:As you said, it's a fine line and it's one that we're always trying to balance because, uh, it's obviously incredibly important that we understand the problem. Um, and we certainly do have, you know, supply chain people with the supply chain experience. We continue to hire there. Um, but, but we want to make sure we, we can bring the best technologies to bear on the problem as well. I think throughout this, whether you have a tons of industry expertise or you're coming in somewhat naive, is you have to be incredibly focused on the customers, right? That you have. They know their problems better than anyone. And so you know, for us, you know, we learn so much from our customers now we, you know, our goal is to, you know, understand their workflows, understand their pain points, how could you know, what would they like the future to be? And then our job is to kind of come back and take all that knowledge, take the things that we know about the broader industry that we've learned from, you know, our experience in and building alloyed today as well as all the things that we know technology can do. And then to start, you're coming up with ideas and then, and then you go through this iterative kind of testing process where you mock things up. You, you kind of, it is trial and error that the best software is built through trial and error to some degree. It's the super iterative process of saying, okay, well here's a couple of ways we could solve that problem, right? I mean, what one resonates to you? And then you, you know, maybe that starts with just paper mock ups and then you proceed to building a prototype. And you, you get to that point, we were like, okay, we know we're addressing the problem. We have a good vision for what the solution's going to be like and we're going to now fully invested in making it. But regardless of level, extra tease, I think companies succeed or fail based on how connected they are and how focused they are on serving their customers and solving their problems because that's what they're paying you money to do. And so we, we get super laser focused on that and we probably have to do that even more. Um, you know, because, uh, you know, because we're coming at it with that kind of strong technology background. So, uh, so that's, that's how we tackle that problem where we focus[inaudible]
Speaker 2:how about hiring? So, you know, that's critical, you know, attracting, attracting new talent, attracting treasure, bland and making sure that you get the brightest and the smartest and the, obviously you started in silicon valley where basically the fight or whatever you want to call it, this is real, right? It's, it's a jungle out there in terms of, uh, of finding and, and getting the right people because there's so many options for them. There's so many startups. Everybody's doing interesting things. How do you, I mean, what's your, some of your current priorities, how do you think about hiring? What's your challenges around hiring? You know, share a little bit about, about that with us?
Speaker 3:Sure. So I think the, the priorities in hiring, I mean, you outlined some of the things that we're really looking for. We look for people that are intensely curious. Um, and I think especially when you're at a startup and you're growing extremely quickly, I mean, our headcount as think tripled, you know, in the last year or so, um, you have to have people that are going to take a lot of ownership and you know, are going to be doing a lot of things, uh, that are going to be new for them, uh, and, and are going to figure it out, right? They're gonna do their best. Nobody does. Perfect. One of our values is iterate to excellence. This idea of like, just start and get going. And the idea is that you're going to get better and better, I mean, the same way that you build products. Um, so you know, our priorities when we're looking for people is those that embrace the stage that we're at like this very fast growing, lots of opportunities to learn and grow quickly. Um, but you have to be able to operate in an environment where there's less predictability than there is going to be at a, at a much larger company. I think in terms of, you know, the, um, the, the challenges, you know, it varies a little bit across our different offices. Uh, and as you said, Silicon Valley is certainly, uh, you know, there's kind of a dog fight here for talent. One of the things we always have to focus on is how we differentiate. And the Nice thing is we're tackling a problem that not a lot of companies in silicon valley are tackling. There's a lot of education that goes on around this is this incredibly huge space. Uh, and there's massive opportunity. Uh, and it's not that there's 10 startups on our street that are all trying to do the same problem, uh, but you have to, you know, help people get excited about it, see the vision, uh, understand the opportunity because you know, there's always other flashier consumer focus things that they understand probably more intuitively. And so that's something you know, we have to kind of continually do is just help them see it. I think you're already seeing this by the way in and supply chain tech, uh, particularly in logistics where there's a lot of investment right now where people are saying, Whoa, there's, there's this, you know, quote unquote unsexy spaced where there's just really interesting technical problems and a lot of data. And that's the kind of stuff that, you know, engineers in particular get really excited about. They just need to know that it's there. Oh, it was, I was looking at the chart, I was looking at the charge recently that kind of outlined the venture capital and private equity investments in, in the space where they did display. And then logistics and the, I think it tripled, uh, from, from last year. So in 2018 it was three times more than 2017. So it's definitely very, very clear upward trend. And also we've seen, and this is not including some of the very large, I mean we've seen some very large, uh, investments of Flexport is probably the, the, the very large one that was just funded billion from, from Softbank. Right? So there's, there's definitely more and more interest
Speaker 2:is more and more, um, money in energy and, and funds being pumped in. It's also a ripe for disruption, right? I mean, there's a lot of ways in which the whole system or supply chain can be done better. And then it's great that people like yourself and other startups that are looking at doing it and yeah. Uh, when it comes to technology people that as you said, you need to sell them the story, the data, but you know, it's that, that's key anyway. Right? So, um, and I wanted also flip the question on the other side because that's also useful for people that are listening to us, for maybe other founders or people that are thinking of setting up. Um, and I are in people. What are some of the mistakes that you made when it came to hiring so far
Speaker 3:where it didn't work out? Yeah, so I think the biggest thing, um, that you see is, um, you know, having kind of, he has been here from day one, different people work at different stages. Uh, and I think this is something, as you're starting a company, you're always trying to find those people who thrive at the stage that you're at and, and hopefully in the stages to come. Um, but I would say if I look at either the mistakes in hiring, it's oftentimes people that are extremely talented, um, and had been very successful often that, you know, kind of the big brand name companies, but have difficulty transitioning to a company where there's much less infrastructure. Uh, and, and so I think that's something I focus a lot. You have to be, it's a balance right in, in the ing process and those you're interviewing, but trying to say, hey, is this the type of environment and the type of size where you might be coming in and being a, a team of one, you'd have to figure everything out on your own and then you're going to set that foundation that is going to support other people who are going to come on a, but you have to do that. And that's something a lot of people haven't done. Um, and it can be really scary and, and it doesn't work for everyone. So I would say by kind of, you know, evaluate the times that it hasn't worked as well. It's oftentimes come back to just, it wasn't the, we weren't the right size, um, for, for the person that joined, even though in many cases I'm sure they go on and, and I've had a lot of success in other places.
Speaker 2:This is good. This is cliff. Come on. And then a lot of people think, and I've seen it a lot and we see the low to low. I mean, and also it has become somewhat, somewhat interesting or sexy for some reason that it's to be, to be part of a startup. Right? But in actual fact, it's hard than the in actual fact. I mean, it's hard with one thing. It's also very interesting, but you gotta have to, you got to have the flexibility to adapt. And sometimes if you've been in the corporate life, all the, all, all your career is not that straightforward to two heads. And then you're used to a certain system and a certain order in a certain structure that startups are much more wild and dynamic
Speaker 3:and
Speaker 2:things, a rollercoaster, I think is the word that you've used and is probably the closest to do what it is. Right? So you gotta you gotta have the number for that, uh, which is very, very tough actually to kind of, to kind of fought in an interview a lot of times. So that's, that's another, that's another thing. That's, that's why also, I mean, I'm trying, I'm trying very hard as a head hunter, a lot of times, even for ourselves, we try to also deploy a psychometric tools. And then there's, there's also, um, uh, all sorts of predictive, uh, analytics on, on human psyche can personality. But ultimately, I'm not sure there's a silver bullet in this. So, um,
Speaker 3:yeah, if you found one, I, I'm, I'm all ears because that, that is it. I agreed to really hard thing to gauge and I also will kind of double down on your point of, you know, and I've seen this in the last seven or eight years, startups have, it's definitely become, you know, people get excited about this idea of working at a startup. And I don't think they always fully recognize what that means. There are pros and cons and until you're trying to educate your, obviously anytime you're recruiting, you're selling, you're trying to get them excited about the problem you're tackling while also being realistic to say, Hey, you've been working at Facebook, it's very different than we are. Um, and you've had a level of board that you're not going to get here. And that is invigorating for some people and it's terrifying for others. And you know, people can change, but, but it's uh, yeah, it's an, it's an interesting balance and I'm sure you deal with a lot in your role too.
Speaker 2:Yes. Um, what do you see in terms of skill? So obviously for, for you in terms of hiring is very much a, I'm going to be kind of keeping this balance and attracting very smart technology people, but also that, that have a strong customer centricity and understand the supply chain, but from a supply chain, from your client's perspective. I wanted to flip it a little bit and ask you from, from your clients, from the manufacturers, from the consumer goods, from the people working in the supply chain of these companies, where do you see the type of skills and people that they will need to be hiring more and more also for them to best utilize your tool? And of course is going to be, it's, it's intuitive and it's simple to use, but there's also a level of, um, I don't know, strategic thinking, be able to analyze data and all of that. So do you see this as there's also a shift in terms of your clients and what type of people they they get in their supply chains?
Speaker 3:Yeah, I think this is a really important point. I don't think this is anything groundbreaking that I'm going to say here, but you compare the arrows of the world. Eero is a customer of ours. They make a wifi routers. They were recently purchased just I think a month or so ago by Amazon and uh, you know, they were founded just a couple blocks away from where we're based. You compare them with, um, you know, the, once again, those kinds of big older CPG companies and the skill set, the way that they, they think about data is just so different. I mean these, these new digitally native kind of consumer brands, it is just, you know, from the get go it's, everything is going to be data driven. I mean, they often come to us because they're just shocked that this is the way the industry operates. They're like, I had no idea when I started working with Walmart. This is the way that it works. When he started selling through retail. Uh, and help me figure this out because I, I just can't imagine operating company where I don't have a pulse and the way that I have historically. Um, and obviously every company is going through digital transformation. Every company understands the importance of you know, analytics and um, you know, using data to drive more decision makings that that's nothing new and been been discussed. But you see such a dramatic difference from the companies that are being founded. I think more recently on the whole, um, with those that, uh, have, uh, uh, workflows, workforce that has an incredible amount of knowledge, um, but hasn't necessarily operated exactly in that way. And I think that trend's just going to accelerate. Um, and you're going to see that, you know, you're just going to see a much more um, datadriven um, uh, your data mindful, you know, kind of set of people. That doesn't mean everybody needs to be a statistician at all. It just needs to be like, hey, we make, you know, a lot of decisions based on, on, you know, the data that we have. We combine that with the intuition and all the smart things that humans can do. I mean, the wonderful thing in all of this is we're going to have a place in this future world, by the way. I mean, I certainly am confident that I'm confident in and people in our brains and our ability to do a lot of things that computers are not good at. Um, but we need to be super comfortable with offloading a lot of that work, two computers and so we can augment them and do the higher value stuff. And, and I think you're going to see the most successful companies are adopting to that. You've already seen it happen in technology and now it's happening across a lot of other industries and supply chain certainly going to be the same way.
Speaker 2:Hmm. I mean that, that broke to my mind. I won't then the company, but it's a major consumer, uh, in appliances manufacturer. And they basically, I was talking to one of the c level executives globally and he was saying that a, that they bought, they had bought a startup that was making some kitchen appliance. Um, and he was, I can't remember what it was, but anyways, it was very specific, but they had tremendous, they had grown tremendously. So I think they bought them within three to five years of their kind of company life. And that, that, um, that manufacturing startup managed to grow to a global scale and distribute that particular piece of equipment for the kitchen, uh, with, I think it was 50 or 60 people with almost everything, outsource source. So they were doing, you know, contract manufacturing and almost everything was outsourced. Um, and they were basically just focusing on a very, very, I think the marketing was something that they were super, super strong at and they had some Paytons in terms of the product, but that was it. So 50, 60 people, they went three to five years, I can't remember exactly. They went global and they basically got this, this top, uh, top global brand to buy them. And the company was bought them because they wanted to learn from them how they could transform and integrate some of the principles that they were using to transform the bigger company in itself. So that's, I found that fascinating and it kind of goes well with your example as well. Um, with a, with a, with a company that you've just mentioned. So I could not agree more. Uh, we are this digital transformation. It's a password. It's, you know, it's been going on forever, but it's really happening, right? It's, it's how can you be faster, more agile and more responsive to the market and it's, it's, uh, or if you don't, then, well, you'll have some serious problems. Um, final question from what, from our, from our discussion, what Joel, what's the best piece of advice that you've received throughout your entrepreneurial journey? Right. So entrepreneurship is hard. It's exciting, is a rollercoaster, is, you know, is a tough, and, and at the same time a wonderful experience, but what some of the best pieces of advice and sharing you receive that could help other people that I think,
Speaker 3:oh, it or on this journey. Yeah. So, um, I was actually thinking about this this weekend randomly, and, uh, I'll, I'll share some advice that I received. It actually has nothing to do with entrepreneurship, but I think applies. Um, the advice was actually given by my oldest sister, uh, before I had children. I have twin boys who are now six years old. Oh. Also, I was about to be, I guess this was probably six or seven years ago. Uh, you know, it was kind of, uh, nervous expectant father, uh, and was chatting with my sister who had a couple children and she said, look, you know, 90% of the time kids are just a lot of work and you're probably going to wonder why you did this is going to be 10% of the time when it's just amazing. It's just, it's transformational. You have these moments that are, that make that other 90% feel like nothing. Uh, and I was, I don't know why I was thinking about that this weekend and I was like, you know what, that kind of sounds like being an entrepreneur. Um, it's hard. Uh, it is, we've talked down multiple times about it being a rollercoaster. Um, you know, things can look externally, like they're going amazing and then[inaudible] you feel like there's, you know, it's chaos and, and um, you know, so you know, a lot of it, you know, and I think any entrepreneur, uh, will say this, I mean, you're just, you know, kind of slogging out in some sense day to day. Uh, but the amazing, you have those 10% of times, right, where you look back and you see, you know, you hear those amazing customer stories where a customer, you know, CEO texts, it was like, oh my gosh, like we, you've changed the way that we operate. I got that text a couple of months ago. And, um, and that's just so invigorating and it makes you feel like you're doing something valuable. Um, and it makes the whole thing worth it 10 times over. So for me that's a way that, you know, you kind of, um, sometimes you know, deal with, with the day to day and the struggles that everybody goes through and any job or any assets aspect of life, but you're saying, hey, there parts of this that make this more than worth it. And those, those kind of tide me over for, for the, the tough periods that come as well.
Speaker 2:Lovely, excellent, excellent. Uh, sharing. Um, I never quite quite thought about it from that perspective, but yeah, very good. Then I have two kids of my own and probably 95% sometimes with kids that are like, oh my God, what did I do? And my sister probably sandbag the numbers to make it seem more palatable. But yes. Oh it's, yeah, but it's um, I don't know if we even get Glen to fight kids, but it's, it's, yeah, it's very similar with the, is very similar with start ups. And that reminds me of honey, the funny sharing or other, cause they asked one another entrepreneur about this and he's, he's actually from Asia and he studied the last Mile Company in Asia. And his, his, uh, answer to this question, you know, what was the, what's your advice? And then he was saying, don't get married, don't get married, don't have kids. You don't have any time for that. So I like you're a, you're sharing because I think we bring balance. I think you can make it no matter the stage at the time of your lives and your lives with it's, it's definitely, you should not, nobody should think it's a, it's a ride in the park. That is for sure not. Um,
Speaker 1:Joel, it's been a pleasure. Thank you so much for, for joining us for the, a great example is in sharings and then real case studies and the, I want to wish a wish you and the other team all the best and um, and keep growing. Uh, keep, uh, keep expanding and um, you know, keep getting us to read about you in that
Speaker 2:news. Well, thank you. I really appreciate you having me on.
Speaker 4:Thank you for listening to a podcast. If you liked what you heard, be sure to follow us on[inaudible] dot com slash podcast for all the show notes, links. And extra tips covered in the interview. Make sure also to subscribe to our mailing list to get the news in the nick of time. If you're listening through a streaming platform like iTunes or stitcher and you like what we do, please kindly review and give us five stars so we can keep the energy flowing. You'd get more people to find out about our podcast. I'm most active on Linkedin, so do feel free to follow me to stay tuned for our latest articles as well as future guests for the podcast. And if you have any suggestions or any other idea, please feel free to write to me. I respond to all and also please make sure not to miss our next episode where we will be having a few other c level and top leaders in supply chain joining us. Stay tuned.