TROND: Jessica, welcome. JESSICA: Hi, Trond. Thank you. TROND: Yeah, I'm excited to have you on the podcast. I thought we would talk about manufacturing, very big surprise. JESSICA: Yeah, [laughs] I know, shocking. Theme of my career. I'm excited for this. TROND: Yeah. How did that become the theme of your career? JESSICA: A bit coincidental. Interesting question. So I studied aerospace engineering, which was a purely technical degree. So, from that path, more of a design engineer, I knew I never wanted to be a design engineer. Frankly, I was an athlete in college, and the thought of a desk job scared me. There's no way I could imagine myself sitting in a desk doing models or analysis. I knew I didn't want to be an engineer. I landed an internship in manufacturing with GE Aviation through my junior and senior year of college. Generally liked it, thought it was interesting. I got an offer to go back pretty early in the school year. So it de-risked my senior year quite a bit to just accept that. And then I did like that the role was so engaging. You talked to people all the time, and you sort hardware. The shop feels like a playground in some ways. And it certainly wasn't a sit at your desk for 10 hours a day kind of job. So somewhat landed there coincidentally, and then after those first few roles, kept coming back. TROND: I mean, you were an intercollegiate athlete, actually. You were pretty serious at figure skating. JESSICA: Yeah, if you're in the figure skating world, collegiate figure skating it's the opposite direction of collegiate football, if you will, as far as seriousness. But relative to going to college, being an athlete, having to juggle schedules, and all of that was there, you know, getting injured and needing to go to practice. So all of that was certainly there from being an athlete and managing that. TROND: But you were never in doubt as to whether you would go into sports professionally. JESSICA: Correct. That was not a route that I was going to go down. TROND: Got it. Well, you did go down a few other routes, actually. So, like you said, undergrad aerospace from BU, and then MBA from Wharton. So you were into business quite early as well. How did that whole journey happen? JESSICA: Yeah. So if I go...from BU, went into aerospace and defense manufacturing. I was on a rotational program with GE, which was a phenomenal experience. I didn't know what I wanted to do. So the fact that I got to see a bunch of different plants, different functions within operations was valuable. By the time...and we can get into this with some of the more in-depth stories and experiences. But by the time I was managing quality for...it was a $52 million revenue year plant. I had very strong awareness that there was so much I didn't know coming from a purely technical degree. I had scattered mentorship throughout my time at GE, bits and pieces were picking up. But I had awareness that I didn't understand anything about the financials of the plant. There were so many things that I needed to understand to make decisions better, and I had no idea what those were. So I had this awareness that I needed to learn more. And I had for a long time been thinking about an MBA as a way to career pivot, potentially broaden what I had seen outside of just aerospace and manufacturing, which is niche upon niche, which is a very [chuckles] tight, little box. TROND: I love how you talk about how you didn't know enough. But, I mean, compared to a lot of people, you must know quite a bit, right? I mean, you've done degrees at very good universities in the right kinds of areas. Manufacturing is this kind of industry, I guess, where either people just work their whole career and don't really particularly have an education for it, or if you [chuckles] actually start looking for education in manufacturing, you could go on forever, right? [chuckles] JESSICA: Yeah. TROND: You could learn this, and then you...wait a second, I don't know anything about that either. Let's just do that. It is this endless source of questions. JESSICA: Yeah, it's broad and deep too because some of my earlier roles I was very deep technically. Like, I will never, I don't think, technically be as deep as I was when I was a manufacturing engineer in some of those processes in anything ever again. There's so much depth. And there's also so much breadth as we come up with new types of manufacturing. Additive, relative to the history of manufacturing is newer. And there's all these, like, specialty processes, and coatings, and platings, all this. So there's so much you can learn from the technical side. And there's, like, constantly evolving business things as well. So it's definitely...there's a lot more to it, but you have to get in first. TROND: Well, and I think that's what I wanted to ask you about. Because you are now at Tulip, right? So here we're talking digitally-enabled manufacturing in all kinds of ways. And we'll talk about that in a minute. But before that, it's just, I think, probably instructive to reflect a little bit back on some of the experiences you had very early on that were deep in manufacturing and perhaps before digital was as mature as it is now. So I wanted to hear a little bit maybe back in your GE days working in discrete manufacturing, what were some of the things that you, I don't know, were frustrated with then or the kinds of problems that you had to solve, you know, with or without digital as a tool? How did you go about it? JESSICA: The role that I had where I was a manufacturing quality engineer at a site up in Lynn, Massachusetts, due to a series of unfortunate events, there was a point at which I was the only person who could disposition hardware in the plant on certain days. We had lost some other engineers. Actually, let me give people a sense because I know this is a podcast. So the visual around that. It's a large plant, very high ceilings. There's two manufacturing cells to it. One of the cells...and when I say cell, these were sort of, like, small business units. We called them cells. Some other plants might call a cell, like, a U-shaped classic for a specific product line. These were almost, like, two separate business units. The plant was cut in half, one made large rotating hardware. Rotating is...I say that because it was all aircraft engine components. So rotating hardware tends to see the highest stress, the most vibrations of a hardware in the aircraft engine. So large rotating hardware for a fighter jet like F-16, F-18 engines. And then the other half made the product that was the bane of my existence. They were called cooling plates, about six inches in diameter, awful little things but high volume. So one side was varied mix but definitely lower volume. The other side was very high volume, small parts. So you got all this hardware running around. It's very different from a manufacturing process standpoint. And every individual piece of hardware was serialized. It's aerospace, so the traceability is super important. And everything is serialized. The raw material, the casting, the forging, whatever it is is also serialized. All that is, you know, genealogy-wise, you can track it. And everything had a paper router, so the paper router tells you what step you're sequenced to go in. And if there is a defect or what we call non-conformance that we were calling them, you had to call it out as an operator on the back of the router and write it up in ink, black ink. That was how I was trained, black ink, no pencil, ever. No blue, always black. So I always had multiple black ink pens on my person. My high-level estimate is my handwritten signature is probably on 2,000 to 3,000 part routers, at minimum, just the back of the envelope of the number of pieces of hardware I dispositioned, you know, number of days a week I was in the shop, weeks per a year kind of thing, minimum 2,000 to 3,000. TROND: So, when people talk about pen and paper in manufacturing, they're not joking. It's literally you're aching at the end of the day here in your hand from writing all these things. JESSICA: Yes, yeah. I mean, truly, it was tough. And there's not just the...sure, that's not convenient. There's a lot of lack of efficiency in writing things out by hand. The lack of visibility to what was going on and the stoppages you could create in manufacturing when you have defective hardware, and the hardware needs to stop until someone dispositions it. I mentioned there was a period of time when I was the only person who could disposition. I usually came into the site around 6:00 a.m., so I could overlap for an hour with third shift before they left at 7:00. And there were days when coming in from the parking lot operators would be stopping me trying to, like, bring me their hardware, telling me...trying to do the right thing. Like, "Hey, I called something out. Let me stop the part, get someone to look at it." But unless they brought me the router or left a sticky note at my desk, I almost had to walk the shop to find the hardware. We didn't have a true MES. We just had this sort of MES-flavored system that was put into place, I think, in the early '80s. It looked like MS-DOS. Like, it was not anything fancy and not always reliable. TROND: And this MES for...so Manufacturing Execution Systems. That's software that runs or tries to capture some of this running data on the floor, right? JESSICA: Right. And it's a world I know much more about now. But a system that, in theory, should be tracking the data that's manufacturing level. So hourly maybe minute-level data of what work order? What serial number is it? What machine operation? Is it called out for something? Did it get put on hold? We had very limited visibility to that. So we would [chuckles] usually print out where all the hardware was at the beginning of the shift. And then you'd go around and mark up that piece of paper as you went out throughout your day. And then if something was non-conforming and it was a key piece of hardware...I mentioned sometimes we had...for the mix, the hardware that we were dealing with sometimes had raw material that had six months lead time, like, a very specific type of material forging or whatnot. That one piece of material gets called out, you can't just replace it. And that's going to, like, a specific aircraft waiting for it. So you would track that on a piece of paper and heaven forbid something goes wrong. [laughs] And now you have to go to your shift meeting and tell everyone that that one part is bad. Everyone memorizes the serial number, and that's how we communicated. TROND: Wow. So this is a while back. Is it your sense that that kind of process now is automated everywhere? Or because it is so regulated, is it perhaps not even automated? Because there are all kinds of other reasons why they need to keep it the way they did. JESSICA: Interesting question. When you say automated, you're referring to the actual manufacturing process? TROND: No. I'm talking about what you were just saying, the correcting the noting errors and steps in the process. JESSICA: Yeah. I'd like to think, you know, with the rate at which we're sending things into space, that all of these shops, especially in aerospace and defense, as you mentioned, tends to be higher tech, highly-engineered products, would be using a digital system to capture data. It seems like it'd be relevant given the number of apps we have for every part of our life. But I would bet money most of those shops, and I probably saw at least a dozen shops while I was in aerospace, most of them are still doing some form of paper-based routing, acknowledgement. I think dispositioning for quality issues should be in some sort of system now, but there's likely still an enormous way they can go to help track data, get visibility to it, give operators tools that they can use. TROND: So, Jess, let's talk about this for a second. Why is it, if that is the case? I mean, who knows? You know, in that particular shop floor you were talking about there, whether they have a new system now or not. But why would it then be that some of these things do take much longer? Is it because it's kind of unclear who should make the decision to make a change here? Is it because there's a little bit of fear to make any kind of change in a process that already...if it isn't well-oiled, well, it's a huge issue. So, you know, you may digitize certain things, but you might destroy other things. Talk about the fear of digital and change in kind of quality overall in manufacturing. It would seem to be a real barrier. Even though in their private life they have apps everywhere, they're sort of thinking, well, you know, it's different. JESSICA: It's an interesting question and an interesting phrasing on the fear of change. Or I'd phrase it as change is difficult. You could apply that phrase to almost anything. In manufacturing, especially in a plant that runs all the time and resources are constrained, change can be really difficult if you don't have someone to help drive it. Like you said, who's the decision-maker here? You probably need alignment between whoever is paying the bills and your quality team stakeholders. The quality team themself...in aerospace, amongst regulated industries, I don't think there's that much fear of change necessarily because the pain is so real. They value data. Like, those functions understand how to look at data. But the path to get from the current state to a future digital state, finding the software, deploying the software and that's where a lot can go wrong. Just finding resources to deploy the software, learn it, manage it, teach operators how to use it, like, teach the teams how to use it, you need someone who can really drive change and action. We used to use the term change agents at GE. Like, we'd drop people and say, "The point here you're supposed to drive change." But it's easier said than done, right? In a shop floor with hundreds of hourly workers, you have supervisors who have been there 20 years and whatnot. To drop in a new system, collect data they weren't collecting before, make that data valuable all of that is so much work. And so many things can go wrong. I think there have been multiple attempts to implement software like that in at least one of the shops I was at, and they didn't succeed. TROND: Right. So, is that one of the reasons why you are at Tulip now? That the overall business case, at least one of the clearest one for Tulip, is sort of to say, wherever you are, you can just start and implement one little thing in digital, and you can get the data immediately. And you don't necessarily have to do everything at the same time. So you can test it, and check, and sort of feel comfortable and kind of move in iterative fashion. I would imagine that when you realized what sort of software that was, you must have reflected back on whether it was your time in consulting, which we can talk about, or, you know, where you were advising, you know, all of these clients on, you know, what should you do. Or it was truly, you know, you were the quality engineer. JESSICA: Absolutely. I was back then...I'm trying to estimate. I've probably been in close to 40-ish plants, mostly in the U.S. and probably only, like, 8 to 10 in Germany I did as part of an academic trip. But in most of those plants, there's so much paper, and writing, and whatnot. But as I mentioned, I understand how difficult it is to change things there. When I was a supervisor implementing small changes, it often took an enormous amount of hand-holding, and time, and whatnot. And Tulip is so interesting in that you can create these small...we call them apps. I know people who aren't familiar with Tulip might think of it kind of, like, a module, except you can configure it to whatever to start solving pain points in bits and pieces. But you can also design a larger system with data structures and make it more robust and get your pieces to be cohesive. But you can do it in a way that matches your manufacturing process. And I think that's one of the key challenges manufacturing has is plant to plant, even plants that do the same basic processing, whether it's CNC machining, injection molding, and grinding, whatever those are, or are the same types of product, there's usually still some differences. They might be because this machine shop has, you know, multiple rough turn machines because one machine is older. They might have just different configuration because of the physical layout of the plant. And so to be able to configure the screens and the data you need to match your actual process is really powerful. TROND: Yeah, and there's something there, too, because I guess the inclination, if you look at it, is sort of to say, well, we're going to switch everything out. We're going to have one digital system that does everything. But that seems somewhat unrealistic given this diversity, both of physical layouts, and talent, and the existing layer of different types of systems that are going to be in any given plant, right? So it's not always like you can actually start from scratch. JESSICA: When you say start from scratch, you mean start from having no digital system in place? TROND: Right. So it's not like digital is this thing that shows up once and then, you know, you can tweak everything into this new thing. It seems to me that you're constantly interacting with other systems, adjacent systems, you know, systems that do different things. JESSICA: Yeah, definitely. And that will be true in the sense of people who are more familiar with the system world. Like, you have your ERP level system, and you have warehouse management systems, and all that, you know, maintenance-specific systems. Like, not all of those are getting replaced tomorrow by one thing. But Tulip is so interesting in that because you can start small and start filling in gaps, you can start replacing systems and maybe end up where you have a more cohesive base. But you can also create a longer-term strategy for what systems you need in your shop, what system is going to control what data, and how they're going to work together. That level of design and a lot of...and I want to say bigger companies but maybe more mature companies, better-resourced companies have people who can sit and think about what that tech strategy is for their shop. Honestly, I have also been to a good number of smaller plants, you know, 50 people; it's still family-run. They're not going to think of systems that way. And so something like Tulip can come in and help them with very basic data structures very quickly without needing necessarily an IT degree to go think about how we're going to architect the whole suite of systems across something. TROND: I don't know if it's a good thing to make you, you know, sort of compare the two situations. But you've been in, you know, in kind of a GE is a real manufacturing situation. And then you've been consulting for McKinsey and been in many, many different factories, like you said, and different clients and different situations. Which one would you say today, looking back, taught you more about what you're now working on which is, I guess, implementing and helping neutral clients succeed with their digital transformations? Is the consulting viewpoint for you still sort of fundamentally the way you approach things, or do you approach it much more like an engineer? Or is it a hybrid of both? Or are you actually doing it very differently from either of those two experiences? JESSICA: From what is the most helpful to what I do at Tulip consulting is definitely...that experience lent a lot. I mean, it's consulting, so it's client-facing. And I've been customer-facing the entire time I've been at Tulip. It teaches you how to estimate value, which is something we talk about a lot with our customers and try to help them get to a way of measuring. Like, if you improve your productivity, your cycle time, your scrap rate, whatever it is, we should be able to measure that. There should be a metric somewhere. And what is that? And that way of thinking came more out of consulting. And that goes back to my comment earlier on there are many things I didn't know when I was a manufacturing engineer. I picked up bits and pieces, but there had to be more to it. That understanding of, you know, I owned, we called it aged WIP. So aged inventory, non-conforming inventory. Like, what is the dollar impact? How much cash am I tying up with that inventory sitting around? That concept I didn't understand until I went into consulting and had a much better sense of, like, cash flow. Like, that's not a concept I knew in manufacturing. TROND: Right. How about the industry? So you've worked in aerospace and defense for sure. And nowadays, I'm imagining you're working in a much sort of greater variety of different firms. And you're certainly working across regulated, unregulated, and perhaps even in pharma and other places. What are some of the things that you're seeing, you know, around the challenges that these different industries are facing? Is it overall the same thing that people are struggling with? And how different is it when you really have to take into account, you know, regulations and things like that? JESSICA: Yeah. I'm laughing a bit to myself because we had a saying in consulting that every client is a snowflake, but they're also all the same. Because once you start going in and you...we called it diagnostic looking for opportunity, you know, how much productivity can you improve and whatnot? All the machine shops start looking the same. Or, like, even you said different types of manufacturers. You go to other SWOTS. You start seeing a lot of similar patterns, similar behaviors, and whatnot. So there's a lot of similarity. I will say there are pretty clear differences between environments, pharmaceuticals specifically that does batch manufacturing in that regulated way. That environment does look and feel different than any sort of discrete...most of my experience has been in discrete manufacturing across. Obviously, aerospace and defense were regulated, very unregulated like kitchen equipment, just random industrial plastics. I've seen shops like that, you know, far less regulated. But a lot of the basic concepts are the same. Like, actually, one interesting example I was thinking of when I was reflecting on my experience for this conversation was, when I was in consulting, I worked with an injection molding manufacturer, and one of the biggest issues that they had that I was personally as an associate at the time working on was changeovers. And so we were applying SMED principles so Single-Minute Exchange of Die principles to try to reduce the time of the changeover because they were these large, multimillion-dollar pieces of tooling that could take six-eight hours to go from last good part to first good part of the next run. And I was like, oh, had I understood these principles back when I was in a machine shop doing landing gear manufacturing, changeovers were also my biggest source of downtime. But when I was there alone trying to figure stuff out, I didn't have this understanding of how I could apply improvements. So there's definitely a lot between the industries. And to your question on what I've seen, a bunch of stuff, a lot more medical devices and pharmaceuticals at Tulip than I ever had previously. And there's definitely common sharings and learnings between the industries for how to manage [inaudible 21:01] look at metrics, digitize, all those things. TROND: If you were to kind of think of more generic advice than for somebody who perhaps either is running a shop floor or has at least some sort of idea that they could perhaps themselves suggest a digital path, or put in a request for something, or maybe they even have the budget and could just say I want to, you know, do a test of this concept here...Pain points is something everybody experiences, right? Because it's a manufacturing environment and, you know, nothing is optimal. But as you're thinking about how to prioritize and how to, you know, what should I attack first and things, what are some of the things that you now have learned that is the best way to start with a digital changeover or a digital change process? JESSICA: I'm going to say it depends, which is a very consulting [chuckles] answer, but it depends who I would be speaking to in the sense of if you are someone who would say...let's stick with, you know, smaller, I mean, not Fortune 100 level companies because those companies are very different conversation, a lot of strategy and high-level vision there, but smaller-ish, maybe 1 to 5, 10 sites. If you're someone in a position of leadership, and decision-making, and power with budgets and whatnot, to think about how might I invest in a tool? I think a major thing there for them is to understand what are you trying to accomplish? Because I say this to our customers sometimes, we're not digitizing for fun. This is not for kicks. We're doing this because we want to see improvement, and we want to make people's lives better. We want to give our engineers better data or managers better information. So you should have or to be able to develop a higher level vision for what you're trying to accomplish by implementing some new digital tool. It doesn't have to be a full-blown five-year roadmap, but at least have some idea of here are the metrics we want to measure. Here's what we want to improve. And that can help get people on the right path, orient them. If you're someone who's more shop floor level, maybe a cell lead or business unit manager, manufacturing engineer, engineering manager, your world is probably a little more focused on specific metrics. And so if you can identify I've got a metric or a pain that can improve by X amount, it would free up people's time, understanding—we do talk about pain points—what those are, and being able to list them out, and what the impact would be is hugely beneficial. And just taking that first step to what do I actually need? What do I want to digitize? Where do I start? TROND: What about on the negative side? Because, you know, in our pre-conversation, we talked a little bit about sort of this idea of scope and kind of accountability in manufacturing. And sometimes that's not the strong point of someone who is very operational. So, you know, it's not to paint this picture but, you know, if your job is to make things moving, and make sure that things are moving, which really is fundamentally important [laughs] in manufacturing, scope, and sort of like, you know, who's going to do all this, and how much should we really invest in this before we start seeing results that's maybe a different mindset, perhaps. Tell me more about that. JESSICA: Yes, it's an interesting comment in that to understand perspective-wise. Today, I come from a very strong project management perspective in that I've been doing that. I did it for a while in consulting. I did it for a while at Tulip. But prior to business school, I was in operations. I was a manufacturing engineer. I was a supervisor. And it's a very pretty significant perspective shift. If you are in a purely operational role...and I mentioned earlier it's really hard to drive change. One of the reasons it's so hard to drive change, as you mentioned, your day-to-day is to keep things moving. You know, ideal state you have leadership that either recognizes the importance of getting projects done and gives you some time and space to work on projects, like, sets the expectation, hey, a couple of hours a week, you need to go work on X project. Or you bring in someone who has, like, a change agent type role or more of a...maybe they're coming from a central team or some other cross-functional team to help out so that they can clear their brain and think about what is the scope of the project that I'm trying to implement? For listeners, the reason we're talking about this is, when you deploy a new tool, it's very helpful to treat it like a scoped project with clear guardrails to make sure that something goes live. That you have a pilot, it gets executed; you collect real data. Because, otherwise, the configuration part can just churn and cycle for months and months because you keep adding things, especially to a point it's easy to add things. So having someone who will keep that clear, controlled vision of where we're deploying; at final inspection, we want to collect final inspection data, three weeks. Get it out there. Train the operators. Make sure operators get trained. Make sure everyone who needs to know within the shop communication-wise is notified of it. Like, someone to manage all that is crucial. TROND: But just to push back a little bit, I also thought that one of the strengths of a tool such as Tulip where, like you just said, can get started so quickly is that you don't necessarily need an entire continuous improvement or a quality team because it actually...a lot of it could theoretically be done almost by the operators or supervisors. So, how do you kind of account for that? Because you were also saying to me earlier that, you know, it's all great to have a business team come in. And, you know, you have been at McKinsey. So you have these smarty pants come in, and then they improve things. But when you are on the production side, it's hard to sort of see the value of that also because you're so ingrained in what you're doing. And you know that when those high top hat people are leaving, you're going to go back and just sort of you have to still get your stuff done. Is it now possible to bridge that gap between operations and planning in a way that wasn't possible before? JESSICA: Yeah, I get your point on mild, contradictory views here. It's not that they're contradictory. It's that the one they can work together and that they do different things, and each one can get you so far. But you'll end up needing a bit of each. So what I was mentioning before on the value of thinking of this, like, an actual project, having someone help manage scope. Someone who can do that who is a manufacturing engineer. They usually have projects. That doesn't mean that somebody who is approaching this from...at Tulip, we use the term bottoms-up, right? Like, I've got a small use case. I'm going to go digitize this checklist, whatever it is, can't go do that. But if you're going to go broader and try to figure out, say, production planning across the site, how do I get work order tracking implemented across the site? That does require at least a little bit of organization from somebody, whether that's internal or not. TROND: Right. Let's talk a little bit about the future. I'm curious. So, you're someone who has basically been in this business now for a bit. And you have seen it both from the engineering side and kind of consulting side and now helping people implement digital change, kind of merging the two, like we just spoke about. When you look at factories of the future, you know, what is next for this process? I mean, do you see robots, and full autonomy, and these, like, clean, white spaces that are perhaps even in space? Like, do you see this radically transforming, or do you see much of the same? You go back to, like, the first factory you worked in. And you go there 20 years from now, and you see exactly what you saw then. JESSICA: I hope it's not exactly [laughs] what I saw. I would certainly hope it's, you know, got a little bit repainted at least. But the concept of, you know, factory of the future, clean floors, a lot of cobots, AGVs moving stuff around, most of the people there are for maintenance, and you see, like, one person for every 25 machines. My answer is going to still be it depends because those concepts of lighthouses and companies who are building out certain factories to digitize certainly that can exist. And there can be, you know, with enough investment, and planning, and whatnot, a jump from a factory that has very little automation to one that has much more automation. But if you look at the rate of change of manufacturing...I was mentioning the site I was in had the equivalent of an MES, whatever their MES was that was installed 30 years ago, or whatever it was. Even if that rate of change accelerates, we're nowhere close to these lights-out factories for the average. And that's completely fine. There's so much space and opportunity to still improve, still provide really interesting jobs, have skilled workers, like, be able to upskill them in these factories, get rid of some of the paper so that you're not wasting people's time and ink and potentially risking also data. But not have to be at a state where you're at a lights-out factory with robots doing everything. So there's many steps in between there. And I think the average manufacturer, again, maybe exclusive of some of the top 100 factories in the world or whatnot are going to go on that journey. And it's going to take a couple of steps before they're even considering more automation. TROND: I guess that's one of the reasons we have this podcast, this discrepancy between these very abstract ideas of automation as an all-in hairy kind of goal we're going to all automate. Versus this idea of augmentation where you kind of are adding capability that may not even be that expensive to do where it makes sense and where it amplifies some role that either was weak or expensive, obviously, or where an added automation or tracking will add to the result. Why is it that we still have this idea of these robots? Or a lot of us when we think of the future, it is this vision that somehow the robots is the ultimate kind of goal here. Is it even the ultimate goal? Is that really what we should be aiming for, full automation? So it's just a cost question? Or would you say that there can be happy, thriving factories with much less? Yeah, like a robotic, you know, automation, but it can be still very, very efficient or doing things at a very high level of quality still with humans. JESSICA: To your question on why do we think robots...I'm like, robots are cool. And I don't think...most people are going to agree. And there are a lot of cool robots out there. But I definitely think that doesn't have to be the only end state. There are some factories, I mean, auto has been in this train for a long time of it's a lot of robotics, automation, and whatnot. But their product lends itself to that. It's high volume, and they can standardize a lot. The standard advocation for some of the manual labor and keeping it manual is the flexibility you get with the labor force to move things around because you don't have to install a conveyor belt where you really need to think about what's the layout. Sure, in the future, maybe conveyors themselves are on some sort of AGV. And it's really easy to, like, redo your shop floor in 10 minutes to develop a new product line. But we're so far from that. There's definitely a state at which you have people...Tulip uses the term augmented worker where humans have the data they need, the work instructions they need, the tooling that they need, and they can account for, interact for, and they work much more efficiently. And they might have some more advanced machines. Maybe instead of humans pushing carts, there are some more AGVs. But there are certainly still people integral to the operations to be able to check things, work on the products. TROND: I wanted to ask you about this, sort of the opposite scenario. What about factories that supposedly are enormously advanced because they have invested tremendously in robots or other types of automation? What is the role of a system like Tulip or any other system that sort of tries to kind of organize and get the most out of whatever is there? I'm sort of thinking that the more you put in advanced systems, the more coordination presumably you actually have to also do. So maybe you can talk a little bit about when you've seen this work really well. Presumably, there are some places that you see tremendous transformation because you go from pen and paper to even just tracking two or three things and then, suddenly, things really speed up. But, surely, in even very, very advanced factories, the kind of coordination that you actually need, the kind of tracking you need, because you have such advanced machines, it would also, I would imagine, just have an enormous demand on fairly advanced human labor to actually make it work. Because otherwise, the automation [laughs], you know, doesn't really do what it was supposed to do. JESSICA: Yeah, you're probably getting it, like, upskilling and reskilling what humans need in those environments versus -- TROND: Yeah. And how they fit in with the various...because it seems to be that digital is often viewed as one thing. And, in our conversation, you talk about it as so many things. So it's not like digital is one thing, oh, we're going to do a digital transformation. It's not one thing. It's a long set of different choices you're making. JESSICA: Yeah, I come from the implementation world at Tulip. So that's probably why, in my mind, when I say digital and digital transformation employment, it's composed of so many different use cases, areas, pieces of data, bits of change that need to happen for sure. It's a lot of smaller things that come together and get you to a future state place. The question around in that more automated state, so some factories that have digitized quite a bit more and automated quite a bit more the coordination between the systems, so there's a lot of complexity there. I'm by no means an expert at this part. There's enormous complexity in the data and the systems, and then also the machinery being able to manage all that. But then, to your point, the skill sets of humans are still needed to be able to manage some of the equipment and also to look at the data and make sense of it. Yeah, there's statistical process control and all of that. But at some point, you're going to want a person to check some of those things. And there's going to be a place for highly skilled workers. And I know there's some anecdote. But there's one company I was working with this client. I was going site to site implementing performance management, non-digital, like, basics on paper. But they had some skilled avionics, and they would fly these avionics from site to site because mechanics, like, avionics mechanics...but they had so few of them. Like, that highly skilled person who didn't go to college, learned on the job, maybe went to a trade school but came up through the industry is being flown around the country on their private jets to go work on different manufacturing sites because their technical skill set was so high. So there's certainly still opportunity. TROND: Those are the funniest signs, by the way. I've seen that on social media nowadays, you know, here and there. Like, there could be this advanced company in some rural area, and there'll be, like, a billboard, inadvertently, that says, you know, become an avionics engineer or some other very specialized thing. And then it might even list, like, earn hundreds of thousands. And it's true, [laughs]. It's like, if you actually go there, sign up, and you reasonably accomplished, you know, meaning, you know, you pass your, you know, whatever examinations, that will happen to you. It's incredible. You just pick your niche there, and you're set. So last question for you then, Jessica, leapfrogging. If you are someone in this situation that is in this leadership role in a relatively small manufacturing company with perhaps, like you said, three to five sites, leapfrogging, how do you think about this? What is the process? JESSICA: So leapfrogging as far as going from all paper to...and instead of doing, like, little by little, like -- TROND: Yeah. Like, making a big chunk change with digital. JESSICA: Change, yeah. TROND: What is your advice? JESSICA: Build a strong foundation on paper. So have your processes defined, know if your parts need to have certain standard routing, have some idea of what data you want to capture. We had one project I had worked with a client who...industrial with discrete stuff. They didn't have standard routing for the operation, like the steps of the operation. So that makes it harder to digitize because what am I going to digitize? You just have Joe who knows what to do when the part pops out. So you need to be able to start with lean manufacturing, basic best practices, like, know what your wastes are, document, know what the process steps are, have standardization. Ideally, have a continuous improvement culture where people can contribute to. That foundation sets you up because you can take that whole thing and digitize it. TROND: Hmm, fascinating stuff. JESSICA: Yeah. It's interesting to see for sure. TROND: Well, it's great to talk to you. You have clearly made a career in something that you're passionate about. And it's interesting to see, you know, how things are changing and how things don't change that fast. It's a moving target, manufacturing, but certainly, the complexity is there. And, you know, congratulations on what you've done. And thank you so much for sharing. JESSICA: You're very welcome. Fun conversation. Thanks, Trond.