GA 528 | The Danger of Automating Broken Processes with Matthew Townsend

You don't know it's broken until you measure it. And that's one of the things that people get tripped up on, is they don't think about measuring their non manufacturing process.

Today I'm excited to welcome Matt Townsend to the show. Now I've known Matt for many years and he's a fantastic continuous improvement practitioner. Matt, and operates a consulting company called Spark Engineers. And today we had a really cool chat about the issue of automating bad or broken processes. And this is a very, I would say, serious issue. And as we talk about quite a bit, it's definitely not just for manufacturing. I mean, this is happening. I even make the argument early in the episode that it's even a bigger issue outside of manufacturing. Because if you think about the whole wide world, you know, of all the things that are happening in the world, manufacturing is a pretty small subset of all the activities that happen in the world. And bad processes are everywhere and automation is everywhere. And when those things marry up, it can be problematic. So we have a good chat about that and then we talk a lot about, you know, how do we go about identifying bad processes and how do we, how do we then go about improving them? So really good stuff. I think you're going to enjoy this one. I know I did. Show notes can be found over@gembapodcast.com look for episode 528 and also be sure to scroll down and go check out Spark Engineers and follow Matt on LinkedIn and all that kind of good stuff. All right, enough for me. Let's get to the show. Matt, welcome to the show. How are you?

I'm doing well, Ron. Thanks for having me.

Yeah, man. Hey, so where are you calling in from?

I'm calling in from Winona Lake, Indiana. Mm.

And temperature today.

Temperature is a balmy 44 degrees.

That's not too bad T shirt weather up there, man.

Yeah, I mean it is. Anything over 35 is t shirt weather.

Yeah, well, I grew up, as you know, I grew up in Ohio, so man, I'm all about the weather bad. And I'm actually going back up to see my mom here and from. I live in Texas now and to see her and I got, you know, I grew up there but my, I've been here so long my blood is thin and my daughter actually attends Franciscan University up in Steubenville, Ohio. So it's basically right on the Pittsburgh border there. And poor girl, I mean she, she was cold in Texas when it was 80.

That's a culture shock, man.

She loves the school. But the. The winter is hard, man, for her.

Yeah.

Anyhow. All right, well, hey, man, I'm excited for the. For today's episode. It's going to be a cool topic, but as you know, Matt, we like to get things going with our guest sharing a quote. So what do you got?

Yeah, so my quote is from Dr. Deming in the New Economics, if you change the rule for counting people, you come up with a new number.

What's that mean to you?

To me, that means that understanding what the data actually means can be misleading when looking at the data. So you need to understand the information behind it in order to interpret what the results mean.

Yeah, yeah. Taking data out of context. Right. Is very dangerous.

Very, very dangerous.

Yeah.

Yep.

Thus the invention of the control chart, right?

Yeah, yeah, yeah.

Very good. All right, well, hey, man, tell us a little bit about your background, Matt. Tell us about how you got involved in all this continuous improvement stuff, and then tell us about spark engineers.

Yeah, so my background is what I like to call eclectic. I have a. I have a bachelor's degree in exercise science with a minor in biology. So I was very science focused, went on and got a master's degree in business administration, focusing on operations management, and then got another master's degree in engineering technology. So, you know, and also growing up, my dad, he was in operations growing up, so it always instilled a lot of things and how to look at processes and how to improve things. And, you know, it's been a big, big part of my life for a long time. But through the years, I worked for large Fortune 500 companies that had these hu. Huge groups of people that were folks focused on Lean and Six Sigma. And so that's how I got started. So I got my green belt when I was at UnitedHealth Group, my black belt when I was at Lippert Components, and my master black belt when I was at Avalon Technologies. So, you know, I spent a lot of time in orthopedic OEMs, which is what's up in this area. I. And then a lot of time in contract manufacturing as well, helping to support oem. So that's been my. Been my path. But the continuous improvement has been a part of my life most of my adulthood. And it's been a lot of fun to get into it and then continuing to learn, no matter what the situation is. You know, if you get your master black belt, you're still learning. If you get your. If you get a master's degree, you're still learning. You know, I mean, PhDs don't even know it all. So, you know, it's, there's always more information to learn and it comes back to the, you know, I was just talking with my kids this weekend about pdca. You know, we're, we're chatting about ways that they can look at their life and plan for the future and try new things and not be afraid to fail. So most of it's just lessons for life.

Yeah. 100% curious. Why, why, what brought you into the, the medical science side of things early on in your education?

For me, that was just looking at the, the way that the world was going. You know, you had the aging baby boomers that, you know, I've always been data driven. So looking at the aging baby boomers, you look at, okay, well, what, what sort of path is this going to take? What sort of trends are we going to see in the world? And that's been a big part of it. So, you know, even, even looking in the current state of the world, is medical device the best place for me to be now? Probably not. So that's why I got into Spark Engineers. I started that. And so we are an engineering consulting firm focusing on providing automation and technology operational health assessments to a variety of customers across the globe. So, yeah, we're doing a lot outside of medical device, which is fantastic because that's, you know, we're starting to see ripples go through the med device industry right now where, you know, they're just now feeling the effects of COVID So we're starting to see some of those things happen. And, you know, with the state of the world, it's like, okay, well, do we spend more time in aerospace and defense? Do we spend more time in, outside of manufacturing? So that's, that's the current state of things is, you know, looking at that, what's the next big trend?

Got it. Got it. All right, well, hey, man, let's jump into the topic of the day and, and just at a high level from a theme perspective, we're going to talk about why we shouldn't automate broken processes. And you know, obviously this is very applicable in the manufacturing world, but I would argue it might even be more applicable outside of the manufacturing world. Because if you think about it and the whole grand scheme of the world and the things that happen in the world, manufacturing is a tiny piece of it, right? You think about all the other things that are going on. So, so yeah, you know, even if you're not in manufacturing, listen, because this, this is, this is applicable to You. So but let's start with like maybe at the beginning, like what, what would you say is a broken process? Like how do we define that?

Yeah. So I think that what, what trips people up a lot is the thought that their process isn't broken to begin with. In, in order to understand whether or not you have a broken process, you need to do process mapping. You need to understand what your customers are needing. And that can be internal customers, that can be external customers when they're needing it. By even with information as well, it's finding that you don't know it's broken until you measure it. And that's one of the things that people get tripped up on is they don't think about measuring their non manufacturing processes.

So give, give an example though, like.

Sure.

Because obviously you could have a process that's not optimal.

Right.

Because all processes can be improved and should be improved. Right?

Yeah.

And so, but is there a point at which you say hey, this is broken?

Yeah. So as an example, looking, looking at a, a process in. It was an hr. So it was an HR process. Human resources and the company that I was working with couldn't understand why their onboarding costs were so high. So what we did was we looked at the process. So got the HR group in a room, put butcher block paper on the walls, handed everybody sharpies and post it notes and we just started to map out the process of its current state. Come to find out that they were doing background checks after people were hired. So after they had started because the way that they were going through it is they were, they were, they were so quick to try and get people in and on the floor and starting to work that they determined that the background check, background check wasn't a critical path for their process. And so when they, when they came through it and this was in a, in a more impoverished area, there was more crime in the area. It was a, it was a larger city, it was a suburb, not a suburb of a larger city, but an area of a city that had had a little bit higher crime rate. And, and so the folks they were getting in had these, the felony charges that they didn't find out about until after the fact. I mean they even hired somebody in that was just recently arrested for murder or let like let out on murder or I forget what it was, but there was some sort of murder involved. And so they, they actually had to walk him out the day after he started. So they paid for all of the drug tests, they paid for all of the, all of the Day one, training, the onboarding information, all that. So they, they then change their process to move that up and say, okay, well this is actually a part of our critical path. So that was a, that was a broken process. I've also done work for a few organizations in their accounting departments as well where we looked at their month end close and with their, with their processes, they were taking, you know, five to seven days to close their books every month. And you know, when they were looking to become a publicly traded company, that was not going to fly. So we looked at building some databases and things of that nature to help improve the flow of their, of their information. But what we found out was that their processes, within their month end close, it took three times too long just because they were waiting on one piece of information from one group of people. And so, so we knew that automating that process was not going to make it go any faster. They were still going to have to wait until they got that information. So then it's, you know, tracking down the, the, the information in the source and working with that group.

Right, yeah. And from a manufacturing perspective, what are some of the classic examples of broken processes that you've seen people attempt to automate?

Yeah, so you have, you have these areas that have high scrap rate, right. So let's say, let's say you're making a high tolerance or tightly tolerance product and your scrap rate is 5%. Now some people might think, oh, 5% is not that bad. But if you're, if you're working in medical device, you're shooting for under one. And with that, you know, with the tightly tolerance process, if you start to automate some of those things, you'll see unless you have automated checks in that system and that information being looped back in, in a closed loop system to the machine to make the offset, you'll see things start to drift just because of tool wear. Your speeds and feeds are different, you'll start to see cosmetic defects, you'll start to see a lot of issues. And so they're always chasing that, that improvement. So in, you know, you can, you can watch a ton of videos online where, you know, based off of the type of insert they have on their tool, can make a difference in the chatter that's, you know, made on their, on their machine. So you can, you can change a few things here and there that help to improve that process to where you can automate it and have less issues and leave it unattended. But it's always important to do that mapping before you Start digging into automation.

Right. Well, and another way area that I've seen a lot of is just not necessarily, well, I guess it would be defined as a bad process, but just the whole concept of automating waste. Yeah, right. Like if we're just, we're you know, like big conveyor systems, you know, that's a, that's waste, you know, like put everything closer together so you don't have to convey things across your facility. Right. Because the customer is not paying for that thing to be traveling on a conveyor route. But I've seen so many people, they're so proud of their conveying system. And I get it, you know, in Amazon or something like that, they need conveying systems. Right, right. But you know, make sure that you've got things as optimized in a point of use and you know, there's, you know, before you, you start automating all that kind of stuff.

Right, right. Yeah. I worked with a distribution company to help automate some of their processes and, and that was their big thing was all we need conveyor systems. We need, you know, this and that to try and make it, to try and make it easier in what, what they, they ended up doing was looking at their, their highest running products, forming those in a U shape. Right. So that way their, their value added time spent picking that product was still there. They removed the non value added by making it a U shape. So yeah, so they're, they're, the time that they were finished picking to the time they started picking again was much shorter.

Right.

It wasn't a, it wasn't a straight line. So then, you know, we had to reconfigure the warehouse and make things faster. But that way then anything that else that was higher running would then come back to that U shape. And then, you know, so it was, that was a fun project, but.

Right.

But yeah, I mean it's everywhere. And you don't want to automate waste, you know, you don't, you want to take that, those steps out of the process as much as possible?

Yep, yep, yep. No, for sure. So I mean, I mean you've touched on this a little bit already, but can you maybe give some more ideas of the issues that surface when we, when we are automating waste and we are. Is it just like you're making bad stuff faster or, you know.

Yeah, you're, you're making bad stuff faster. You're, you're adding labor into the process. So non value added labor is being added into the process. So you know, if you look at it from a cost per piece view, right? It's, it's, you are adding more non value added time to that process because of what you're doing than what it should be. I mean the goal in a lot of these manufacturing settings is either to reduce the cost of the product being made so you can improve your margin or give cost concessions to the customer if you needed to, or help to reduce a lot of those costs. But what can happen is that you're adding costs in the back end because you're having to hire 10 more people in your quality department to catch bad.

Product before or you're spending millions of dollars on inspection equipment. Right?

Yeah, yeah, ye, yeah, yeah. You know, it's quality is not the problem. It's the cost of poor quality. Your cost of poor quality goes up. So, you know, that's the thing that people need to keep in mind when they're automating is you need to get in someone, you know, whether it's internally with your engineering department, if you're smaller and you don't have that, hire a consultant to come in and look at the process and you know, focus on, you know, work to the wise here. Focus on consultants that are, that are looking to improve your processes and give you value as fast as possible rather than, you know, milking that cow dry.

So okay, well let's transit, that's my next kind of topic or question is. All right, we get it. Don't automate waste. Don't, don't automate bad processes. So what are some practical tips or some things that you've seen work for folks who was like, all right, well how do I identify and how do I improve bad, broken, wasteful processes?

Yeah, so the identification is the key there. So looking at the process from a perspective of you know, you need to do your sipoc, your look at your suppliers, your inputs, the process, the outputs and the customers look at that gives you the overall view of your process. And this works for manufacturing, it works for distribution, it works for the back end processes, it works for, you know, health insurance companies, everybody. If you look at it that way, right, that breaks it down into five buckets. And then you can, then you can look at the process map within that and that can be a business process map. You can look at, you know, your value stream mapping, you can look at all sorts of different process maps. You can do, you know, fishbone turtle diagrams, all the, all the wonderful tools that, that you know, I've, I've been able to experience on Gem Academy which has been fantastic. But that's a topic of conversation for another time. Right. But they. You can look at all these things and start to map out the improvements. So. And you could do it in a Kaizen event. You can do it over time, but you need to have the people that are involved with the process doing the work in order to understand the ins and outs of the process and understand the different paths that a process can go down. You know, I've been lucky enough to be around a lot of people that are implementing systems, a variety of systems, whether it's a machine monitoring solution or erp. Mes. I mean Salesforce instances, all these things and that's the number one thing that I tell them all is you need to map out the process of the way that it is, understand what you want it to do, map out the way you want it to be, and then determine the delta between the two.

Yeah, yeah. So talk maybe a little bit more on just specific mapping techniques that you've seen work in, in maybe a front office environment. Are you throwing down traditional value stream maps? Is it more of a detailed process map or what? What have you seen work?

Well, so I've, I've used both value stream maps and, and detailed process map and for a lot of the areas in, in the front end. Right. Or even the back end, it's. It's that. That detailed process map for a lot of those that I have found that provide the biggest bang for the buck in understanding those processes. Even in standing up a PMO project management office. It was a detailed process map to help make it as lean as possible with being able to achieve fantastic results. So it's that the sipoc, the detailed process map and, and really trying to streamline the operations.

Yeah, I like swim lane. I like marrying like a swim lane detailed process map together just to see all the handoffs of all the various people. Like why does it got to keep going back to such and such 42 times? Right?

Yeah. Right, right. Yeah. And then sometimes it can be an infinite, infinite loop. You know, if that information is not there. I mean I've even seen it where a process can be. They tried to automate a broken process at one company I worked with and they, their order intake would come from their fax number and automatically go into a, an email bucket. But if it had any Chinese characters in it, it threw it into a black hole that it never made it into their, into their email inbox for the incoming orders. So they had, you know, hundreds and hundreds of orders that came in that were never serviced because it was automated to a fault where it just went into a black hole. So that brings me to my next point, that even after you do these, these process maps and understand the process, you need to test out the solution before you go live with it. You know, my, my, my brother in law and my wife, they work in the tech space and you know, they're, they're like, oh, when people test things in production, it's always going to be a recipe for disaster. So always test things out in a sandbox or you know, side environment to where you can see if the solution is going to work. But yet the business process maps, the detailed process maps using swim lanes and you know, there are a lot of products out there that are extremely helpful in business process mapping. Like Miro, you've got Visio, there are some others out there as well. But those are some fantastic products to really map out processes and they're, they're simple to use. They've got, you know, tons of information on there about how to use the product, the different types of operations within the product itself within your process, you know, understanding whether or not something's paper documented or in a database. I mean those are, those are key pieces of information in addition to the swim lanes to understand the, the overall process.

Got it. Good stuff, man. Well, I, I appreciate you coming on and I think this topic is something that we all need to really reflect on because we all have opportunities to improve. Right. And so no matter how good you are, I mean Toyota, these companies at the, at the top, you know, obviously look at Toyota, I mean, you know, they haven't been stellar lately. Right. And so they, no one's, no one's infallible here. Right. So I think we all need to be looking at how do we improve. And yeah, getting back to the basics of map out your process, even if you think it's working well, map it out and then figure out how can we make it a little bit better, you know, even just 5% better. If you're doing very well, customers are happy. Fantastic. How can you make it a little bit better? Because then you could potentially win new customers that you don't even have today.

Yeah, yeah, yeah, good stuff.

All right, well, hey man, how can people best connect with you, Matt?

So the best way to connect with me is either through email or through LinkedIn. So through email it's matparkengineers.com and that's Spark with a C, S, P, A R, c or my LinkedIn. It's just Matthew Townsend within LinkedIn, and either way they can get a hold of me. We can talk about processes, we can talk about automation. Spark Engineers is actually a certified systems integrator for a large collaborative robot company, but we always go into it with that mindset of all right, let's improve the process first before we automate and try to help out as many people as possible.

Yeah, man, awesome. Awesome. Well, again, thanks for coming on and keep up the great work and yeah man, let's go fix some processes.

Heck yeah.

All right, take care.

All right, thanks.

Thanks for listening. Whether you've been on the continuous improvement journey for many years or perhaps you're just getting started, Gemba Academy is here to support. And while we're best known for our more than 1500 Lean and Six Sigma teaching and virtual tour videos, we also have a team of experienced Lean and Six Sigma practitioners available for one on one coaching, as well as a variety of Lean and Six Sigma certification options. To learn more and to schedule a demo, head on over to gembaacademy.com.