Full Transcript for Connecting the Systems Behind the Sales Process with Andy Janaitis

business. 00:04:25.279 --> 00:04:32.000 Because I think so many, especially in the small to mid-market space, they miss that boat entirely. 00:04:32.079 --> 00:04:34.639 They they think of it as two totally different worlds. 00:04:35.040 --> 00:04:36.000 Exactly, exactly. 00:04:36.079 --> 00:04:40.399 And I think one of the pieces that you kind of hit on there is, you know, the process around it. 00:04:40.480 --> 00:04:46.079 You know, I always say for my own business, you know, we we run a small shop, it's myself and and two employees. 00:04:46.319 --> 00:04:50.720 And everybody, you know, the answer for a lot of this stuff is always, hey, let's let's build an SOP, right? 00:04:50.800 --> 00:04:56.879 Like let's, you know, at some point we're years into our growth, like we we really should have a standardized process around this. 00:04:57.120 --> 00:05:01.839 And I always say that the SOPs really never took off for our company. 00:05:01.920 --> 00:05:09.759 You know, like I would, I can't tell you how many Google Docs we wrote up with the sales cycle, with how we're doing our project management, all these different things. 00:05:09.920 --> 00:05:11.600 And it really never stuck. 00:05:11.680 --> 00:05:14.720 Um, and we realized that it wasn't an SOP problem. 00:05:14.959 --> 00:05:18.399 It was that the systems that underlie all these problems weren't connected. 00:05:18.480 --> 00:05:32.160 So once we started to do some of the standardization work so that, you know, for instance, on the sales side, making it so that as you're moving up somebody through the pipeline, you have to send certain emails at certain times or you have to enter with the proper fields. 00:05:32.399 --> 00:05:37.040 The SOPs only go so far because it's one document that probably nobody's ever going to read. 00:05:37.199 --> 00:05:37.519 That's right. 00:05:37.600 --> 00:05:49.600 Um, but these days, you know, especially in the age of AI where we can write these into standardized processes that are actually woven into the way people are doing the work and doing the same thing in this in the underlying systems. 00:05:49.759 --> 00:06:02.399 Um, that's where I think you get the most value is by not only coming up with the process, but then, you know, and this is a kind of the real industrial engineering take, is like, hey, let's make the system such that you can't break process. 00:06:02.560 --> 00:06:05.360 Um, you know, it's the the the way forward there. 00:06:05.600 --> 00:06:06.639 That's exactly right. 00:06:06.800 --> 00:06:13.759 I can't, I mean, so everybody understands how to write an SOP that just documents how the process should work. 00:06:14.480 --> 00:06:15.519 That's easy to do. 00:06:15.759 --> 00:06:24.160 But writing an SOP that does exactly what you're talking about, that incentivizes the correct behavior, that's that's where the good engineers function. 00:06:24.319 --> 00:06:29.279 You know, I did an RFID project, which I get is is totally separate from

what you're talking about. 00:06:29.360 --> 00:06:38.000 But that RFID project took years to get right because all we did was write an SOP that said, you know, scan the RFID tags, right? 00:06:38.240 --> 00:06:39.040 And guess what? 00:06:39.279 --> 00:06:40.079 They didn't do it. 00:06:40.319 --> 00:06:43.839 But once we were able to put in place systems that forced them to do it, right? 00:06:44.000 --> 00:06:56.480 One-way RFID combines where you couldn't get material unless you scanned that tag, then we started getting the correct behavior because we had we had figured out how to incentivize it to kind of force what we what we wanted. 00:06:56.639 --> 00:07:00.240 So yeah, I think the same thing, it applies to all processes. 00:07:00.319 --> 00:07:06.720 When you're writing that standard, when you're thinking through how are we going to roll this out, that has to be one of the questions you're asking yourself. 00:07:06.959 --> 00:07:08.399 How can I force this, right? 00:07:08.480 --> 00:07:11.759 How can I make it so that it doesn't work unless you do this, right? 00:07:11.839 --> 00:07:12.720 That pokeyoke. 00:07:12.800 --> 00:07:19.600 How can I pokeyoke this process so that it doesn't work unless unless you you know unless you follow the correct standard? 00:07:19.680 --> 00:07:21.279 And that's that's not easy to do. 00:07:21.439 --> 00:07:25.600 Uh most of the time that's not easy to do, especially on the administrative side. 00:07:25.680 --> 00:07:32.480 So I love the fact that you're kind of thinking through that and and putting in place processes that force it. 00:07:32.639 --> 00:07:41.279 And I also think, and I'd love to get your your feedback on this, you know, I know this is cliche, but it's but it doesn't make it any less true. 00:07:41.839 --> 00:07:44.079 What gets measured gets managed, right? 00:07:44.319 --> 00:07:57.120 Being able to put in place you know, KPIs that you can actually track to show or to to indicate performance is going to also drive that correct behavior. 00:07:57.439 --> 00:07:57.839 Exactly. 00:07:57.920 --> 00:07:59.920 Yeah, we're we're a big proponent in our shop. 00:08:00.000 --> 00:08:08.879 We run EOS, the uh the entrepreneur operating system, but you know, a big tenant of that, and there's a million different frameworks that kind of all roll up to the same kind of idea. 00:08:09.040 --> 00:08:27.920 But one of the big tenants is building a scorecard, and especially a scorecard that's updated in real time, so that at any given point you have a clear picture of what's going on in your business and you know what the key, you know, not only those key KPIs, you know, typically you're gonna be looking at, you know, revenue and margin, you know, right at the end of the day. 00:08:28.079 --> 00:08:40.480 But let's take those down a level or two and find some of those leading indicators that actually roll up and and you know, at the end of the day, what's gonna make your revenue and your margin get to those targets that you want to set in any given months, quarter, or year. 00:08:40.799 --> 00:08:45.120 Yeah, you you know, in in my business, we call we develop KPI trees, right? 00:08:45.279 --> 00:08:47.440 Which is a bowler chart or a scorecard, right? 00:08:47.519 --> 00:08:54.879 It starts off at the top with net profitability, and then we break that down into the the KPIs that make up each of that, right? 00:08:54.960 --> 00:08:59.679 So revenue, cost, what goes under cost, material cost, labor costs, overhead cost, right? 00:08:59.759 --> 00:09:04.240 And just keep driving it down until you get to those leading indicators you're talking about. 00:09:04.480 --> 00:09:08.799 But yeah, it's always amazing to me that companies don't take the time to do that. 00:09:08.879 --> 00:09:10.159 I think they just don't understand. 00:09:10.399 --> 00:09:17.679 But if you don't have that level of detail in your business, it's really hard to figure out where the break is, right? 00:09:17.759 --> 00:09:18.639 Where the problems are. 00:09:18.720 --> 00:09:24.639 You you gotta have that level of data to be able to make sure that you're working on the right things in the business. 00:09:24.799 --> 00:09:29.440 Otherwise, you could be spinning your wheels, banging your head against the wall on the wrong thing. 00:09:29.919 --> 00:09:46.559 And we find one of the big breaking points, it does, you know, not to use my uh my hammer for every nail in the world, but it comes back to a systems problem often in that these owners know that they need to pull some of those numbers and they roughly understand how some of these KPIs roll up. 00:09:46.799 --> 00:09:52.000 The challenge is, you know, if you go out and say, you know, what exactly is your scrap rate generally? 00:09:52.159 --> 00:09:53.519 Like, good luck finding that. 00:09:53.600 --> 00:09:57.759 You know, you know, probably you could run a study and figure that out over the course of months. 00:09:57.919 --> 00:09:58.879 You're not gonna do that. 00:09:58.960 --> 00:10:11.840 So finding ways to get that data into a dashboard or into something that you can see and use on a regular basis, as opposed to just saying, like, you know, notionally, I understand how some of these pieces work together. 00:10:12.000 --> 00:10:30.639 Um, I think a lot of the real reason why a lot of these people, um, a lot of these founders kind of run their business either by rule of thumb or just by kind of gut feel, is just because they, you know, it's it frankly takes too long to create the reports they would need to do to see the data that they really need to um to be able to pull this data. 00:10:30.720 --> 00:10:34.399 And and the answer is connecting systems and automating the processes to get it. 00:10:34.720 --> 00:10:38.639 Yeah, you know, uh you touched on the the scrap topic. 00:10:38.799 --> 00:10:40.480 That always blows my mind. 00:10:40.559 --> 00:10:45.440 Uh, you know, I came from automotive, which is you know

not smaller mid-market manufacturing, right? 00:10:45.519 --> 00:10:46.639 I mean, that's definitely large cap. 00:10:47.279 --> 00:10:53.440 And uh, you know, we tracked our scrap rates down to like the second decimal, right? 00:10:53.519 --> 00:10:57.600 I mean, it's it was it was a big deal, and and we looked at it every day. 00:10:57.679 --> 00:11:01.279 And and then to go in these small to mid-market firms and ask them, well, what's your scrap rate? 00:11:01.519 --> 00:11:05.519 Like, you know, not a lot. 00:11:05.919 --> 00:11:06.080 Okay. 00:11:07.279 --> 00:11:10.480 That uh that always is just kind of a culture shock for me. 00:11:10.559 --> 00:11:16.159 So yeah, I you know, I and I think to your point, the reason they don't have it is just because it's difficult to track, right? 00:11:16.320 --> 00:11:30.960 I mean, unless you have it, unless you have an MES that's tracking it for you, and most of these small to mid-market firms don't, then you're stuck with having Billy Bob write it down at the end of the shift and then enter it into some Excel spreadsheet. 00:11:31.200 --> 00:11:34.799 And we all know that that's unreliable at best. 00:11:35.200 --> 00:11:36.000 Exactly. 00:11:36.639 --> 00:11:38.559 So, you know, I don't think so. 00:11:38.639 --> 00:11:40.080 Moving on to the next topic. 00:11:40.159 --> 00:11:47.360 I I'm pretty sure it's a law that you can't have a podcast in 2026 without referencing AI. 00:11:47.600 --> 00:11:50.799 So I've always been a rule follower, so here we go. 00:11:51.200 --> 00:11:51.679 That's it. 00:11:53.759 --> 00:12:00.320 Andy, you know, your read is that when an AI pilot flops, it's not the AI and it's really not even the people. 00:12:00.480 --> 00:12:02.879 It's that you're not feeding it the context. 00:12:03.120 --> 00:12:13.519 Walk me through what feeding it context actually looks like and where you tell someone not to use AI uh at all and just write a deterministic integration instead. 00:12:13.840 --> 00:12:14.399 Exactly. 00:12:14.480 --> 00:12:14.799 Yeah. 00:12:15.120 --> 00:12:18.960 So, you know, we we often forget that AI is so, so new, right? 00:12:19.039 --> 00:12:29.120 So when we kind of look at the you know, the evolution, you know, was it two, three years ago, Chat GPT comes out and people start plugging some things into it, and it's like, oh my gosh, like this is amazing. 00:12:29.200 --> 00:12:32.960 We're getting like kind of human readable answers to questions and things like that. 00:12:33.120 --> 00:12:42.480 Um, but then you know, pretty quickly you kind of go through the hype cycle and you realize, like, yes, it's a cool demo, but you know, ultimately it's just not close enough that it's not really production ready. 00:12:42.559 --> 00:12:44.320 It's not what I'm gonna be using. 00:12:44.480 --> 00:12:59.759 And I think at that point, people started to realize the value of you know what we would consider like prompt engineering and kind of coming up with like, hey, you've got to really detail the question, really detail you know, the context and the situation that you're putting into it, um, and and kind of how you can use that. 00:12:59.840 --> 00:13:11.840 And again, you know, you started to get much better results once you wrote these books of a prompt, essentially, that that would you know, write your emails for you or whatever kind of test project you were throwing it at. 00:13:12.000 --> 00:13:26.879 Um, but now where I think we sit is that a lot of times, you know, companies have have gone through that process and they're starting to understand, you know, if I take the time to really feed all this information in in a in a really well-constructed prompt, I can get a pretty good output. 00:13:27.120 --> 00:13:31.200 But realistically, I'm not gonna do that in my day-to-day and use that all the time. 00:13:31.519 --> 00:13:38.399 What we're saying is, you know, the only way to get good value out of these tools is the context that you feed into it. 00:13:38.480 --> 00:13:42.639 You've got to let it see, you know, I love using the example of writing emails for you. 00:13:42.799 --> 00:13:48.480 You've got to let it see the corpus of emails that you've written in the past and learn what your voice is and all of this. 00:13:48.799 --> 00:14:07.120 And if you're only doing that in a, hey, let me pull five emails and drop it in and let it try to write the next email, it's just too manual that you're just not gonna do it, especially when you think about pushing that down to your team who maybe doesn't really have any real interest one way or the other and whether or not we make this process a little more efficient or not. 00:14:07.200 --> 00:14:10.480 It's you know, it's they're not thinking about the bottom line the way that you might be. 00:14:10.639 --> 00:14:15.039 Um, so the only way to really get this done is to start to connect those systems. 00:14:15.200 --> 00:14:23.200 So, you know, in that email example, I love what we do is every time an email is going out, we're having the AI draft it. 00:14:23.279 --> 00:14:32.000 But then crucially, the human is going out and reading that draft and probably rewriting about half of it and you know, getting some of the ideas, but really making it their own. 00:14:32.159 --> 00:14:40.399 And then after that goes, um, our AI tools are then reading, hey, not only the email you sent, but what was the email that was drafted? 00:14:40.559 --> 00:14:48.639 Let's diff those two things together and say, hey, what can we learn about your voice and how you like to respond to emails based on what those differences were? 00:14:48.799 --> 00:15:04.639 And we find that by doing that over and over and over again, you get this continuous improvement cycle that by the end of it, all of a sudden you've got a really strong tool that can that can um you know help you get through your inbox in minutes where it used to take hours, kind of thing. 00:15:04.720 --> 00:15:20.000 Um, and

really the only way that that all works is by having the direct automated integration with the back end that us as users don't have to say, hey, go out and read that that email that I sent and you know, diff it, and here's all the different instructions. 00:15:20.080 --> 00:15:28.960 It's just, hey, you're already connected, you have the data, and you know the skill that every time I send an email, I want you to go out and collect that data and make sure that you're improving the system moving forward. 00:15:29.120 --> 00:15:31.679 Um, so I really like to use that example. 00:15:31.919 --> 00:15:38.879 Obviously, that's a pretty simple example, and there's a lot more we can do across the business stack with kind of similar principles. 00:15:39.200 --> 00:15:52.159 But you know, at the end of the day, I think where a lot of these AI pilots fail is that we're not feeding the context in and we're not connecting the tools that would really, you know, make the AI pilot work properly. 00:15:52.559 --> 00:15:57.440 Do you think here's something I'm running into in my business, right? 00:15:57.600 --> 00:16:02.879 So in addition to consulting, we also do some bespoke software development for for firms. 00:16:03.360 --> 00:16:10.720 Um and what we're finding is our biggest competitor really isn't other software development shops, it's our own customers, right? 00:16:10.960 --> 00:16:20.960 Um, because now with Claude and Codex, uh, you know, they feel capable of developing their own web applications to solve to automate these workflows. 00:16:21.440 --> 00:16:33.440 And you know, we're kind of in this this time where my software developers are like, you know, Claude and Codex can get you a certain point. 00:16:34.000 --> 00:16:54.159 But when you're really talking about complex workflows, like a quoting tool, right, where you're actually gonna go look at a repository of drawings and previous quotes, compare that to an RFQ, uh, you know, and then make a determination on similarity and then make a suggested sales price, right? 00:16:54.240 --> 00:16:56.320 To expedite this this quoting process. 00:16:56.480 --> 00:17:04.400 When you're talking about doing something that complex, you know, you better have more than just Claude behind you. 00:17:04.559 --> 00:17:09.359 Like you need to have some some some software development experience to do this. 00:17:09.680 --> 00:17:20.000 Do you think that AI maybe has allowed us to get out over our skis a little bit and in in and actually what we're capable of doing with it, or or not? 00:17:20.079 --> 00:17:25.279 Do you really think, no, man, uh, I hate to tell you, Justin, but your customers are right. 00:17:25.359 --> 00:17:28.240 They don't need you, they can do this completely on their own now. 00:17:28.720 --> 00:17:37.359 No, I I think you're spot on there is AI is you know one of the most powerful tools to have in your tool belt, but there must be a human in the loop. 00:17:37.519 --> 00:17:45.359 And to your

point of you know, understanding that software development, I uh, you know, we've built this tool, kind of an operating system for our entire business. 00:17:45.519 --> 00:17:49.279 And I I would tell you, you know, we've been building this thing for probably about a year now. 00:17:49.599 --> 00:17:54.400 And literally every day, something will come up that's just flat wrong. 00:17:54.559 --> 00:18:00.799 That if I just took it at face value, I probably would would not, you know, the tool is the most powerful thing we use. 00:18:00.880 --> 00:18:04.559 Like I would say, you know, probably cuts our overall workload in half. 00:18:04.720 --> 00:18:12.799 But if I'm not sitting there and actually giving a critical read of every single thing that gets output, then you know, it can go off the rails pretty quickly. 00:18:13.039 --> 00:18:18.480 Um, you know, I kind of like to think of it almost like, you know, to bring it back to a shop floor perspective. 00:18:18.720 --> 00:18:23.839 Think about you know, the manager coming in, think thing thinking that they're going to optimize a particular process. 00:18:23.920 --> 00:18:35.839 And they just say, hey, you know, we're just gonna switch this up, we're gonna go ahead and you know, flip this process before that process without ever talking to the actual people who are operating the line and figuring out, you know, why you could or couldn't do that. 00:18:36.000 --> 00:18:49.440 I I think that's in some ways, you know, where where AI can come up with some of those great ideas and some of the brainstorm and say, hey, no one's ever thought to, you know, let's see if we can get it through paint before we do finish inspection or something like that. 00:18:49.759 --> 00:18:58.400 But if you don't have that intelligent person who understands the full process and who's actually done each of those steps and can say, wait a minute, wait a minute. 00:18:58.480 --> 00:19:03.039 In this particular, you know, that that's a great idea, but in this particular case, here's why it wouldn't work. 00:19:03.200 --> 00:19:08.319 I think those are the types of things where you've got to have a human in the loop and you've got to have an expert. 00:19:08.480 --> 00:19:21.839 You know, it's almost like um, you know, when we're writing new AI skills or AI processes, I always like to kind of encourage the team, like, hey, do it manually first once, you know, or you know, figure out how to how you would do this manually. 00:19:22.000 --> 00:19:24.400 And I think it's the same exact thing from a software perspective. 00:19:24.480 --> 00:19:32.240 It's like if you're trying to build complex apps and you've never written a line of code before, like it's not gonna work. 00:19:32.319 --> 00:19:33.200 It's gonna flop. 00:19:33.359 --> 00:19:35.759 So just you got to have that basis. 00:19:36.160 --> 00:19:39.200 I mean, I think too when we rolled out our first application, right? 00:19:39.279 --> 00:19:41.519 I mean, like with any software, there were bugs, right? 00:19:41.599 --> 00:19:42.960 Hey, why does this not work? 00:19:43.039 --> 00:19:43.920 Why is this doing this? 00:19:44.000 --> 00:19:47.119 Like, oh, you're right, let me let me go figure that out, right? 00:19:47.440 --> 00:19:50.480 And we did that for a month, six weeks, right? 00:19:50.559 --> 00:20:00.400 Where we were finding bugs, fixing bugs, and then we reiterate and and I just look at that, I'm like, well, how would you I mean obviously AI is the one that created the bugs, right? 00:20:00.480 --> 00:20:04.400 I mean, yeah, i if if it's right in the code, so how do you how do you fix it? 00:20:05.359 --> 00:20:13.279 So anyway, I you know, I um I guess from my perspective, the jury's definitely still out, right? 00:20:13.359 --> 00:20:15.839 I mean, I kind of I I I agree with you. 00:20:16.000 --> 00:20:26.799 Um, I feel like you probably need to have some coding experience if you're gonna write complex applications, but but that's not the prevailing sentiment right now in the market. 00:20:27.039 --> 00:20:33.519 So I think we're just gonna have to let this thing play it, run its course and and see who lines up being right. 00:20:33.759 --> 00:20:36.160 The good news is we're gonna find out real quick, right? 00:20:36.240 --> 00:20:38.319 We're not gonna have to wait around a year, right? 00:20:38.400 --> 00:20:41.519 I mean, folks are trying to use this thing now to build these applications. 00:20:41.599 --> 00:20:45.680 So we're gonna find out pretty quickly is this uh is this something they can do or not. 00:20:46.000 --> 00:20:54.720 I I always come back to the the Gardner hype cycle when I whenever kind of thinking about evaluating any new technology, and it always blows my mind how how often this comes up. 00:20:54.799 --> 00:21:02.720 But you know, with every new technology that comes out, you know, you've got that kind of you know peak of inflated expectations, and all of a sudden you start to see what it could do. 00:21:02.799 --> 00:21:21.680 And you know, I think we're starting to get over that hump, but you know, even six months ago, you know, I was listening to podcasts where people were saying, hey, you know, software engineering as a as a skill is going to go away because you can go ask Claude to refactor an entire code base and you know, rewrite it from one framework to another framework, and it'll just go and do it. 00:21:21.759 --> 00:21:25.599 And you know, in in one minute, you could do what your team would do for in years. 00:21:25.759 --> 00:21:32.799 And you realize that there's so many reasons why that's not not how it works and just not how it's ever going to work, you know. 00:21:33.599 --> 00:21:40.640 Not to mention, you know, the energy and the cost that actually goes into running some of these tools, you know, once you start building some of that extra complexity. 00:21:40.799 --> 00:21:44.400 But um, yeah, certainly I think it's a really, really powerful tool.

00:21:44.559 --> 00:21:49.759 But um, you know, Claude, I feel like is like a really, really strong junior developer. 00:21:49.839 --> 00:21:59.759 But if you don't have that systems architect and you can't, you know, kind of check their work and understand why certain decisions are being made, certain design decisions, um, you're gonna be in trouble. 00:22:00.240 --> 00:22:01.200 Yeah, yeah, yeah. 00:22:01.359 --> 00:22:02.160 I think you're right. 00:22:02.240 --> 00:22:04.480 Or I mean, I guess for my business, I hope you're right. 00:22:04.640 --> 00:22:07.680 But uh, you know, time is certainly gonna tell. 00:22:08.319 --> 00:22:09.279 This is true. 00:22:09.519 --> 00:22:09.839 Yeah. 00:22:10.000 --> 00:22:13.599 Well, Andy, you know, you describe yourself as the systems and numbers higher. 00:22:13.680 --> 00:22:17.519 A two to twenty million dollar manufacturer really can't afford full time. 00:22:17.839 --> 00:22:23.279 You know, a real posting around$130,000 that three or four shops can't fill. 00:22:23.519 --> 00:22:29.039 Who's doing that job today in most of the shops you walk into and what breaks because of it? 00:22:29.359 --> 00:22:30.000 Yeah. 00:22:30.480 --> 00:22:32.480 Yeah, I mean, there's a lot of answers for that. 00:22:32.559 --> 00:22:35.599 And I I think you know, the real problem is it's an unanswered question. 00:22:35.759 --> 00:22:54.880 Um, but you know, it it thinking through, you know, hey, if we're gonna bring on a new ERP system at probably the most complex project level, you know, or even just like, hey, we're gonna, you know, run a spin-up harvest to do time tracking for our team and just make sure that we understand, you know, when people are coming and going kind of thing. 00:22:55.119 --> 00:23:17.279 I think those types of projects, um, you know, integration projects, start to fall to uh, you know, maybe the founder, if the founder is really involved in the day-to-day, maybe uh, you know, director of operations, somebody in that role, oftentimes maybe even an admin person, if just like, hey, you know, we gotta, we gotta get this tool figured out, um, you know, go ahead and and figure it out, um, which is fine. 00:23:17.359 --> 00:23:35.680 And I think that's where some of these SaaS products, it is easy enough to spin up and to, you know, enter the data in and start to get running, especially with like those more, you know, more simple business systems, you know, things like time tracking, CRMs, you know, those are things that somebody on your team can spin up. 00:23:36.000 --> 00:23:42.480 But where there's a real gap is in that kind of systems architecture of understanding, okay, this is great. 00:23:42.559 --> 00:23:48.160 We have time tracking now, but you know, it doesn't speak to our uh quoting tool or our invoicing tool. 00:23:48.400 --> 00:23:52.720 It doesn't speak to our financials when we want to pull reports at the end of the month. 00:23:52.880 --> 00:24:00.079 Um, you know, it it doesn't speak to project planning as we're forecasting, you know, what's what's coming down the down the pipeline. 00:24:00.240 --> 00:24:16.319 So I think those are there's a lot of really great tools out of there, out there that pretty simply out of the box, you can get a lot of real value out of them, but you can kind of take that value to the whole next level when you can start connecting a few of them together and making sure that you're using all the data you have available to you. 00:24:16.400 --> 00:24:20.160 Um, and that's the role that I think just doesn't really exist. 00:24:20.319 --> 00:24:24.480 Um, I think the other the other place we actually do see it quite a bit is in IT. 00:24:24.640 --> 00:24:32.079 You know, so you have a an IT, either, you know, a full-time IT person or even you know, you've outsourced to an IT team. 00:24:32.160 --> 00:24:38.400 Um, and we see this a lot where you know we'll come in and we'll start asking some questions about, hey, how is this tool used? 00:24:38.559 --> 00:24:40.000 Why is it used this way? 00:24:40.160 --> 00:24:43.519 You know, what other what are the roles that that can go into it? 00:24:43.680 --> 00:24:45.359 What um you know, how is it connected? 00:24:45.519 --> 00:24:48.480 And we'll always get passed back to you know an IT person. 00:24:48.640 --> 00:24:52.160 Then you get the IT person on the phone, they're like, man, I like I don't know. 00:24:52.319 --> 00:24:59.200 Like somebody asked me to install this on the network, I got it installed for them, and now they're coming to me asking like how this tool works. 00:24:59.279 --> 00:25:01.359 Like, this is not my thing at all, you know? 00:25:01.599 --> 00:25:11.519 So I think that is uh another kind of gap there where um you know that that people don't realize that, yeah, it's more than just getting it installed and spinning up a few user accounts. 00:25:11.599 --> 00:25:19.119 Like you've got to really understand what the tool does, how it interacts with the tools around it, and and why you would make some of those decisions. 00:25:20.000 --> 00:25:32.880 I think this kind of points to a broader topic that that I experience a lot in my consulting business is small to mid-market manufacturers don't take the time to think through their org chart, right? 00:25:33.279 --> 00:25:43.839 You know, it kind of grows organically, you know, hey, we're we're building more, we need to add on a second shift, so hire a supervisor for that and a couple hourly guys to run it. 00:25:44.160 --> 00:25:55.039 Um you know, but they really don't think through here's my organization, here's where I'm at, here's where I want to go in terms of revenue dollars, units produced, whatever. 00:25:55.599 --> 00:25:59.599 What skills do I need to help me scale to this, right? 00:25:59.759 --> 00:26:04.319 You know, it they kind of look at it as well, we just need to build more, build more, right? 00:26:04.400 --> 00:26:07.440 We need to hire more supervisors and hourly

folks. 00:26:07.839 --> 00:26:20.160 And and they don't think through well, no, what what what are the support functions that I really need to be really good at to ensure that when I bring in these supervisors and hourly folks, they're able to execute at a high level, right? 00:26:20.240 --> 00:26:33.440 So, you know, your systems guy, your IT guy, you know, obviously, uh, you know, I I I I can't remember the saying, but it's like, you know, everything is everything's tuberculosis or something, you know, everything is industrial engineering to me. 00:26:33.680 --> 00:26:38.160 Like, hey, here's uh you know, do we have the industrial engineers on staff that could drive this? 00:26:38.400 --> 00:26:55.359 Um you know, I I do think that exercise of thinking through that org chart and what not just the direct labor, but the indirect piece, like what at what point does it make sense to bring in, you know, these these uh full-time roles? 00:26:55.440 --> 00:27:04.720 And until you can get there, how are we gonna backfill like with a with an Andy or with a Justin to do some of these things that we can't quite afford yet? 00:27:05.119 --> 00:27:10.640 I think a big thing there is right, people think in terms of the position that they're hiring for. 00:27:10.720 --> 00:27:13.359 So, like, okay, how many uh supervisors do I need? 00:27:13.519 --> 00:27:15.759 How many industrial engineers do I need? 00:27:16.000 --> 00:27:20.559 But really, we should be flipping that on its head and thinking through like what are the roles that need to be filled? 00:27:20.640 --> 00:27:22.480 You know, what is, you know, what are the different jobs? 00:27:22.640 --> 00:27:27.359 And especially in a small manufacturer, a lot of those jobs are gonna have the same person associated with them. 00:27:27.440 --> 00:27:35.119 So one person may be wearing five or six different hats, but until you take the time and sit down and say, like, what are all those different hats first? 00:27:35.279 --> 00:27:41.359 And then we can go and overlay, like, okay, who would be the right people to try to fill as many of those roles as we can? 00:27:41.440 --> 00:27:44.799 And then, you know, ultimately a lot of it ends up falling back to the founder. 00:27:44.960 --> 00:27:52.400 But um, I think that's, you know, don't think in terms of how many of X people do we need? 00:27:52.559 --> 00:27:55.200 Think like, what are the jobs that actually have to get done? 00:27:55.279 --> 00:27:57.759 And then get creative with like, okay, how can we solve those? 00:27:57.839 --> 00:27:58.720 Are we gonna outsource it? 00:27:58.799 --> 00:28:01.119 Are we gonna have one person do two or three of them? 00:28:01.200 --> 00:28:03.519 Or, you know, how are we gonna get that done? 00:28:03.839 --> 00:28:04.000 Right. 00:28:04.160 --> 00:28:11.519 Maybe even a little more um high level than that, what skills do I need to be able to execute this, right? 00:28:11.599 --> 00:28:15.839 I mean, what are the skill sets

that we need to have on the team to be able to do it? 00:28:15.920 --> 00:28:25.200 And to your point, you know, when I got five people, you know, one guy filling five roles, I think you you gotta have an honest assessment of what is this person's skill set? 00:28:25.599 --> 00:28:32.559 Where can they be, where can they really be a good fit here, and where do I need to figure out how to offset, right? 00:28:32.640 --> 00:28:38.960 How to supplement, because it's just not look, man, you know, I if you put me as a quality engineer, I'm gonna do a terrible job at that, right? 00:28:39.039 --> 00:28:42.559 That's not my that's just not my gap, that's just not my area. 00:28:42.799 --> 00:28:45.039 Um and and that's fair, like that's okay. 00:28:45.440 --> 00:28:49.359 That doesn't make me a bad employee because I'm not great at this this one job. 00:28:49.759 --> 00:29:00.000 Um but if you're a small manufacturer and you need me to be your quality engineer, we're gonna have to figure out a way to supplement that so that we can actually make sure we're putting good parts out the door. 00:29:00.319 --> 00:29:00.960 Yeah. 00:29:01.279 --> 00:29:02.559 So well, good deal. 00:29:03.279 --> 00:29:10.799 You know, Andy, an MA attorney told you that if you are your company's sales process, that's a valuation discount. 00:29:11.039 --> 00:29:14.960 For a family business that grew on grit and is now scaling past it. 00:29:15.200 --> 00:29:18.480 What's the minimum set of leading indicators before revenue? 00:29:18.559 --> 00:29:22.240 And how do you get reps to actually fill in the source fields? 00:29:23.119 --> 00:29:24.720 That's a great, great question. 00:29:25.119 --> 00:29:27.519 I love that you've done your research for uh for this podcast. 00:29:27.599 --> 00:29:29.599 I love you've been uh digging into my content. 00:29:29.759 --> 00:29:34.079 So that's you know, something that we we do bump into all the time. 00:29:34.319 --> 00:29:51.599 Um, you know, especially, you know, you hear this this idea of like the silver tsunami and a lot of you know, a lot of these manufacturing shops, especially small, medium, then somebody's been running it for for 30 years, maybe the kids don't want to take over and they're they're getting ready to start thinking about moving towards acquisition. 00:29:51.759 --> 00:30:01.039 Um, and we like to think about process in the sense of like, hey, let's make it more efficient, let's you know, get the systems connected because that's gonna help your business to run better. 00:30:01.200 --> 00:30:12.720 Um, but another kind of big piece, and and this really struck me when when uh this this MA attorney kind of mentioned it to me, was like, you also got to think that's a big part of how valuation works. 00:30:12.880 --> 00:30:19.440 So if somebody's gonna buy this business, not only what kind of revenue are you putting through today, what kind of accounts do you have? 00:30:19.599 --> 00:30:52.079 A big question is gonna be like how repeatable and you know, if I'm gonna buy this business, can I step in and run it today, as opposed to, you know, if it all kind of runs through the founder's head, you know, I always like to think on the sales side is a really clear example of like, hey, if the whole sales process is the founder's network and people pop up and you know he ultimately closes a sale based on relationship alone and nothing ever hits any systems, then you know you're gonna have a really hard time moving on from that business because there really is no business without that person still in that role. 00:30:52.240 --> 00:30:56.720 Um, so that's that's where we kind of get that idea of like, hey, it's a valuation discount. 00:30:56.880 --> 00:31:16.000 If you're thinking about, even if you're thinking about, hey, five years, 10 years down the road, or even if you're not thinking about selling, having this in your back pocket that you know at any given point, if I had to step out of the business, someone else could step in really makes, you know, gives you some peace of mind, but also ultimately makes your business a lot more valuable. 00:31:16.240 --> 00:31:16.880 Yeah. 00:31:17.119 --> 00:31:21.119 Yeah, I think uh I think of that in my own space all the time. 00:31:21.200 --> 00:31:25.519 Like, you know, I have I created a business or have I just created a job for myself, right? 00:31:25.759 --> 00:31:27.200 Because those are not the same thing. 00:31:27.359 --> 00:31:30.400 Yeah, those are very different, very different things. 00:31:30.640 --> 00:31:36.960 So, well, Andy, thanks again, man, for agreeing to come on the show, share your knowledge with the folks uh with our listeners. 00:31:37.200 --> 00:31:38.960 Um, you know, really appreciate it. 00:31:39.039 --> 00:31:45.839 And guys, if you've enjoyed this conversation between Andy and I, please do, you know, like, subscribe, share the podcast with a friend. 00:31:46.000 --> 00:31:59.359 Uh, and if you're in a small to mid-sized manufacturer and you're looking to improve that sales cycle and and you think, hey, you know, we've got some systems, but I really don't have a clear picture of of how we're doing and how we need to improve, you know, reach out to Andy. 00:31:59.440 --> 00:32:03.279 I'll put his uh link to his LinkedIn and his email address in the show notes. 00:32:03.440 --> 00:32:05.920 But uh again, Andy, thank you so much for the time, man. 00:32:06.319 --> 00:32:07.680 Thank you for having me, Justin. 00:32:08.160 --> 00:32:08.960 Absolutely.