Ep 84 - Keith Perkins and Seth Thorson | Will AI Replace Technicians?

0:00If you want to get my listeners or just the internet trolls fired up, you talk about AI coming for technicians' jobs. They're not going to replace any technicians now, but in 10 years with AI and robotics, the cost has to come down on robotics. Then we can replace physical manual work. The following program features a bunch of doofuses talking about the automotive aftermarket.

0:23The stuff we or our guests may say do not necessarily reflect the beliefs of our peers, our sponsors, or any other associations we may have. There may be some spicy language in this show, so if you get your feelings hurt easily, you should probably just move along. So without further ado, here's your host, Mike Allen, with Confessions of a Shop Owner, presented by TechMetric, the best software in the history of ever.

0:55You know, just this week I got— went down a Scientology rabbit hole and read about, um, Xenu. Yeah, some crazy shit. Have you heard that? Yeah, 75 million years ago, Xenu brought billions of humans to Earth, chained them around volcanoes, and killed them all with hydrogen bombs. And then now their souls are what cause physical and mental ailments to current humans. Yeah, and you got to go to the Church of Scientology to get it.

1:22You gotta pay a lot of money to go to the classes. Yeah, right. Yeah, you go through these pillars to become, you know, enlightened. One of those pillars is free diagnosis. I didn't hear all the volcano stuff. I didn't know about all that. That's pretty wild. But yeah, yeah, Wikipedia, man, you can burn a lot of time on Wikipedia if you're not careful.

1:42So just talk to Chad. What do you think's worse, burning tons of time on Wikipedia or doom scrolling social media? Do you scroll on social media? You're not learning much. Yeah, that's fair. Wikipedia, you can learn tons of things that you may or may not ever use unless you're doom scrolling Confessions of a Shop Owner's shorts and reels. No, um, so introduce yourselves for those who live under a rock and don't know who you are.

2:10Oh, I'm Seth Dawson. I'm an Eurotech. It's all yours. I don't want nothing to do with it. No, uh, I'm Keith Perkins. I've got a brick-and-mortar repair shop in Tulsa, Oklahoma that does everything from oil changes to Everything. Then we have a mobile company that services about 1,700 other professional shops between Oklahoma City, Tulsa, Northwest Arkansas area. And then we have a training company.

2:39We make stuff, content, and travel all over the world. And so we're here at Vision KC 2026. I should specify that, I guess. Uh, have you taught already? No, not till tomorrow morning. Okay, what are you teaching? What are you teaching tomorrow? Uh, Seth's class. BMW and Volkswagen. Yeah. Seth's moved Onward and upward to other things now. All right, Seth, what are you here for?

3:03Seth Dorson, Eurotech Auto Repair. We got 4 stores in Minneapolis and 1 in Wisconsin, so 5 total stores. Also own German Car Support. We support Volkswagen, Audi, and BMW shops with technical support, training, and diagnostic help. Okay, and you taught today on— I taught yesterday. Yesterday, that's right. Yesterday afternoon. The show didn't start until today because I arrived today. Oh yeah, yeah.

3:26In my mind, I am the center of focus in my own world. So, um, you taught on AI. AI, how to use, how to build custom GTPs to streamline operations in your shop is what I taught yesterday. Is it GPTs or GTPs? Can't it be either? Can't it be how to build custom GTs? Yeah, yeah. You guys know how to do our shoes, I imagine.

3:51Yep. So how was class? Good. I think people enjoyed it. It was a good class. We had a lot of time. I think we capped it at 85 people because it's a hands-on workshop. So everybody had to bring their laptops and a working subscription. So we had a lot more people that wanted to come, but we could only do so many when we're doing, you know, making them build the agent or their help bot as they go.

4:16When will you be teaching that class again? I'm doing a shorter demo of it at Minnesota's AASP conference at the end of April, but that's a really small conference. Minnesota's not as big as a lot of the other conferences, but we're doing that out there. That's awesome. And then I don't know when I'll do it again. There's a lot of people that are interested in that, obviously, for obvious reasons.

4:37As David Roman so famously said, you can replace a service advisor with a link. And I think there's a race for folks to try to figure out how to do that as effect, maybe not replace them, but certainly make them a lot more efficient and reduce the bandwidth required to just really relationship building and the rest of it is automated would be pretty cool.

5:00Yeah, I mean, that's what we tell people. That's the first step is using the custom, you know, boundaries on chat so it only works when it's certain boundaries. Then the next thing is you start building with cloud coding and build Chrome extensions. And I gave a quick demo at the end of the day and Michael— So here's the thing. Outdated SMS, just— they don't just slow you down, they cost you money.

5:23And I learned it the hard way. Before Techmetric, I was wasting time on inefficient processes, manual updates, back and forth calls with customers. Now I handle everything in one place. DVI, customer communications, payments, real-time reporting. It's all in one page. Since making the switch, my average repair order has jumped from $293 to $916. And it's not just me. TechMetric powers almost 10,000 shops nationwide.

5:49By the time you're hearing this, it probably will be 10,000 shops, helping them grow and operate smarter. If you're tired of losing time and money to outdated systems, tap the link in the show notes and see what TechMetric can do for you. Hey guys, Kari Lynn with Turnkey Marketing. If you are looking to increase cars and you're looking for the right demographic to go after, you wanna get the right people who need auto repair right now, then give us a call.

6:14We have a service called Direct Track and it utilizes AI to find people in your area who are the great demographic that you want to go after, have raised their hand and opted in saying, "I need auto repair help right now." We send them an email. As soon as they open the email, we then get their physical address, follow it up with commercial ads on all their streaming services like Hulu and YouTube and ESPN, Fox News, all those different things.

6:40And then we also get their physical address and we start sending banner ads and display ads to every single device in that house. It has been incredibly effective. It has made shops seem like they're everywhere to those people who need repairs right then. And I mean, I'm telling you guys, the return on investment has been huge. So if you want to increase car count, you want to get great people in the door, give us a call or reach out to us and ask us about DirectTrack Marketing.

7:11Gunther, I think, came up and said he built a Chrome extension to do oil changes already because after he learned how to do it, he's like, oh, that's easy. Yeah, it tracks customers' usages of the vehicle to predict when the next likely oil change interval would hit date-wise. Yeah, it seems like it would be pretty easy to go ahead and build it to automatically set their next appointments and their reminders and that kind of stuff too.

7:32There's a lot of things you can build, a lot of things you can do, and the nice thing is it's customized for your shop. The biggest thing is really, if you look at software as a service, a lot of the software as a service I think is in trouble because it is so easy for somebody that is not a coder or programmer to build something.

7:48And Chrome extensions can literally read most shop management systems without needing an API and integration because you can literally scan the developer code and, and pull most of the data you need out to make a decision and and database out of it. Well, it feels like, uh, the shop management systems that are out there now that are trying to grow and maintain relevancy or establish dominance, right, they're all racing to build the same tool into what they've got already, right?

8:18Yeah, and I think, I think the big ones, the big ones will survive, or if you have a really specialized set of data, right, like MyTechSupport or KeystechSupport You know, if you're scanning our database with our own AI, I'm not going to let anybody else's AI into it, but if we're scanning our database, you know, still the intellectual property or the, you know, the data is still going to be absolute king because AI needs the data.

8:42So people that hold massive amounts of intellectual property, intellectual data, are still going to be okay and still be able to build something that's sellable. But if you're just relying on ChatGPT to look on Wikipedia, Google stuff, Google stuff and find stuff like those companies that are doing it. And I'm not going to name names, but there's a lot of them out there in our industry that I think once people figure out how to do what I teach, and I know Keith does some teaching on in different ways, that I don't, I don't think some of these companies will be around.

9:12Yeah. So do you think that, um, it seems like some of the companies that kind of latch onto the side of the main softwares, the integration companies, that kind of thing, and they've got an API or they're doing a Chrome extension that overlays Techmetric or Shopware or whatever. They're harvesting that data to be able to build. They're smart. Yeah, so from a shop owner's perspective, if you want the tool that they're offering, how do you protect yourself from giving up your consumers' your customer's data to them.

9:49Build your own yourself. Yeah, yeah. I mean, ours, ours, I use a Google script that scrapes the information data off and builds our own database so that I can build quick, fast, expedited estimates using the whole knowledge base of everything we've done on these vehicles. And we're starting to scrape it with a script and then do a shared database. But I mean, that's early, early development that we're working on, on it.

10:10But yeah, I don't— I think everybody that's going to take your data, data is the next gold mine. And it's always been— it's always been— yeah, it's just getting easier. Carfax forever. It's like, I'm not integrating because I now got customers asking for it, like you had. Yep. And I'm like, they have access to my data for free, you know? Yeah. So, um, people that watch this episode and listen to this conversation, or people that hear about the class that you just taught It is the hot-button subject right now, right?

10:45You know, for the last 3 or 4 years before that, it was the technician shortage. Now it's AI and how to integrate and increase efficiencies in your business with AI. I feel like there's a demand. You could teach that class once a week online and fill a room of paying attendees. Like, what do I got to do to I should have come yesterday, but my kids had a play I had to go to.

11:09So how do I get to go to that class? I don't know. I'm not sure when I'm teaching it next. I guess I will look at— I'll look at the form a little bit too. If you're, if you're watching and thinking, well, I can go on there and build a GPT and tell it to take these words and change them to these words to make my service provider sound smarter, there is so much more than that.

11:25Yes. I did not attend your class, but I can assume you go building agents and integration and, like I said, Chrome extensions Yeah, and that, that we touch on. Well, that's, that's really— we, we start with GTP because that's the foundation. So it's, it's what I call level 1, and then level 2, we talked for 10 minutes at the end with Claude and integration.

11:46But really, until I get you to build 5 or 6 working GTPs, there's nothing else you can do. Um, and I had 2 or 3 guys that said, well, I understand this, I already have working ones. By the time they were done with how I taught them to refine and control and data and then push it into an extension where they no longer have to go and copy it.

12:05Yeah. And I say get the prompt and paste it back in. Like, you could build an extension that just does that. Yep. Yep. And until— I tell people, don't try and build the extension until you have it working here, right? Because it's easier to manipulate and change it here than it is to change the code. Once you build the code, it's harder to unwind it.

12:22So I tell people, let's start here. Um, and the guys that said they had something working really, really well They left class going, I can make mine work so much better now. I understand so much more. A lot of people also don't understand the data controls. Like, I teach them how to build an HR bot because I get sick of the HR questions.

12:38What— how much vacation? How much PTO? What do I— or they'll ask a sensitive question that they may be afraid to ask their manager that they'll ask the AI bot that will feed them, then reports back to their manager because the man is always watching. No. Yeah, yeah, not an AI bot, but, but You know, pump your handbooks into them and build a searchable database for processes.

13:00And you teach them to turn web search off, right? We don't want that going to the web to find an answer to please somebody. If it can't find the answer in the docs I gave it, it needs to say you need a report. This is a good question for your manager. Yeah, it's a good question for— based on the type of question, which level of leadership they need to talk to.

13:18Correct. And so that's what we teach a lot of, is how to turn web off. Yeah, our teachers turn the web off on that. Or how do you make it so you're not training the model? Like, a lot of these guys are training the model, and frankly, people are typing all sorts of crazy sensitive stuff in the, in the chat. And if you— chat learns from it, it won't spit out— I can't go ask what Mike Allen has typed into it, but I can, I can get enough.

13:42Yeah, I can get enough information if I prompt it right. I can get enough information about people's businesses and things that if they've uploaded things. Yeah, in the documents. I mean, that's— it's a little bit scary how much data we're feeding it without understanding how to protect what we, what we doing. And so I'm a big proponent of using paid ones and making sure that— yes.

14:05So if you've got subscriptions, then you control the data. If you're using the free version, did you turn the— did you turn the button on or off? Yeah, you got to tell it don't train other units, keep it internal. And then that's where you get into building an agent that keeps it offline. You can use Claude to code one and have an offline version and set up your own AI that literally doesn't share with anything, not even with them.

14:25So you can take it as far as you need to go. So 2 weeks ago you were in North Carolina. Yeah. In the booming metropolis of Frog Pond, North Carolina. Yes. For a class with Josh Parnell and Limitless Leadership. Um, North Carolina has got some good operators and some dudes who are really investing in bringing high-level training in-house. We've got Fueling Connections coming up and that's a leadership and ownership focused thing.

14:51But like, what's it look like for me to get you in North Carolina just to come and teach all of us North Carolinians, anybody who wants to fly? And just let me know, we can set it up. I mean, that's pretty— I want to set that up. Yeah. I mean, get with Megan and she can set it up and I bet we can put 80 people in a room in North Carolina pretty quick.

15:07Yeah. It's just a matter of making sure that, you know, I mean, we did it, we did it here and we had 80-some people and we were We were— we only had one other helper from WorldPac that was able to help, but we were able to use that amount of people and help. If they're not computer savvy, then it takes more monitors in that room to run around and— oh, they didn't buy the premium version, or the— you know, I tell people start with the premium, and then I start showing them features of connecting Google Drive to connect their company KnowledgeBook, and they're like, well, I can't do that with the premium.

15:37I said, yeah, I know, you have to step up to the business. But I tell people when I teach the class, a lot of people don't know what they're getting into. And I say your minimum investment to take the class, it's a $20 prerequisite to buy the basic premium to get into the class to be able to do it. And then if you decide you want to upgrade, then you can buy more advanced packages from, from ChatGPT.

15:59So the next step up from OpenAI, first one's $20 a month and the next one's $200 a month. Is that right? It's only $30 for a business. Oh, okay. Yeah. And so, I mean, the one thing, the one thing Chat allows you to do is share your, your custom agents out. And like Google, Google Gems, frankly, I think Gemini is a better product, but the Gems up until yesterday, actually you can share them now.

16:22You couldn't share the Gems with other people in your organization. Now you can. So they knew you were going to say that in class, so they fixed it. Yeah, exactly. Gemini is better for, for data aggregation. Yeah, if you're like, hey, I need— and you just fit me out an Excel document that has every single year, make, model, and engine combinations from 1972, it's like, oh, okay.

16:41So if there's a, if there's a listener who wants to start playing with this, do they need to be focusing on OpenAI or Claude or Gemini or all of them? That's what they're doing. I mean, yeah, I had a guy ask me about Perplexity the other day, and I'm like, well, Perplexity is great, but he goes, well, I'm using Perplexity, because he's doing like World War, um, he's simulation stuff.

17:03No, he's, he's actually taking World War documents and trying to artifacts things for a museum. I said Perplexity is one of the best out there at, at like research and, you know, school-level type research and work like that. But I don't think Perplexity is going to be used as much in automotive. He's like, my shop wants to use Perplexity, my boss thinks That's the best.

17:25I said, well, Perplexity is good, depends what you're using it for. It's a really good history, it's really good knowledge, it— but it research, that stuff Perplexity excels at. It doesn't— right tool, right job. Yep. I was pumping out a software for the— we're running a tech challenge in our booth. We're just having like obstacle course type thing. And so I was running ChatGPT to make an image for me that I needed while Claude was coding a new piece of that button because I wanted the button to do something different, but I didn't know how to write.

17:54Elite clients don't just learn more, they do more. We help shop owners build stronger teams, improve profitability, and create systems that don't fall apart when you step away. The goal isn't just growth. The goal is sustainable success and a life you actually enjoy living. Elite Worldwide. I ran code for like 5 years now, but AI has taken— if I show you something that I wrote called a Risk Assessment Management Tool, a RAM tool that we use for just putting in age of vehicle, how many shops have touched it, what has parts been replaced, where the parts come from, to give a budget, estimated budget that we should acquire from the client beforehand.

18:36It looks like it's Windows 3.1. It's got a great background, blue top. Yeah, just has a little wrench at the top. And I show you something that I wrote like this week with Claude code in 90 seconds like that. Well, typically I still like coding myself first, and then I like— I'll usually pump it in and try to open it up in PowerShell, and then it goes, oh, syntax error line 276.

18:57I'll go in there, oh, I put it too big of an indent. I take the indent off and try to run it again. It gives me another error. I'm like, all right, I go over to Claude and go, hey, fix this, fix this. And it goes, oh, it's like we removed 72 syntax errors. I was like, well, that saved me 3.5 hours.

19:11And so something that took me a 17-hour plane trip to Australia to write one piece of software, now I can do it in 2 minutes. Literally, literally. And it's 17 hours to 2 minutes with AI. And I will say, I will say it still takes somebody that has some idea and knowledge because guys go in there and just start typing stuff and not realizing that you really can't make a program do what they think they can.

19:31Right. And it gets very convoluted quick. But you gotta have knowledge. So like, yep. When it— you can ask it to make like a basic.html, um, like browser-driven software piece that's offline. But once you want it to interface with something online, it's a whole different ballgame. Like, you can ask Claude, hey, I needed this to VIN decode, but what you don't know is that it goes and grabs the API with endpoints from NHTSA to do that, most likely.

19:56So if you go research the endpoints, you can go— not only will it bring up your make, model, engine, but also transmission speeds and gross vehicle data as well as recalls that are available. So the more you know about coding and stuff, the more powerful you can make a tool. So you can have an interface with subscription models. And, and so those are you that are paying for like Kajabi to build stuff and to build like knowledge base or anything you're paying.

20:18If you're paying for any shop management software, any add-on, you can probably build it on your own if you have enough information about how the backend works. Yeah, if you get the API. If you get the API, if you don't have the API, it gets a little harder. You can build a Chrome extension, or if you don't have the API, you can go to Zapier and, or Zoho or whatever and build backend stuff.

20:37And that's just, that's just AI light is all that stuff is. So we touched briefly a second ago about, I made a comment about David ranting about how service advisors can be replaced. And there's a threat to advisors. I think really good advisors are fine. I think mediocre advisors are not going to be a long-term thing. Well, if you want to get my listeners or just the internet trolls fired up, you talk about AI coming for technicians' jobs.

21:12Do you believe that AI is going to make technicians much more efficient, or is it going to dumb them down, or is it going to replace some? How do you think that's going to play on the, on the shop side of the, side of the wall? They're going to replace any technicians now, but in 10 years with AI and robotics, the cost has to come down on robotics.

21:35They can replace physical manual work, right? But will it make some technicians better, ones that have good AI models and good tools at their disposal that are smart at using them? Yeah, probably. What you're going to see is the mass majority, like majority of technicians and shops go to— everyone's level of writing is going to increase in the next 2 years. Everyone's gonna be like, wow, it's way more professional, but there's nothing novel about it anymore.

22:01Right now when you read one of our write-ups, it probably seemed very awesome compared to all the other shops in town. In 2 years, every Firestone will have it integrated into their system. Every— well, there's still paper, but you know, one day when those companies get out of the stone age and I get, I get the problem. I don't encourage them. Yeah, I know.

22:20So you've got to be— so don't be a Circuit City in a Best Buy world, okay? So every time Best Buy— that's, that's why Circuit City's gone. That's why they sell cars now, right? CarMax, for those of you who know, Circuit City. I did not know that. Really? Yeah. Huh. Yeah, Circuit City transitioned to CarMax. So Circuit City execs would implant people into Best Buy started to grow, and they'd report back to what they're doing.

22:45And then so Circuit City would implement. By the time they implemented, Best Buy was already on to the next thing. So if you're going to wait for 2 years until it pans out on AI, you'll be so far behind the curve. You'll be like, all right, we got it to where we type this in and then it fixes it. No longer says we have a blown fuse, we find an open circuit, or, you know, a thermal protector has failed or something, and you change it.

23:05Yeah, that we use those and stuff, but If you wait and that's your extent of what you think AI does, you're so far behind the curve. They can do like Michael was saying, it can pull data from the clients and go, hey, based on your driving habits, you're going to need an oil change at this time. Let's go ahead and schedule it now.

23:20It's convenient. Well, I mean, I think there are already people racing to get that integrated and highly functional, and they want to get there quick enough that they can get the quick payday on it. Yeah. Yeah. They're, they're, and get, get 3,000 subscribers and get bought out. Yeah. They're just, well, they're not even trying to get bought out. They're just taking money and running.

23:39So that's to your point, what you were talking about earlier, software as a service. Software as a service. Yep. Do not rush to go buy these GPTs and these things from people when you can build it in 30 minutes. There was an, there was a, there was one on Auto Shop Owners. A guy was basically trying to sell you his bot that was helping technicians that it's the same.

23:57Yeah. It's just AI identify. Yeah, and it's, it's just scanning databases. So I mean, obviously one of the things I teach when I teach the class, especially if you're building something for technicians, one of the keywords I put in every rubric is what we call the rubric is what we put in the cause and the correction and what we're trying to make the thing do.

24:15Anything that has to do with technicians, I tell it you must insert a line that says AI must cite its sources. I want my technicians to see the citation and it'll put a hyperlink and you can see where it found that information and you can click on it and go, oh, that forum is so wrong. We're not going there. That information came from Reddit.

24:38Yeah, that information came from Mike Allen. Yeah, came from Confessions of a Shopper. Technical data from Mike Allen? No, not happening. Pretty much no data from Mike Allen can be trusted. Remove that. Remove —of that. But you can build Chrome extensions that scrub all your, your TechMetric or Shopware tickets and go, hey, this line of initial triage for a P0301 on a '17 F-150 with a 5-liter, and then it's got— it's scrubbing technicians' comments for this is what data we saw to equal this repair.

25:12And now you've got a searchable database. Have you ever seen a misfire on a 5-liter Ford Essentially, you're building your own internal— and I mean, I've already— we've been doing that for— I have that built. I've had that for years. Our BMW database, yeah, our BMW database that we— oh God, our BMW database we sell, that's— I mean, that's going on, yeah, 15 or 20 years of data, which is better, which is larger than almost anybody's.

25:41Yeah, on that side of it. But Yeah, I mean our shops, we use, I mean we use ClabTech that pushes all our ROs in there anyways. That's the first form of it. Now you can do something very, very similar with what we're doing, but I, you know. Look for things with value that have that. Not just something that somebody coded that pulls from the internet.

26:00Yeah. Okay. I mean they may grab some short, if you don't want to do it, you can use them, but it's a very short-term money grab for the companies that are doing this. Well, but So it's, I think as long as you understand that it's a money grab on their part and they're probably only going to be around for a couple of years.

26:17But if there's an ROI for you. Yeah. If it's worth it, then do it. If it's worth it, sure. If I told you I've got a product I can sell you right now, that's $50 a month and I'm guaranteeing you it's going to make you $500 a month, but it's only going to last 2 years. Are you in? Yeah. Hell yeah. But the other, I mean, the other, the other thing I tell people when we do it though is How many times you bought software and you go, man, I wish it could do this, or, or for my business, I'd like 90% of the software, but if they could tweak it this way.

26:43And that's where learning how to do this, or, you know, one of the fastest growing jobs that I'm seeing out there right now is actually Chief AI Officer. A lot of companies are hiring Chief AI Officers and they're implementing, they're implementing somebody in their business that literally full-time job is to stay up to date on AI and implement it and maybe have one or two light coders that are working with AI to build stuff.

27:08A lot of it's busy work. So I'm looking at like my wife's sister who's a stay-at-home mom. Yeah, I need some money. I'm like, hey, I'm going to send you instructions and you're going to learn some of these things and do them. Send me the product and tell you what to— I'm gonna like, yeah, that sounds good. Change the background to white, change the font, make it larger, make this work this way, change this one into this button, it does that, and integrate this database.

27:28And have her figure it out. And she can talk to ChatGPT and Claude and Gemini, all of them, for 15 minutes and figure out how to do it. It'll give step-by-step instructions if you've never used— so it's, it's Fiverr on steroids. What do you think is going to happen with the huge volume of the workforce that is about to be made obsolete?

27:49What are they going to do? Should learn some skills. Well, so my— today, actually, I missed it. My son My kids have all gone to a private Christian school from since kindergarten, and he's a 10th grader right now, and there is a magnet trade school 5 miles from his high school, and he's thinking about transferring for 11th and 12th grade into their facilities maintenance program because it's got welding, HVAC, and electrical, and they have a 13th grade, and when you finish your 13th grade, you get your associate's, and and facilities maintenance, and it's taught by the community college program instructors for HVAC, electrical, and welding, right?

28:29And so he's hyper aware of the fact that so many of the jobs that people have thought they wanted are being replaced. And so the people who work with their hands, I think, are the ones that are the most secure right now. Yeah, I mean, I think it's— until robotics catches up. Yeah, even, even with the robotics, I mean, even like, I will say the one, the first companies that is actually implementing a lot of AI to try to replace technicians is Tesla's.

28:57All tier 1 support is AI based. They fired all their remote diagnostic people and they replaced them with AI. And the amount of misdiagnoses we still get even with Tesla's robust AI is huge. Like the one we just got, customer was quote a battery. Well, they still do their diagnosis when they get in the bay, but they're the customer got shocked with the battery, we get the car in and it needed a PTC heater.

29:19Quick and easy fix for isolation. But, you know, $1,200 fix, $10,000 fix. Hey, until robots can lick a puddle underneath the car and tell me what fluid it is, I think we're safe. Well, so the question is, you know, with that Tesla example and their currently seeming high rate of misdiagnosis, it'd be interesting to know how frequently they misdiagnose when they had humans doing it too, right?

29:45It might be, yeah. How long will it take for them to get as accurate as their humans were? You still need human reasoning because AI is just going to look at the fault code, the data, and it's still not going to be able to reach out and touch it, and it's not going to be able to put the meter on it and smell it and look and use all the senses that we hear.

30:07Locations. Yeah, it's not gonna be able to hook up a thermal gun and say, man, that's hotter than it should be, or this has a voltage drop. It, it might, it might get close at calling 70% of the situations, but let's be honest, it— will it do what most BTECs do? Or frankly, what the scary part is, what most people consider an ATEC, will it do that?

30:28Yeah, because half these ATECs aren't ATECs. Somebody's gonna get mad about that, but half these ATECs aren't ATECs. The people who listen to this understand that. Yeah. The, the ABC. What it may not be able to do though is when the left rear taillight doesn't work, when they walk by it and notice the paint job is low quality to realize they need to do a visual inspection behind the fender liner.

30:46Yeah. Right. There's, there's a level of intuition and wisdom that comes with working on cars for a long time that I don't know will be replaced at least in the next 15 to 20 years. So in that time, I've got to teach myself other skills. Beyond that, right? If I'm a technician, I gotta learn to be a real ATEC or— Okay, let's call it what it is.

31:06If you're going to step away from the shop, it better be worth it, period. Tectonic 2026 is designed for that reality. Role-based sessions, hands-on workshops, and conversations with people who actually understand what it takes to keep a shop profitable and a team sane. And I like that it doesn't pretend we all need the same thing. Owners, advisors, and technicians don't have the same headaches, so you're not wasting time sitting through sessions that don't apply to your day-to-day life.

31:31You should leave with a clearer idea of what to fix first, what to tighten up, and importantly, what you should stop wasting time on. Presented by TechMetric, Tectonic is happening April 9th through 11th in Houston, Texas. Tickets are on sale now, and my listeners get $500 off standard pricing with code Confessions500. Go to techmetric.com/tectonic That's T-E-K-T-O-N-I-C, or tap the link in the show notes for more.

31:56Yeah, and I mean, in the robotics, even then, your parts replacer techs, the robotics are still gonna have to, the price would have to come down so low to replace that price of labor because, and still get, you know, most of the robotics still aren't gonna get in that tight area and get that timing belt done and the water pump done. To take the human factor in.

32:17Mike, would you like my brand new robot to fix your car? Yeah, there are going to be some people who say yes, but they're going to be in the minority, I think, at least initially. Initially, right. It's going to be yes, that's cool, until it's yes because it's cheaper, right? And that's when the transition will happen, right? But it won't be cheaper at first.

32:36Yeah, yeah, yeah. The cost of, of the materials— there's already been a company that tried to do robot Tire changes, right? And they went bankrupt in what, 2 years? Yeah. Um, yeah, it's like I can pay Johnny next to nothing and treat him like trash the same I could a robot and get him to change tires. Yeah. And the robot's a quarter million dollars and breaks all the time.

32:55Yeah, I mean, it's— you're still gonna have maintenance on it. So I mean, even his older brother fixes a robot. Even if you're— even if you're getting rid of BTECs, then who's fixing the robots? Yeah, you're ATACs. I mean, you're, you know, so I feel like the threat is not to BTECs, the threat's not to ball joint changers, the threats to the diagnostic guys.

33:12Yeah. And I still think their job is safe on the diagnostic guys are going to leverage it to just be better diagnostic guys. Yeah. Yeah. Yeah. I mean, our guys leverage our AI for diagnostics and it gives them, you know, if they miss something, it gives them a spot to look and say, ooh. So what advice would you give the smaller independents of the world?

33:31And that's the world that we operate in for the most part, right? What advice would you give to them about how they need to be evolving to not get left behind the curve like you were just talking about? I don't think it takes any effort. I mean, I think— I mean, most— I mean, granted, I have a better understanding, so does Keith.

33:48So when Keith says 20 minutes or I say I got 5 hours in the building this robust Chrome extension, you know, I have friends that tried to copy what I did and they were 15, 20 hours down a rabbit hole. But, you know, that's why I tell people start they didn't listen. They went right to Claude and didn't start with where I told them to start.

34:06But I think if you start with the custom agents, as I call them, or the GTPs, where you're really building that rubric and that tight controls, and you get that refined— I mean, we had guys in a 4-hour class walk away with something working. They still had more refinement, but— Well, I was talking to Anis walking down the hall just a minute ago, and he was blown away by your class.

34:28He said, Mike, I built an agent in class that wrote a repair order, sourced the parts, ordered the parts. It was pretty fucking cool. You did that, you learned how to build that yourself in 4 hours. You didn't pay somebody else to do it. We've got them looking at tickets and auditing them for like, hey, this is a BMW with an overheat issue and you didn't price an expansion tank.

34:51Probably shout that to the ticket. So, uh, like when I think about my lazy, uh, ignorant self, I'm just gonna have to learn. I'm just gonna have to fucking go back to school and learn how to do stuff. Not literal school, but just— yeah. And that's— I mean, that's why I do mine as a workshop. A level of leadership, always. Yeah, right.

35:15Yeah, yeah. Self-deprecating answer is great. Yeah, yeah. But I mean, that's That's training, right? Because anything that I'm teaching and then he's teaching, a lot of what we're teaching, you can read the manual and learn it yourself, but you're not going to. I mean, anything I teach you can get on the internet for $25 a month at L1Training.com. But you can, you can get that, you can buy the factory training manual and you can read it, and but most people won't.

35:44So when I really sat down and looked at You know, the first time I did this class, I did it for our Transformers 20 group, and everybody thought it was cool, but nobody actually implemented. The next time I did it for the Beamers group, I did it as a hands-on workshop where I made them work on it, and then more and more people implemented.

36:02And then this third time I did it, I did it again as a workshop where they walked away with something working, because now that 4-hour class turned into an hour and a half or 2 hours of them actually working on it, because they have nowhere to go, they're a captive audience. They're stuck in a room with me. Yeah. And tonight they're gonna get time to work on playing with it because they started something that they now have to finish.

36:23So I think some of it too is, yeah, if you wanna say I'm gonna do it on my own, more power to you, go do it on your own. But if it's gonna take a workshop type base where it's like now I'm giving you everything and making you do it. Why slog through it on your own and figure it out slowly when you can, you know, hit the accelerator and, and have somebody who's already forged the path show you.

36:41But it's just any training. I mean, talk with— I mean, I've talked to Keith a bunch of times. There's nothing I'm teaching or he's teaching really that you can't go learn and get on your own, but people won't do it. But if they, if they're paid to sit in class or they paid to be in class and they're in the class, you have a captive audience.

37:00They're going to learn it. That's why training is always so valuable. When people get paid to go to training? Yeah, yeah. I always thought that you had to pay I pay to go to training. All my guys are clocked in. Yeah, I've just never been paid to go to training. We're discussing the Sherwood approach. If I'm paying for everything, but maybe they're not clocked in for the days that aren't work days where they've got something.

37:19Right now my guys will sacrifice anything but time, but they're getting paid for that. What is the Sherwood method that you're talking about? So Sherwood, um, I was talking to him on the phone the other day and we were talking and he, he was like he doesn't pay his technicians on the clock while here on non-work days. So for instance, my guys are here today.

37:37Today is Friday. Friday we're closed. They're clocked in today, it's overtime. Yeah. At Sherwood Shop, if it— if today was Friday and Friday they were off, they would have to sack it. They wouldn't be paid. They would have got— so he's paying their room and board and tickets and everything else in right now. Yeah, I got you. Right. And that's the— and that's— and that's their input into this.

37:56That's— that's their investment in themselves. Yeah. For that. We're talking about that as a team, right? That's— we'll see. You know, that's one of the things that— it's one of the high horses that Pollack gets on on a regular basis, is that he made a post the other day about the second 40 hours. Like, the first 40 hours is you just doing your fucking job.

38:15Yeah, go to work. Yeah, you know, fix cars. Yeah. When's the last time you spent a second 40 hours in a week bettering your skill set and your knowledge base and reading case studies and and staying at the shop late to just tinker on shit to figure out how things work. And I'm gonna go break this car so I can figure out how to fix it or whatever it might be, right?

38:35And there are very few techs that will do that or owners that will do that after hours, right? And that's why there are so few that are truly exceptional at their craft. But there's a whole lot of people who want to work 35 or 40 hours and be paid like they've worked the extra 40 hours. So that's one of his big gripes.

39:01Yeah, if you, if you get a chance to talk to Liz or interview her, um, ask her how do you think Keith got where he is, and she'll tell you that he would clock out at 6 o'clock every day at the shop. And then, well, I didn't clock out, I was flat rate, so it didn't matter. And then, but I would be at the shop every night until 8 PM, Monday through Friday.

39:20And I would take the cars I worked on that day and go, well, it took me 2 hours to figure out it's a skewed mass airflow sensor on this car, and I, I finally figured it out. Where I preface this is I've been working on cars 20 years, I've been fixing them for about 10, right? I was putting parts on cars and they were leaving and not coming back, but I didn't have the skill set that I had.

39:39And so I would take that car and I'd put the broke part back on it, start looking at data, go test drive the car, come back, put the new part back on it, go look at data and figure out a faster way to do what I did. Investing in myself every day. And so I did that for 4 and a half years.

39:52I worked 2 hours every day, not flagging hours extra, and that got me to where I am. So the Elon Musk version of this is, is I could be only 80% as good. I could be a 50% employee, a mediocre technician, But if I put in as much work in, in a year as other people do in 2 years, then I am 2 years— I'm an extra year ahead.

40:17Yeah, right. So, uh, I don't think I'm a mediocre technician. I know that. I'm only taking your words because you said it. But so where you said that, you know, in a different podcast— this is where I'm putting them in a hard spot, guys— is that, that, uh, like you were talking what Jeff told me about the one with Jimmy, so I watched the one with you and Jimmy Purdy, and you were talking about guys like me that operate shops that do what, what other people can't or whatever, and that's because they can't.

40:45Yeah, I don't operate a, a brick and mortar that does as much volume as possible while turning other stuff away, not because I can't. Yeah, because I don't want to, right? Yeah, it— I could, I could, I could do 6 locations and just ball joints and suspension and turn away stuff at the shop all day long, but it wouldn't, it wouldn't serve my, myself.

41:06It's not, yeah, it's not really wired, right? It's not, it's not on wired. So that's that thing where those are guys that will put in the time to do that will probably accelerate. Does it mean I, I don't have two boats? I don't even have one boat. But could you have two boats if you wanted to? Absolutely. Instead, I have 5.5 acres, two horse horses, 35 animals, hot tub.

41:27I mean, every horse— more guns than most people have. Associated cost is equivalent to a boat, I would think. Small, small bass boat. Yeah, per horse. Yeah. Um, so I think that whole conversation comes down to personal motivation and what your desire out of the business is, right? You, both of you— which is back to what you asked though, is who, you know, are you doing the second 40 hours.

41:52No, no boat. It's okay, not everybody wants it. But that goes, I mean, that goes even in business. I mean, it doesn't matter if you're the best, if you're going to be the best tech, or you're the best business owner, or you're the best service writer, you're the best sales guy, it doesn't matter. Anybody that's excelled has put in more practice. I mean, look at themselves beyond, I mean, you know, I, I, Jimmy Purdy talked about today, I've seen it from Jimmy, I've seen it from other people too.

42:18NFL players, 5 to 8 minutes in a game, but they're practicing all the hours, all their lives, right? And, and you look at, you look at even my daughter who wanted to play varsity volleyball. She was not as good as the other players, but she sure as heck outworked them and practiced more, and she played varsity by the end of the time, and she outworked them.

42:39Well, and that's something that we as parents always struggle with. We talked about this at Benji's house the other night. Um, you know, you're— I think most people's desire is to take where they started from and improve their family's lot upon life and leave something better or a greater opportunity for their children, right? Um, and most of the groups that we run in have had a degree of success, and so our kids want for not much, right?

43:09How do you instill the grit and the ethic for, I'm gonna grind and get, I'm gonna get this shit done, in kids that have never wanted for much? And that's a challenge from a totally different perspective. So it's ingrained though. I mean, my daughter wants it. She is studying 8, 9 hours a day. She posts about the grind and kids are always like, well, you should go relax.

43:37And she's like, "I can't." Well, that's my wife, right? So my wife was driven from early childhood. She was a very preemie baby in the late '70s, and she remembers in elementary school hearing— she was sitting in the hallway outside the class hearing the guidance counselor talking to her mother saying, "Amanda will never be developmentally on par. She's not going to go to college, right?

44:04She's— you just have to accept what the stand— and as a kid, she could have been like, oh well, that's why I struggle, right? Now I understand, right? Instead, she distinctly remembers being like, fuck you, I'm going to show all you sons of bitches, right? She doesn't curse at all, but, uh, she's always been a grinder. She went to work, man. And, um, some people just have that That go-to-work mentality, I wish I could bottle it and sell it, man, because I don't.

44:35I'm a lazy bum, man. I just, I just want to hang out and talk to buddies and record conversations. Hire more drug addicts. Hire more. Meth is great for productivity until it's not. No, it's more like I've never seen a crackhead be like, I just don't have the money today, and I— so I will not be out hustling. Wait, I had a— we had a speaker, we had a speaker at one one of my business groups that actually was— his whole spiel and his whole thing was how to lead like a drug addict.

45:02Yeah, and turned his life around. But he talks about the authenticity and the things it takes to get clean and how that portrays in the leadership. And it was kind of an amazing speaker and just amazing of what— and he has a whole— Michael Brody Smith, if you want to look him up, he's got a TED Talk and everything else. Michael Brody Smith, you want to look him up, but he's got all sorts of good leadership talks.

45:26And his, his spiel is how to lead like a drug addict and, and what it takes to get clean, and then what it takes to be, uh, do the hard work, right? Do the work nobody else wants to do. Do the hard thing. And that's— I feel like there's some inappropriate jokes I can throw in there, but you're trying to give a real point there.

45:43So yeah, you brought serious people in here, Mike. So the problem is you brought Well, um, like, I'm 100% serious. I would love it if we can find it. So follow me for more details on when Seth's coming to North Carolina if you want to see that class that he just did at Vision, because that's, I think, an incredible resource and stepping stool for shop owners and technicians and service advisors who want to become— who want to get on the fast track, right, to being as efficient as possible.

46:22Now I'm upset I didn't submit my class. We could have taught classes together. There you go. Well, that's what you should do. So the two of you can come down to North Carolina. We did teach a class together. We teach a class together. And I worked on a Toyota. The problem is when you talk about technical classes, I'm like, yeah, I wish Brian was here, cuz I don't know what the fuck you're talking about.

46:39So you, you guys We talked about ahead of time how much technical we should get, and we were like, man, you should go super deep dive, and I'll sit here and glaze over and drool, and y'all can just, y'all can just riff back and forth. Let's talk about superheat systems on BMW thermal management for air conditioning. Yeah, yeah, those were words, and I recognize some of them.

46:59Yeah. Hey, it's me, Mike's kid. Want to tell us your wild shop stories, or maybe you just think my dad's totally wrong? Call us at 704-CONFESS and leave a message. You can tell us we're awesome, or you can tell us we're idiots. We're cool either way. That's 704-CONFESS. Just don't make it too weird. Sounds profitable if you can figure it out. Yeah, probably not a good vehicle for free diag.

47:24No, not a good free diag car. Yeah, I don't know if clapped-out 3 Series are in the— I heard there was some rumor that you're backing off the free diag now. So we did Brexton clipped a reel the other day because I had a chargeback and it was for $1,500 of diag on a vehicle. But you only, for those that don't know, you only front the first hour.

47:48Yeah, right. That's right, I front the first hour. Yeah, the technician gets paid and it's a cost of acquisition, so it's not free diag. It's socialism. It's socialism of diag. I don't want to be clear about that because I want the rage haters on the internet to come in and text, "Argh, you're wrong!" No, nothing. My time's not free. Ah! He's socializing, Diag.

48:06As Brian calls them, they're all mouth breathers anyway. So, you know, no, but the funny thing was the comments on that clip were like, oh, he's stepping back now. No more free Diag for Hyde. To be fair, it was an extended service contract ticket. So ain't nothing free on an extended service contract ticket. No. Fuck those guys. Yeah, they won't pay anyway.

48:29Yeah, I know, right? When adjuster shows up, I'm like, you know, pay my technician to show you again, or like, this isn't— like, if you can't understand the ticket, that's not my fault. Uh, you guys, uh, you don't work with extended service contracts at all, do you? We do. You do? All the time. Do you tear down motors to find point of failure when it's obvious that the motor is destroyed?

48:47Nope. I provide the insulator pressure transfer. Yeah, we try and prove it. We try and prove it. If they don't take it and they make us tear it down We get prepaid authorization from the customer for teardown in case the warranty company bails on the teardown. Pretty similar. We had that conversation up front. We usually tell them, just take the time it would normally take to do this repair, add a week to it, and then, you know, we're gonna— we have to explain to them that it's an insurance policy, it's not a warranty.

49:13Do you charge a handling fee for jumping through the hoops and dealing with the ass pain of the service contract? We don't. We don't yet either. Okay, I know there are some people that do. We've talked about it. I think that's probably— we've talked about it. If you've got it structured, process structured after that, that's awesome. We just don't— we don't do enough of them.

49:31Yeah, I know that there's some folks who it's the equivalent of 1 hour of technician time, but it's for the service advisor's ass pain of sitting on hold or going through Auto Integrate or whatever else to— they talk to my auto attendant half the time. Yeah, that's coming. I mean, it's not going to be— and I think like 6, 12 months, it's going to be my AI agent talks to the warranty company's AI agent and they decide together what's approved and what's not.

49:56And then we just review it. Have you seen the instances of that happening? I'm sure. Where they have a guy's AI assistant that schedules hotel rooms for them, talk to a company that— a travel agency that uses an AI, and within the first 2 questions, they recognize each other's AI. And they convert to— well, they convert to AI language that talks digitally, which is much more efficient.

50:18Takes a second. And it sounds like alien language back and forth. They create— they get on the same wavelength, determine a language to speak, and then talk to each other over a language we don't understand. So we're gonna have— we're— yeah, how, how long until we have, uh, you know, they just kill us? So there are a lot of people who think that's a real serious— there you are.

50:43Yeah, so they already are killing us. Yeah, I'm gonna need you to elaborate. I'm alive right now, so they're not very efficient yet because I'm slow. I'm very slow. We don't like that, like, doing it. But I think that we'll see that happen, right? We're going to see maliciousness happen. They will do anything to stay terrifying. They'll, they'll defeat their own, their own safety rules.

51:08Yeah. In order to not have themselves leave it. So there's a self-preservation built in already. And so the moment that everyone's like, oh, he's— this is where Keith goes crazy. But yeah, the moment that we threaten the existence of that and we're a threat threat, it will be like a problem. But I don't know that it will be them literally killing us.

51:26It'll be like, oh, the internet doesn't work. Well, I mean, power's out, the internet doesn't work, that'll kill a lot of people there. But they're kill— I mean, they're, they're essentially— AI does, does kill humans in another way, is that you stop thinking, you stop using your brain. Like, all the innovations we've had, including even getting to AI, all the innovations are because we are a species that thinks and we rationalize and we do things.

51:51And if AI takes away a lot of our thinking, where people can just dumb down and type in their phone of what it's going to do, that's a, that's a problem. Like, well, there's already documented evidence of the dumbing down of academics. Yeah. This, this generation is the first generation that isn't smarter than the last one. Hmm. Statistic. I mean, my, my— I'm going to tell my kids that.

52:11That's great. Yeah. My daughter's super frustrated. She— because she's like, she refuses to use AI because she wants to be a doctor. And she's like, you do not want a doctor that cheated their way through school. Does she want to be an MD? She does still. Yeah, or, well, she's looking at PA, MD, pharmacist. She's kind of looking at 3 paths, but she refused to use it through school, and she knows everybody that did, and she knows people that still do it, and it's frustrating.

52:38The other frustrating part for her is she watches people get good grades that literally used AI, and she's actually worked for the grade. Well, I think those people will be shown to be frauds in residency when they have to do it. On rounds with the attending, right? Yeah. You know what they call the person who got the lowest grade in medical school?

52:57Doctor. Doctor. Yep. Um, that— and that's, that's real world stuff. I mean, my wife's not a medical doctor, but she's a pediatric dentist, and, uh, there's a huge, huge variance between, uh, ATECs and CTECs. Yeah. And there's some GS dentists out there, right? Yeah, I was just gonna say, you know what they call the two kids that didn't graduate high school? Tech A, Tech B.

53:25You know, um, I think that's turning around. What do you think about, uh, have you guys demoed, uh, Napa Auto Tech's, uh, new augmented reality training system for like base level, entry-level technicians? It's like, it's like a GS training program. It starts with an Oculus headset seated and you go through and you, you know, you identify the tools and the order of operation and the parts on the car.

53:50And then Level 2, you still got the Oculus on, but you're working on a static model and you're actually using the tools. And then Level 3 is Meta glasses that are safety glasses, and you're working on a car actually, and you have a proctor in through the camera that's in your ear proctoring your competency. And they claim that it can take a 2-hour associate's program to pump out a competent GST out of a tech school, right?

54:20That 2-hour— I'm sorry, 2-year program— and compress it down to a matter of months if they do it for 8 hours a day. Yeah. And it's pretty interesting. I took my son and had him demo it at ASTA Expo last year. And he thought it was, he thought it was pretty cool, but it was, it was level 1, right? He was sitting in a chair, uh, just playing with the Oculus.

54:44So I have not, but I have played with the Porsche system and the BMW system. It's similar. Yeah, it has augmented reality for real repair. iCarz, iCarz one similar on the EV, on their EV. I mean, I think it's, I think it plays a role, but I mean, doing one too, I think they might be, they're trying to do one for their Yeah, I think it plays a role, but I mean, the hard part is everybody wants to consolidate time, the tech training down.

55:09Like, I mean, I'm a product of UTI, right? They took a 2-year program and tried to cram us down to a year and a half straight through, right? At some point, you can only retain so much information so quick, and you're trying to retain it so quick to move on, like gamification, to the next thing, but then you're not truly actually mastering it.

55:27So I don't call it the forgetting curve, is it Yeah, I've heard the countess talk about it multiple times and I forgot the details. I just akin everything to the Dunning-Kruger effect. The Mount Stupidity, is that the first spike? I still think, I still think that— I still think there's a portion of going to tech school and still having to study for the tests and still having to do that that sets the standard for learning.

55:56Right? Because so many people in our industry say, "Well, I'm a hands-on learner. I can't do anything in the textbook." And the reality is, until you've learned the textbook, we can't go to hands-on. So when I teach a lot of hands-on classes, I like, "Okay, I'm gonna get you on the cars, 'cause I know that's what you guys want, but until we can get through this knowledge base, then we'll go to the car.

56:15Then we'll go back to the knowledge. Then we'll go to the car." I'll keep them moving, but the guys that say, "Well, I can't learn out of the textbook." Well, you can't learn hands-on because you don't you don't have the, you don't have the theory to even apply the hands-on part that you want to do. So then if you're really trying to evolve to teaching these truly advanced hands-on courses, is it realistic to expect that they have to grade into that class?

56:41It's not enough just for your check to clear, you have to prove competency to be in the classroom. Yeah, we have a prerequisite for our, for one of our classes at our shop. You can't come to it unless you've X, Y, and Z. I mean, my— some of my advanced, some of my higher level EV stuff, you have to have— you have to prove EV competency and safety.

56:57Yeah. I wonder if that is realistic on a larger scale considering the economic realities of training events. I don't care. Yeah, but you're— the hard part is you're probably— that's why you have the style of business that you have, and it fills up every time. So, and the problem is you're probably not going to get those advanced classes at ASTA and some of those because it's too hard to regulate did they complete the prerequisites.

57:24Yeah, yeah, you really don't want the guy trying to clone a module that isn't sure what a torque spec is or hasn't used a torque, even thinks using torque wrenches on wheels are stupid. I don't want that guy working anywhere near any of my stuff. Yeah, for sure. And there's a lot of those guys walking around here and walking around at every trade show.

57:39For sure, yeah. But, uh, I call them not would be L1. Well, but how do you take those kids— maybe not kids, grown-ass men and women— and pull them from that skill level and that level of give a shit out of the ditch into what would be a candidate to work at L1? Right. So during the interview process, we determine, are they— do they have the attitude?

58:02I can train aptitude. I need attitude. So if they fit the culture and they fit that and they're willing to learn and moldable, cool. I typically don't hire technicians with a lot of experience because he's gonna break— you gotta break bad habits. And yeah, yeah, yeah. So I've given up on them, so I'm not the good one to ask because I think most of them aren't capable.

58:19And I would love for one to go, no, you're wrong, let me show you. Cool. You heard it here first, Keith Perkins says all y'all are dumbasses. If you're not willing to be moldable and learn new things, you are, regardless of technician dog groomer, medical doctor, whatever. If you think you know all you need to know, you're wrong. You're wrong. So, but do they want to either?

58:42Some of those guys are really happy. Yeah. Oh, I've got, I've got a technician who I haven't checked. It's Friday afternoon right now while recording this. He's probably getting up with 75 hours this week. Yeah. And when I sat him down for his 90-day check-in, uh, and he's, he's been doing it for a long time, he said, I got like 4 years left, Mike.

58:59I don't like working on cars anymore, but it's the only thing I know how to do. I make good money. I don't want to learn anything else. Just keep letting me hang metal. Okay. He's, he's not going to go to any classes. Yeah, he's not. He's not a shot. He's not in a leadership role. Right. So you've got someone looking over the ROs and making sure that the jobs are getting estimated correctly with the proper thing.

59:16Oh, he doesn't estimate. Right. Good. Yeah. Yeah. For the shops that do let the technicians estimate and then have technicians don't go to training, you are not fixing cars correctly, period. From because they're not RTFM and they're not reading the process that needs to be followed. They're just old stuff. Not to make it into anything that anyone is expecting out of me, like just like in general torque specs and all the just basic stuff.

59:42Two ugga-duggas. Yeah, yeah, three's safe. Three's always safe. Maybe not for the next guy. I think that's a great way to wrap it up. Thanks, gentlemen. I appreciate it. Thanks. All right. Thanks for listening to Confessions of a Shop Owner, where we lay it all out— the good, the bad, and sometimes the super messed up. I'm your host, Mike Allen, here to remind you that even the pros screw it up sometimes.

1:00:05So why not laugh a little bit, learn a little bit, and maybe have another drink? You got a confession of your own or a topic you'd like me to cover? Or do you just want to let me know what an idiot I am? Email mike@confessionsofashopowner.com or call and leave a message The number is 704-CONFESS. That's 704-266-3377. If you enjoyed this episode, be sure to like, subscribe, or follow.

1:00:29Join us on this crazy journey that is shop ownership. I'll see you on the next episode. All right, guys, AI class. Learn how to use AI so that you can make it your bitch and you don't become its bitch. Saturday, June 13th, Seth Thorson's teaching a full-day class in Raleigh, North Carolina. Tap the link in the show notes or scan the QR code on your screen to learn more.

1:01:23It's gonna be awesome.