Latest / Exit Algorithms / Warehouse Automation for Business Owners: What to Automate and When with Dan Cahalan (#70)
Transcript
- Pete Vera, Exit Algorithms: Welcome to exit algorithms, the podcast where we decode what it really takes to unlock growth, streamline operations, and prepare your business for a high value exit. I'm your host, Pete Vera. And today I'm joined by Dan Kat Kahalan, sales director at Swiss log, one of the world's leading warehouse automation companies. Swiss log operates in over 50 countries and is behind some of the most sophisticated automated distribution systems running today. Dan, super excited to have you here. Welcome to the podcast. Dan Cahalan: Thank you, Pete. It's great to be here. I will say you had a mustache at MODX, which is why I shaved my beard into a mustache so we could like do double barrel mustaches here. But I understand with the wedding coming up, you don't want the mustache tan line. So it's all right. I will immediately go shave it in the bathroom after this, but we'll let it Pete Vera, Exit Algorithms: Yes, sir, maybe we'll have to do a follow-on one day and I'll grow it back Dan Cahalan: Yeah, please. I was really hoping we could do like a Mario and Luigi looking like, you know, side panel here with the red mustache and the brown one. But yeah, thank you so much for having me. It's a pleasure to be here. Pete Vera, Exit Algorithms: Absolutely. Yeah. Do you mind starting off? Can you share a bit about your background and career journey? Yeah. What led you to Swiss Long? Dan Cahalan: Yes. Yeah, absolutely. I'm not sure where the camera can see back here, but you'll see a Purdue helmet back here somewhere. But so I grew up in Ohio. I went to Purdue for industrial engineering. I actually started in aeronautical engineering and then decided that. the like the grain size of what you study and can control in aeronautical was a little too fine for me. And I really wanted to kind of zoom out and have a more systems perspective. So the degree is actually now called industrial and systems engineering. At the time that I went, it was just industrial engineering. But yeah, I went there. Over that time, I interned for for GE in South Carolina, which was your neck of the woods, interned in fastener manufacturing for a summer and finally ended up interning with Cole's department stores, which is For those of you unaware, that's retailer headquartered in Milwaukee that has about 1,150 stores nationally in the US. we're in turn with them, worked for Kohl's for about six years as an industrial engineer, as a project manager for deploying automated systems that we were purchasing turnkey automated systems from others. So project managed and did some design, a lot of design for those facilities for primarily Kohl's e-comm fulfillment. a team of industrial engineers doing similar things as well. So, spent some time there and then kind of I realized that what I really, really loved to do was... learn a business extremely well, in this case it was Kohl's, and then design solutions that fit that business's needs, and then kind of support that long term, learn how the system behaves over time, try to squeeze the most out of it. So I learned that I love to do that, and my experience at Kohl's was tremendous, but we were only doing that for a building every maybe two years, and I really wanted to kind of increase the frequency at which I was doing that process of really thoroughly understanding a business, designing a automated solution for it, deploy that automated solution and watch it show up in 3D, which is like the most unbelievable experience is when you get to design this like beautiful automated system and then watch it run, there's nothing better. So anyways, I moved over to the integrator side to be able to do that on behalf of companies like Kohl's. know, when I was at Kohl's, I would go to an integrator and say, this is approximately what I have in mind. Here's my data. Can you help me design the system and deploy it? So I kind of had an understanding for how that already works. So I moved over to the integrator side. Pete Vera, Exit Algorithms: Hmm. Dan Cahalan: to do the receiving end of that. I worked for Bastion Solutions for four years and was really fortunate to be able to sell pretty much the whole breadth of what Bastion had to offer and really, really loved and did get that experience of, instead of working on one solution every two years, I was working on five solutions at a time for four years. So got to do that at Bastion for four years, which was great. It was an excellent experience there. About a year ago, I transitioned over to Swisslog to approach the industry from a different angle, from a different company with different primary verticals, different value proposition and different software. So really have loved my time at Swisslog. I'm doing a similar role to what I did at Bastion and my role now is titled sales director and effectively I'm responsible for the commercial and technical offering to a customer for an integrated system. you know, from the point of first engagement of I always make I call it Bob's Boot Mart, right? Like Bob's Boot Mart says, hey, we have a warehouse we think we want to automate. They engage us, I'm immediately involved. And I work with them through data analysis, visits, concepting, pricing, scheduling. And then eventually when the project sells, I'll hand it off and it will be executed by a really talented execution team at Swiss Log. So anyways, that was my journey from basically back at Purdue to now where I sit at Swiss Log. Pete Vera, Exit Algorithms: Awesome. Yeah, I love that story. You know, I'm sure it helps tremendously. It's probably a requirement to have that engineering background or at least understand the technical aspects of the job well, right? Dan Cahalan: Yeah, it's pretty helpful, I think, ⁓ to be honest, and I'm sure I'll mention this later, but I think like the best thing that anybody that operates a warehouse can do is hire an industrial engineer, someone that's dedicated to, you know, observe the operation. ⁓ recommend things that can change to make it more efficient. They pay for themselves in like a month, typically, you know, the first month of just improvements of like, why are we doing it this way? So I think, getting that being an engineer is really helpful, but ultimately, I think having kind of a systems mindset, which is a more like qualitative feature, I think is really important. ⁓ And actually, I will say that having worked on the customer side at Kohl's really prepared me well for doing the integrator side design because, you know, like I've lived at a building and supported an operation and felt the pain of having the building director call me at two o'clock in the morning and be like, what the hell dude, the print supplies are down or whatever. You know, like I think that that experience of understanding like, you know, transparently the pain that you go through in deploying automation ⁓ or not even just automation of deploying warehouse management systems or new processes I think is really helpful in how to design them to be robust for the future. ⁓ But yeah, I'm super, super fortunate to have I this industry that I absolutely love. Like it does not feel like work a lot of the times to get to go see a site, recommend improvements, envision with a customer how their solution can change or how their operating model can change. And yeah, it's just, it's a joy. So, so excited to be able to talk about it. Thanks for having me. Pete Vera, Exit Algorithms: Definitely, yeah. No, that was actually going to be one of my questions. Does it help having that customer experience first, you know, before entering sales? Are there any lessons you learned from your time at Kohl's? Dan Cahalan: Yeah, it's how much time you have. Yeah, I think it's a... I have found it to be critically important because I mean, for the reasons that I just mentioned of being able to put yourself in the shoes of an operator. But I think there's a lot of, you know, softer stuff that can't necessarily be trained when it comes to designing automation solutions that you kind of just have to look and feel and see of, you know, the way that a very small change to how a user's workstation is set up can have drastic performance, ergonomic sustainability impacts. ⁓ Like maintenance access is something that I think is a big difference that I see between those that have experience in warehouses on the customer side to those that don't. It's like, have you designed a way for someone to change this motor out on this piece of conveyor? So yeah, I think that it's been extremely helpful to have had that experience. I think that there are a lot of people around the industry that, on the integrator side that have that. And then I think the folks that don't, ⁓ the most successful ones are the ones that are able to really empathize. empathize with the folks that are running operations, those that intentionally get operations involved ahead of time and make sure that what they believe and prioritize is ⁓ considered in the design. So I think it's really helpful. It's not necessarily required, but yeah, think it really, it all comes down to how well you are listening and empathizing with and understanding an operation. So. Pete Vera, Exit Algorithms: Yeah, it makes a ton of sense. Yeah, I'm curious, what does a typical engagement look like with Swiss log? ⁓ And you know, is there a certain size of business where it makes more sense or yeah, to just kind of get an overview of the type of customers you work with. Dan Cahalan: Yeah, so Swiss log specifically, and I know we're talking more generally, so I'll try to keep the sales pitch part to a minimum here. But yeah, think Swiss log, and the reason I'll answer it this way is because each person, provider, integrator, OEM. industry ⁓ kind of approaches sale differently. They have different verticals that they're really strong in different technologies they offer ways they go to market. So ⁓ Swisslog specifically is a master solution integrator, which is we manufacture some of our own components, but ultimately we provide an integrated system of things that we manufacture and things that we procure. So we manufacture a really effective unit load crane and pallet conveyor, but then we also procure and integrate with our controls and software, AutoStore, EuroFork, FourWay Shuttle, Conveyor, Robotics. So I think there are people kind of in that cohort of integrator, like Bastion, I used to work for, example, Ford and others. And then there's like, ⁓ on the OEM side, which are people that sell primarily what they manufacture. I'm sure you can think or name a few of those. And Swisslog exists kind of somewhere in the middle, which is ⁓ OEM backbone of things we manufacture, but also things that we integrate like AutoStore. ⁓ And just to be fair to like fashion manufacturers, Great Conveyor and Robotics, for example. So I think everybody approaches the industry a little bit ⁓ differently. For Swisslog, think what ⁓ because we are providing an overall solution that is tied together with our software, which is called Sync, ⁓ a lot of our engagements oftentimes start with customers who... ⁓ who have some data about their operation. They have some vision of what they want to do in the future. And I think they're really looking for the integrator to kind of guide them towards, we've evaluated all of what you want to do, what your particular order profile and customer base looks like. And ⁓ here's a couple of potential solution options. yeah, think a lot of, and it really depends, ⁓ the way that we engage with the customer depends a lot on Pete Vera, Exit Algorithms: Mm. Dan Cahalan: ⁓ I think their own experiences in general with automation. like when Kohl's would approach an integrator having worked for Kohl's, know, like they have huge experience with automation at a lot of different levels, right? So as opposed to my usual example, like Bob's Boot Mart that may not have any automation, maybe approaching it for the first time. So the amount of, you know, mutual education we have to do or site visits we have to do or It really oftentimes depends. think it also often depends on whether that organization has their own engineering infrastructure. like when I was back when I was on the customer side, you know, would lightly, lightly engineer a solution. I'd know approximately what I wanted and I'd come and say, here's approximately what I want, but I'd like for you to beat it up and. Pete Vera, Exit Algorithms: Mm. Dan Cahalan: make it really come to life in 3D. And then I would say there's people on the complete other end of the spectrum that come and say like, here is exactly what I want. I've already drawn it. Please sketch this for me and tell me how much it's going to cost. So I think we, as integrators, whether it's Whistlog or Bashin or whoever, you're provided, you kind of have to meet a customer where they're at. And ultimately you want to build a relationship. Pete Vera, Exit Algorithms: Mm. Dan Cahalan: understand what they are really seeking to accomplish in order to be an effective advisor for where they can go. ⁓ For those that are not really involved in this industry much, ⁓ the thing that is incredible about it is that the most successful solutions, relationships are the ones that are very openly collaborative. That are the ones where you take the time to really work together to concept and reiterate And it kind of takes a mutual investment, is the integrators will kind of put up a lot of reiterative design work, which frankly costs a lot of money on our side to have resources do that. But then also on the customer side, there's transparency about, you know, how the bid is progressing or how they're like where the price point needs to be. So I think that the most successful implementations are those that are the most collaborative. So I kind of got off the path of your question a little bit, but I'll just say like it's really an it depends thing on. the customer's experience, how transparent, collaborative it is as opposed to transactional. And I think lastly, it often depends on what the customer is seeking to accomplish. Like some people are saying, hey, I want to understand if automation's even in the future for me. So like, can you tell me if this thing that I want is five million or 50 million or $500 million or $500,000 for that matter? That's what I'd call like a ROM level of detail where we'll do some concepting. a little bit of light data, we'll make a lot of assumptions and then come up with something that they can at least inform, are we in the right ballpark? And then there's another level of fidelity called budget, which is another step further, which would be, Hey, we did more data analysis. We visited your site. visited ours. We've kind of decided on a concept. Here's approximately what it's going to cost based on the real recent costs coming for this type of equipment. And then there's a firm engagement, which would be like an RFP where someone says we already know We are going to do a building. We already have a budget approved for the most part. We just need a price that is completely accurate so we can decide between two or three vendors, put one on contract and run. And so the higher the level of fidelity, the more intense the effort is. And to be honest, the more detailed the data analysis needs to be, the more, ⁓ you know, the more ⁓ scrutinized the documentation needs to be, et cetera. So I think this is all like an, depends answer to your question, but I think at its core, the best. Pete Vera, Exit Algorithms: Hmm. Dan Cahalan: thing is that the typical engagement starts with openness. It starts with a willingness to be transparent about here's the data that here's me as the customer this is the data that I have and the integrator saying hey with that data we can do X, Y, and Z but we're gonna need to make assumptions on AVNC are you good with this? So like I think that the most open and transparent in time both sides can dedicate to each other the better an engagement goes in any case. So long answer to your question sorry about that. Pete Vera, Exit Algorithms: Yeah. No, yeah, no, I appreciate all the context. ⁓ Yeah, I think one of the takeaways I got at MODEX where we met was it's not like automation isn't necessarily just for Amazon scale operations. ⁓ I'm curious from your perspective, is there like a certain metric you look at, whether it's revenue, whether it's square footage of the warehouse? ⁓ Dan Cahalan: Bye. Yeah. Pete Vera, Exit Algorithms: to where it makes sense to start automating, or would you say, you know, start automating from day one in your operation? Dan Cahalan: You know, you, sorry, you asked this question, a part of this question is part of your previous question. It's like, who is automation really for? Right. It is the question. I love it. think it's great. Take my answer with a grain of salt because I sell automation, right? But like, I really think that automation is for everyone. Like at some level, there is something that you can do to improve your operation. I said at the top, like hire an industrial engineer. That's not necessarily automation, but Pete Vera, Exit Algorithms: Yeah. Dan Cahalan: there's process improvement. I think that in terms of actual mechanical automation, there's something for everybody. You could buy a five to $20,000 AGV to do something, or you can buy a $50 million rack-supported building, Pallet ASRS. So I think that any business... can probably find something where automation is justified in some way. I think oftentimes the mismatch is between the vision for automation and what is actually justified. like the... Not everyone is going to have justification for a $10 million auto store, right? Like you're gonna, you may have justification for a, you know, a lighter duty, sold as a service, like AMR style picking robot that you follow behind and pick into, right? So I think there's some level of automation that makes sense for most, not all operations. The things that I really look for is not necessarily square footage or company revenue. I think it's more, ⁓ I start with what are the problems that you're seeking to solve? What are your goals? And I think if you're really looking for like who is a candidate for automation and who is not, it's more of a question of I would ask, like show me what you're seeking to accomplish. And I think that in terms of like justification, the easy ones, and I mentioned this when we went in the panel in MODX, but I think like there are what I would call like the frequent flyer. Pete Vera, Exit Algorithms: Mm. Dan Cahalan: justification methods, right? Like removal of F, for those of you that don't know that term, that's full time equivalent. So like you have... ⁓ Pete Vera, Exit Algorithms: Yeah, what are those? Dan Cahalan: That's headcount reduction transparently or if you're growing it's it's preventing needing to increase headcount to accommodate for growth So you're solving I mean simple ROI based on labor There's labor availability. That's also an issue. That's like again. These are in my frequent flyer categories There's also, you know Accommodating for growth there's storage density and then I think there's more abstract things that you could evaluate like saving Pete Vera, Exit Algorithms: Mm-hmm. Dan Cahalan: on cost of quality to customer, where you may have a big issue with errors to customer. Like oftentimes automation is a way to reduce errors and can measurably reduce errors. It's an ergonomic issue where, you know, if you're picking TVs or refrigerators off one pallet onto another, like automation may be justified on the basis of reducing ergonomic strain on individuals. There's also, you know, kind of safety, ⁓ Pete Vera, Exit Algorithms: Hmm. Dan Cahalan: entitlement so to speak which is like you're running into rack every 10 days and you have to replace rack like what's the cost of the rack you're replacing in the labor in the downtime so so I think really to come back to the core the question is who is automation for it is for anyone really that can think of a collection of things that could justify automation that is willing to to kind of accept the automation that they are a fit for does it does that make sense Pete Vera, Exit Algorithms: Yeah, yeah, think definitely. So starting out from the goal that you're trying to achieve, do you see any any mistakes companies make when trying to trying to automate, you know, an existing facility? Dan Cahalan: Mm-hmm. Yep. ⁓ Yeah, I definitely, definitely. ⁓ think that I can talk about this in its own hour long podcast episode, to be honest. But I mean, I think that mistakes get made by everyone in these situations, small mistakes, the severity and impact of them vary, of course. But yeah, I think that people oftentimes... ⁓ I mentioned this in the panel, but I think a big mistake that people make is thinking that there will be a silver bullet ⁓ solution to their entire problem. So I call this the silver bullet fallacy, which is like, hey, I have this big warehouse with all these different sizes of boxes and units and palettes and whatever else. I need a solution that solves all of this. Typically, the solution that solves 100 % of your problems is going to be five times the solution that solves 90 % of your problems. So I think a big mistake that people make is not saying like, okay, it is still acceptable to take the bigger, weirder freight that's more difficult to automate and still handle that manually. I think a really good example of this is like, let's say you're an automaker that has spare parts, know, like handling nuts, bolts, cables, spark plugs, like super easy to automate that for spare parts distribution. The hood of a car, extremely difficult. So if you go look at a Pete Vera, Exit Algorithms: Mm. Dan Cahalan: building that's a parts distribution center, for example, and you say, I want to automate this building, automating 90 % of the throughput, which is the small parts, let's just throw a fake number at there. $20 million, right? Automating all of the big parts for 10 % of the throughput, $50 million. You know, so I think it's a, it's a really, the mistake people make is thinking that they have to solve the whole problem all at once. And I know I'm throwing out big numbers, but I just want to give the sake of scale. Pete Vera, Exit Algorithms: Yeah. Dan Cahalan: there. So I think that's a big issue. think another one that happens kind of on a more ⁓ like a deployment basis is not involving operations early enough. ⁓ So like I'm sure anyone who listens to this has an operations team, maybe part of an operations team. I think that a huge mistake people make is not bringing operations in to weigh in on the requirements that the system has to accomplish, that an automated system has to accomplish, because you'll miss critical ⁓ critical components, right? Like the operations team might know that you need to scan the lot number when you're picking an item, for example. The difference between picking up an item and putting it in a box or picking it up and scanning a lot number could be, you know, half the user could be doing half the performance. And at that point, you've drastically under designed your system. So so I think getting operations involved early to not only weigh in on how the system needs to be designed, but then to also really have some ownership and buy in as Pete Vera, Exit Algorithms: Okay. Dan Cahalan: what they're eventually going to need to end up running, I think is really an important thing to do that a lot of people kind of fail to do. ⁓ And then I think ⁓ another one is that I think there's this, ⁓ especially as you start to get into bigger automated solutions, there's this perception, it's an easy perception that if let's say I have a task that a human is doing and I would like to automate that task, I think oftentimes the thought is like that the automation has to be, automation has to be human shaped, you know, like a robotic arm or an AGV that travels where a forklift otherwise would have traveled. And I think, Pete Vera, Exit Algorithms: Mm. Dan Cahalan: more fundamental, taking a step back and thinking more fundamentally about how could I design a different model of processing is a mistake that people make a lot. And then the last one I would say is that, just speaking of the human, like humans are incredibly valuable, dexterous, wonderful machines. Like the amount of time since robotic arms have come out where humans are still picking shows that human dexterity is something incredible, right? So I think the other mistake that people make is they really get attracted to this idea of a lights out warehouse, you know? And I think ultimately, humans are so valuable at solving many, problems that... Pete Vera, Exit Algorithms: Hmm. Dan Cahalan: considering that there is a failure to consider that humans can be a really important part of an automated ecosystem, of an automated systems ecosystem. So yeah, I think that's another big mistake that I see is not only thinking that you could replace a ⁓ working position with something that's shaped like a human, but also that humans have no place in the warehouse because that's simply not true. Pete Vera, Exit Algorithms: Hmm. Dan Cahalan: most automated warehouses that I've ever been in solving incredible problems can still have dozens, hundreds of people there that are doing tasks that humans are so uniquely qualified for. ⁓ So anyways, that's just my thought after having been in many, many, many, warehouses. Pete Vera, Exit Algorithms: Yeah, no, I appreciate it. Yeah, very detailed answer. A lot of examples. Yeah, that that concept of lights out, I think not maybe not all of our listeners are familiar with, but it can you describe that a little more? And does it really exist? Dan Cahalan: Sure. Yeah, I mean, it's a good question. Lights out is like, you know. If you imagine, I'm trying to think of a good example from like from a movie or something, but if you like close your eyes and imagine like robotic arms attaching doors to cars where there's no people that you can see, right? Like where where everything is done in an automated fashion, that is what people call lights out, which the term is just like the robots don't need lights to work, but the people do. So that's where the term comes from. ⁓ And so, yeah, I think that there are warehouses that have trended that direction. Certainly, I'm sure someone watching this be like, well, I went in warehouse that blanked such and such a thing didn't have any operators, which might be true. think like if you have manufacturing style tasks that are really consistent, don't require decision-making, do not have a lot of permutations in how they operate like that. that may, there's probably warehouses, manufacturing lines that are close to lights out. I think that a lot of the things that Swisslog, Bastion, Domatic, others in our, ⁓ in kind of the material handling industry interact with, a lot of those. warehouses, whether they're manufacturing or otherwise, the whole point is to solve for variability, right? So if you're interested, I can expand on this a little bit, but I think like, yes, there are probably lights out facilities around. I'm sure there are. But I think the more variability there is in terms of size of product, what you have to do with it, how it is ordered, the harder and harder that becomes. Do you want me to expand on an example for this or no? I'm happy to. Yeah, like, so consider like an automotive manufacturing example, right? Like, Pete Vera, Exit Algorithms: Yeah, sure. I'm curious. Dan Cahalan: When you're manufacturing a car, seven, you, I don't remember the exact number, but it's like every 50 seconds, a car runs off the line, right? And there's a lot of like, bring the door, the door can attach. It's for a Silverado, it's going to be largely the same every time. Maybe the color changes, maybe the trim changes, whatever. But like, it's all what I call linear. Like it's all moving in one direction. And a lot of like warehouses operate in what I would call like non-linear fashion, which is like, let's say you have a hundred thousand skews. Skew being unique items. So like this shirt in this size and this color is a skew. The other eight sizes are different skews and the other eight colors are different skews. So this shirt could have 64 skews right in that case. So consider a retailer that has 600,000 skews right. Anybody goes online and can purchase any combination of those skews and they'll need to be picked and packed in a certain way. Maybe there's a gift wrap, maybe there's a label, maybe there's a sample. So there's all of this variability that goes into combining this order into a box to ship to you that is not necessarily linear because you're have to pull it out put it into a storage position then pick it then pack it maybe there's a wrong you put it back maybe there's a cancel there's just so much uncertainty is basically the entire point of the automation is to weather uncertainty so I think in situations like that where there's quite a bit of variability you You oftentimes need humans to solve the dexterity as opposed to automation where it's like move this thing from here to here over and over and over again. That kind of stuff has been automated forever, right? Like you can go on how it's made and watch chips get made. And like the people are basically there to maintain the machine. ⁓ So it's a, I think that in warehousing specifically, maybe in 50 years or 25 years or 10 years, who knows, there's drastic changes in that regard. But I think ⁓ in many, warehouses humans are still going to be involved. And so part of our goal is to design a user experience for those that are involved to be as efficient, as ergonomically comfortable, and as safe as possible. anyways, I can talk about, like I said, I could talk about this all day. I think this is just such a fascinating topic. Pete Vera, Exit Algorithms: Agreed, yeah, no, it's ⁓ very interesting and exciting to see where it's headed. ⁓ But yeah, that makes sense. And most operations have a variety of different SKUs and types of boxes, packages, so forth that would make a lights-out facility very challenging. Dan Cahalan: Yeah, and I think if you have that kind of money and you have that kind of product, go for it. Certainly possible. think the problem is that, I mean, it's a matter of who is automation fit for. It's more of what automation are you fit for? You know what I mean? Pete Vera, Exit Algorithms: Yeah, definitely. And we touched on it a little bit, but what are some of the main, you know, ROIs that the ⁓ warehouse operation can expect from implementing automation and, you know, for example, working with Swiss log, what would they expect to get? Because it's not just cost savings, right? It's also ⁓ ability to generate more revenue in some cases. Dan Cahalan: Yeah. Definitely. Yeah, I think what you're exactly correct. think cost save, like a very simple ROI calculation, you know, it's like, how much am I investing divided by... Or how much am I investing divided by how much am I saving? How much am I saving? It could be recurring basis labor. It could be storage. could be, I no longer have to buy 200 forklifts at $15,000 a piece or whatever. Like that's your relatively simple ROI analysis. think the more abstract version is adding in additional components, as I mentioned, like safety, ergonomics, et cetera. I think to your point about further abstract stuff about growth, like the other thing that ought can do is you're able to... increase your pick facings. For example, so like in my situation of 600,000 skews, like close your eyes and consider a grocery store. Grocery store might have 25,000 skews or something, right? Like imagine having to store whatever that difference is to from 25,000 to 600,000, multiply your grocery store by that much. And like that's how many individual slots that you're going to need to put in there for a human to go by and pick that product. And then you'll need to create aisles. and more floors and fire expression etc. ⁓ Automation really allows you to scale your pick face quantity in a way that is much more affordable incrementally as well as much more densely. So one really interesting thing that's happened that is like a constant pendulum in warehousing and fulfillment is the difference between really big sites that have a ton of skews that are centralized, know, like mega warehouses, for example. And then the other swing of the pendulum is localized fulfillment. So like, Hey, we want it really close to the customer to ship it to them now. And there's this like kind of constant pendulum in the industry of like we're going to build these mega warehouses so that we can take the five SKUs that Dan ordered on Kohl's dot com and get them all in the same box or swing the other way and say Dan ordered these today he expects them tomorrow we're going to ship them from the nearby store you know so I think like there's this constant pendulum swing depending on customer expectation at the time depending on how much it costs to ship at the time so that those kind of fluctuations I think automation allows you to be a little bit more flexible and reactive to whether it's at the node level. So you put a small piece of automation in a store that can now hold more skews and fulfill more locally or whether it's at the major node level to say, hey, we're shifting back into the major node. need to increase the amount that we store here. I think that there is a lot of flexibility that automation allows. ⁓ But yeah, I would say on the abstract basis, a couple of other things that just immediately come to mind is like, ⁓ Pete Vera, Exit Algorithms: Yeah. Dan Cahalan: You can, you can flatten the curve of how much staffing you need to accommodate peak periods. So people often talk about labor availability, but they don't really talk about is what that means, which is like, yes, on a day-to-day basis, I might have a difficult time staffing, but what's especially hard to staff is during Black Friday and the two weeks following when everybody and their dog needs to quintuple their staffing to handle a quintupling in demand. So what automation can allow you to do is instead of five X in your normal Pete Vera, Exit Algorithms: Yeah. Dan Cahalan: non-peak operation to peak. You could maybe 2 or 3x your non-peak operation to peak. it helps you solve that kind of labor shortage ramp up. ⁓ Additionally, it helps ⁓ the surplus labor that you hire. Automated jobs are typically A, much faster to get up to like a reliable planned rate. And the quality is also typically much higher. When I present you a bin and say, pick two of these, that's typically much more reliable than me saying, go walk Pete Vera, Exit Algorithms: Okay. Dan Cahalan: quarter mile and pick two of the white v-neck t-shirt out of this specific bin that you need to go find. know, so I think there's the quality, we already talked about quality a little bit, but I think ⁓ that's really helpful. The other thing that kind of automation allows you to do is potentially you can collapse yourself down into a shift that is better for associates, right? So instead of having to run a day and a night shift and manage both of those staff, both of those, maybe automation allows you to have enough throughput in an hour that running eight hours is as effective as running 16. and you could say you guys don't have to be on nights anymore you can be on days. So I think there's a lot of like kind of soft benefits to automating because automation at its core is basically going to increase throughput per hour it's going to increase storage density in the same space and it's going be an ergonomically more effective ⁓ solution for your operator. I mean again I could list a lot more things about you know what you're no longer Pete Vera, Exit Algorithms: Mm. Dan Cahalan: buying, but those are the big ones. I think, yeah, a couple other ones are popping into my mind here too. Like if you have a specific ship cutoff window that you have to ship by, let's say you have to get it by 5 PM, you get orders at 3 PM, this is much easier to accomplish with automation than having someone run across the building, try to pick it, try to build it into a wave even in a semi-manual ⁓ operation. it's a, operations just allows you, think, Pete Vera, Exit Algorithms: Yeah. Dan Cahalan: or pardon me, automation allows you to empower your operations to be more flexible to changing order profiles, to achieve their goals in a way that is more consistent without having to ramp up staffing and have every day be a fire drill. Pete Vera, Exit Algorithms: Love it. Yeah. ⁓ thank you. That's so many good ideas, insightful things to think about, know, as business owners are listening and looking to grow their operations. I know we're coming up on time here. We probably could talk all day. There's a lot of other topics I wanted to bring up, but. Dan Cahalan: Yep. Thank We're good. We'll just stop recording. can just chat for the next three or four hours or something. Pete Vera, Exit Algorithms: Sounds good. Yeah, at the end though, I always ask my guests to condense what the conversation was into one practical takeaway. So yeah, what's one practical takeaway business owners can implement ⁓ and start automating their operations? Dan Cahalan: Yeah, I would say just don't go it alone. I think it's really easy to kind of get caught up in trying to research using Claude, like what automation should I use or whatever. I would say don't go it alone. There's just knowledgeable people in the industry both. suppliers, integrators who can help you, but also others potentially in your cohort of companies that potentially have automated in a way that you haven't that you could learn from. And a lot of the people on the supplier side are more than happy to have a conversation with you, briefly discuss what you've got going on and tell you if they think you're a good fit for what they offer. And if not, recommend to somebody else. I recommend, hey, go talk to this labor management person, go talk to this ACR person, go talk to this other company. I do that. maybe for half of the people that I interact with. So I would say don't go it alone and be prepared for, you know, to walk a journey with the support of a lot of people, both in your cohort, on the customer side. One of your previous podcasts, Pete, a guy said, don't befriend your competition, I think was what somebody else said on your podcast. And I thought that that was like a really great line that I think goes to this don't go it alone thing, which is like a rising tide raises all the Pete Vera, Exit Algorithms: Yeah. Dan Cahalan: in this industry I think and so like befriend your competition ask them what they're doing you can share learnings they share learnings and and I think it's a really great way to go so ⁓ that's condensed is a stretch but that's my condensed ⁓ statement there. Pete Vera, Exit Algorithms: Well said. Yeah, thanks so much for your time today, Dan. Where's the best place listeners can find you at? Dan Cahalan: Sure, you can find me on LinkedIn, just shoot me a message. My name is down there, just look for Dan Cahalan at Swiss Log. I don't think there's any other Dan Cahalans that work at Swiss Log probably, but yeah, find me on LinkedIn, shoot me a note, I'd love to chat with you. And Pete, thanks, it's been a pleasure. Absolutely love the opportunity to talk about this. if, yeah, please don't hesitate to reach out if you ever wanna get into other stuff, at other detail that your audience might potentially be looking to find out. Pete Vera, Exit Algorithms: Absolutely, Dan sounds good. Be in touch. Dan Cahalan: Thanks Pete.