Latest / Auto Collabs / The Reason Your Dealership Tech Stack Sucks (It’s Not What You Think) with Nick Askew

Transcript

The Reason Your Dealership Tech Stack Sucks (It’s Not What You Think) with Nick Askew · Auto Collabs

• 0:00
  You know those situations where you go and you get a couple friends together, and you realize- [singing] This is Auto Collabs...
• 0:12
  two of your friends have, like, this connection and something in common, and they just- [laughs]... like start to talk about it? It's how I feel like people are around me when I meet another Eagles fan. [laughs] Right?
• 0:22
  Oh, that's the worst. And then, see? That, that's the feeling I'm talking about. You're like, "Oh, I'm gonna go- [laughs]... I'm gonna go find the mini donuts," right? Well played. Time for me to find...
• 0:28
  Well, that's kinda what it's like when, the last time I was with Kyle and Nick, and they just started, like, nerding out over things that I would, like, nerd out over the Eagles.
• 0:37
  So I think, you know, I, I think I un-volunteered myself for this next conversation. [laughs] Not because I don't like Nick. He's one of my favorite people. Every time, though, it's like I f- I, I feel you.
• 0:46
  It's like, it reminds me of my first day in the Philippines where everyone was just blah, blah, blah, blah. And I, I was like that. I was like, "Uh." [laughs] "Uh." They're not that bad.
• 0:56
  And when you look at, you look at both of them, you look at both of them, if Kyle wore glasses, he'd be this guy. He'd be the one that's like... [laughs] Sitting, biting on one, biting on one of the sides.
• 1:05
  [laughs] Well, hey, look- And Nick would be smoking a pipe. [laughs] Like a cork pipe. I think, I think I'm just off to go hang out with Nick. See ya. All right? We'll, we'll see you guys later. Call us when you're done.
• 1:18
  [laughs] [upbeat music] All right, Nick. So we're just gonna jump right in. I am constantly paying attention to your social profiles right now, whether it be Instagram or LinkedIn or anything like that.
• 1:33
  Just so you know, just so you know, I'm following, I'm listening. [laughs] You, you are obsessed right now, and I'm gonna use that word, and hopefully you're okay with it.
• 1:45
  You're obsessed with where AI is taking us in, like, a bunch of different channels, whether that be generative search or AI coding or, like, anything that AI is doing right now.
• 2:00
  D- it, was there a moment that you can go back and say, like, "I realized in this moment that this was gonna be a pivotal shift that we were making culturally, not just in our industry"? Yes. It was November 2023,
• 2:19
  and
• 2:22
  I first started using the client-side ChatGPT to do a bunch of marketing work and charge the same amount of money that I was charging specifically before [laughs]
• 2:41
  ... to get, like, 10 times more work done. And I, I, I realized, I was like, "This has got a very limited shelf life before everybody else catches on to this." [laughs] Yeah. Um, so yeah. No, no, it was, you know,
• 2:56
  a, a time in which I felt like, you know, we've always wanted the, you know, end-all be-all, uh, intelligent solution that knows everything and can do everything.
• 3:11
  And, you know, even early iterations of, uh, AI when it started getting serious, even like Midjourney for generating images and ChatGPT for an LLM.
• 3:22
  Like, when I started seeing multiple companies and multiple, m- you know, multimodal, uh, pieces of AI start to really get some serious traction, and it'd be less of a, um, you know, a, I guess
• 3:41
  a, a party trick and more, "Oh, actually, this has a real business use"- Mm-hmm... that's when it really clicked for me. Yeah, 'cause November '23 was when, uh, it was when GPT-3 hit, right?
• 3:59
  Yep. Uh, 2.5 had been available to some beta users, and then they gave GPT-3, they made that available generally. It wasn't login. Didn't, you didn't, there was not even a paywall behind it, if I remember.
• 4:10
  It was just, like, generally available as a free application initially. And we realized, like, oh my goodness, this large, the, like a larger context window, a larger ability for people.
• 4:23
  Like, and it was, I think it was, like, 16 gigs of data or something like that was all that it was trained on. Is that, i- is that right? Do you remember what that, that was- Oh, it wasn't-... back then?...
• 4:37
  it wasn't a huge amount of data, and it was only- Right... up until, like, November 2021 or something like that. So there was, like, a- Right, yes... a context window to where it was a definite hard cutoff.
• 4:50
  Um, so, you know, you, you couldn't ask anything about, like, current news cycles, about anything predictive. [laughs] Uh, it was just- Didn't have access to the internet, anything like that. Yeah, d- yeah. Yeah.
• 4:59
  It had no search, but it, it, you know, it could synthesize, right? And that was it. It was, it, y- things could be synthesized with just us interacting with it. Right. Um, you know?
• 5:17
  S- so we're less than three years from that. We're less than two years from that. Golly. It's o- just over 18 months from when, like, a, an LLM was generally available to the public.
• 5:30
  And the pace at which things are changing now are almost breakneck. It's almost impossible to stay up on.
• 5:38
  How are you, like, running a business and staying up on the breakneck pace of this and also then dragging that back into your business? Yeah, it's t- it's difficult, and it takes a lot of experimentation.
• 5:54
  I mean, I, I think enabling a lot of people internally, uh, to go and spend time exploring, uh, and solving problems in different ways. I'll give you a couple of examples.
• 6:06
  Um, you know, developers-Uh, using things like, you know, it starts with using Cursor, and it starts with using, you know, other pieces of software. And for people that don't know what that is, Cursor is a...
• 6:20
  It's like ChatGPT for code, right? It's your co-pilot that looks at your code base. You can ask it questions. You can ask it, "Hey, QA, you know, my, uh, new feature here and help me make it better."
• 6:33
  And you gotta realize that a lot of people, it's not just a cultural shift that we have to, uh, adopt, and, like, technology's great and all, but you've got all of these people that are fighting for their jobs.
• 6:47
  So, a- and you, you've got a lot of people that say- Mm... you know, engineers going, "Oh, no. The AI can't do a great job at that.
• 6:54
  They can't do as a good a job as a, as a real engineer because it'll, you know, hallucinate, and it writes code that we're gonna go and, uh, you know, have to go and fix in the future," until they have that point to where they realize they can use it as a tool as well.
• 7:10
  So- Hmm... in- internally with marketing teams, with, uh, development teams, with, you know, everybody internally, I think the first, uh,
• 7:24
  hurdle that we had to jump through was, like, a n- no man left behind approach, right? To where everybody had to realize that they could improve their lives by utilizing, uh, AI tools in any capacity.
• 7:40
  It wasn't there to replace them. It was there f- for them to solve, like, three months worth of problems in a night. Uh- Yeah.
• 7:47
  So- I, I almost imagine it's like, it's, it's like Google came around, and if there was, I don't know, let's say a marketing manager that was like, "Ah, Google's here.
• 7:57
  I'm gonna lose my job, 'cause now my general manager can search everything that has to do with marketing," right? Yeah. Said no marketing manager.
• 8:05
  And I think that there are obviously, there are people, there are elements of job loss out there in the world because of, you know, efficiency or, or ways of working, um, that are a lot different, whether that be with engineers or others.
• 8:21
  But by and large, people are leveraging this stuff as a tool as opposed to a replacement. Um, now, maybe that changes over time. But it's, it's, it's definitely a tool to be wielded and, and used really, really well.
• 8:36
  I'd like to focus it now on, 'cause you're developing some specific AI uses for end users in the tools that Space Auto's bringing to market.
• 8:46
  How are you thinking about, like, guardrails for data and structuring your, the data so that the AI tools are relevant and useful for someone that's sitting in a showroom or sitting in a BDC or, or, or working with a customer working a deal right now, uh, and not overwhelming them, but giving them access to something that's actually relevant to their job?
• 9:08
  Yeah. And unlike the guy that called me out on LinkedIn the other day and just said, "Oh, you just propose everyone just throw all that PII into ChatGPT," um- [laughs] Just, just kidding. Um- [laughs] N- no,
• 9:22
  that's what, that's just what I'm telling everyone to do nowadays. Um- Right. Yeah. Yeah, of course. So- Yeah. Into the free plans, right. Yeah. Just, yeah, into the free plans.
• 9:30
  Give them all your customers' da- uh, make sure to include socials. Right. Side note, I'm not being serious. Just [laughs] Very not serious here. We will- Very not serious here. We will not clip that out, I'm promising.
• 9:40
  [laughs] Yeah, don't clip that. [laughs] Just take that one clip. Um- Yeah, right. Uh, so, you know, for us, it's always been about, okay, taking an experience that somebody already needs to go through
• 9:55
  and taking the monotony and the manual entry out of that. And you think about, you know, why integrated systems exist in the first place.
• 10:06
  It's like, because nobody wants to submit a credit app online and then come to the dealership and do one again, right? Mm-hmm. So then we have integrations, and we, we try and solve that with processes. But
• 10:19
  what we're trying to do is take all of these pain points, and for a long time we've been... Like, dealerships hate CRMs. Let's, let's be real about it, right? I think actually the- They hate the process...
• 10:30
  the prop- the population by and large hates CRMs. Yeah. Like, I've never heard, like, an industry partner rep be like, "I love logging into Salesforce. My favorite thing of the whole day," you know?
• 10:39
  I think, like- Yeah... let's just generalize it, you know? I, I, yeah. Uh, let's take dealerships out of the equation. Um, we use HubSpot internally as our CRM. Sure. And I l- love my HubSpot rep to death.
• 10:53
  Uh, but at the end of the day, like, everything's a chore. Right. I mean, it really is. Like- Doesn't matter. Yeah... keeping up with tasks. I mean, it's just... And a lot of times it's, it's internal marketing, right?
• 11:04
  It's like, "Look, look, boss, I did all of my stuff for today. I made all of my phone calls, and I completed all my tasks, and my KPIs are this." Right. [laughs] Um, but,
• 11:14
  you know, it, what it's created is extra workload for us. So I, I, I take... My approach is take these things that we know have to be done anyway, right? Mm-hmm.
• 11:24
  And use AI to automatically give someone a cheat code to getting to the next step without manual input. Because think about this, right? What is slowing us down in devices that we use?
• 11:41
  It's our stupid little fingers- Right... whether it's on a keyboard or a mobile device. Like, we've still gotta go through user experience, and we've still gotta load the deal, and we've still gotta set the appointment.
• 11:51
  So I just take things like that. Like, okay, I'm on the phone with the customer, and we have AI transcribing the call, and the customer says, "I'm gonna be there tomorrow to look at that, you know, Ford F-150.
• 12:07
  See you then." There is no reason we do not have the data to automatically set an appointment-... automatically put it in the CRM- Yeah...
• 12:16
  l- log it in there, make a note in the CRM, send an appointment confirmation to the customer. All of these things, they get done manually right now. Yeah. So it's, it, there's no reason that we can't. Yeah.
• 12:33
  So I'm just giving one example. They hope- We'll call it, like, well, they hopefully get done manually. But the auto- Well, they hopefully get done manually, right. Yeah, exactly. So, so- Yeah...
• 12:41
  that's, I mean, that's what you take is, is you take jobs to be done that you want to be done in a really efficient way, and you want to be done in a way that is, um, that takes some of the pain out of the manual input.
• 12:57
  Mm-hmm. A- and just give us the time back in the day again. And whether that's, you know, background agents doing admin work in the CRM, or it's, you know, to do with sales and marketing.
• 13:08
  Whatever that approach is, you know, we're just looking at the jobs that need to be done in a dealership regardless, and seeing if we can take some of the pain out of having to manually do it.
• 13:22
  Because I think in technology in general, we have now reached the pinnacle of what we can do with classic computing and manual input. Mm. Right? That, that's it. Interesting, yeah.
• 13:33
  The next phase of that is, well, what can we do without manual input, just from the data that we have? Yeah, it's almost as if we've developed so much so quickly that we now...
• 13:49
  Well, one, also I think just, like, the, the volume of things that we are personally asking ourselves to do in a day, and that our devices and our digital interfaces are asking us to commit to on a day, right?
• 14:03
  Like, just think about the number of emails, the number of tasks you have in the CRM, the number of distractions. Like, the, the inundation of required task load on an individual's- Yeah...
• 14:15
  brain is so high, and it's because of the speed of development. Oh, anybody can spin up an app. Oh, anybody can spin up a website. Oh, I, I've, I need to buy this. I need to buy that.
• 14:25
  I need to communicate with this person. I have to, I have a text, a WhatsApp, and a, and an email that I've got to communicate with. And so it is.
• 14:31
  It's like, okay, now we have to figure out systems and processes to get the lower level things handled so that we can do the higher level, you know, interpersonal tasks at a much greater percentage of our day, right?
• 14:46
  Yeah.
• 14:46
  And spend our time doing the things that actually matter, because I think that's, you know, for, for anybody, whether you're in a leadership position or not, it's the general manager manually signing checks on a Friday, you know.
• 15:00
  Yeah. That, that has to do that. Or, or someone that is, you know, manually going out and, you know, ch- checking and con- you know, sending a text to confirm every appointment.
• 15:10
  Like, yeah, there, there's, there's a lot of these things that we've just become executors of tasks and processes that just ta- that... And, and it gives us anxiety. I mean, I look at my notes- Oh, for sure...
• 15:22
  my texts and my emails and my CRM and my, "Oh, I've got to follow up with Kyle because he needs some notes for the..." Like, all, all of that is, it causes people to try and, like, efficiency hack, I guess.
• 15:36
  But the one true efficiency hack for me is to not actually have to do that in the first place [laughs] you know? Right. Yeah. Yep.
• 15:44
  I wanna, I wanna h- go back to a con- we had, we started a conversation, or maybe it got started for us back on podcast that we did, and I think Michael and Paul were with us.
• 15:54
  But there was this, like, rabbit trail that we almost went down around whether kind of like any, with any tech, and, and obviously you're building tech that has more of a one-stop shop feel.
• 16:06
  But whether the one-stop shop or kind of the bolt-on, you know, API integrated approach is, is kind of the right way to go.
• 16:15
  Can you give me a little bit of insight into your thoughts on why you're building one way and, and why maybe you see that as, as the best way for this industry to develop tech?
• 16:30
  If I recall, you and I have very potentially different views on this. [laughs] Yes. I think may- maybe the conversation was steered away because of that, so. Maybe. [laughs] Yeah.
• 16:41
  [laughs] Well, and we can go into that, but I'd love to hear your perspective, and then, you know, I think my perspective is, is a bit different. But, you know, with some context. So.
• 16:51
  Yeah, no, and I'd love to hear that too. So I'll, I'll give you my, my quick view of this, you know. So when we say one-stop shop, we're talking centralized infrastructure, right? Um, for me, the
• 17:09
  number one pain point for any customer experience, or, and, and when I say customer, I mean a dealer customer or a dealer themselves, right? Using one of our applications. Any user experience is just unmet expectations.
• 17:27
  So that's a big frustration, when expectations aren't met. Mm-hmm. Um, so when I have a thought to implement a centralized data structure that has everything,
• 17:43
  website, your marketing, your deals, your desking, your, all your SMS, every, tracking every single phone call, uh, your, your DMS pulls, like, everyth- like, your real CDP, right?
• 18:00
  And that, that's a actual centralized platform.
• 18:04
  What you're able to do is so much more flexible, because you're allowed to activate people and activate tasks and, and, you know, communications with customers or do certain things all across the funnel.
• 18:20
  And there's a lot of great companies that do-Some of those things very well and have very great policies when it comes to API integration and, you know, they, they wanna work well with others.
• 18:34
  In fact, one of our core values at Space Auto is there's room for all of us. Yeah. That is a core value, right? Now, you have those companies that that is not one of their core values. Right. Right?
• 18:49
  Hoarding data is their business model, right? So integrating with them is a pain.
• 18:56
  And then everyone points fingers at each other and say, and says, "Oh, well, you know, Space Auto can't do that because they don't have access to this data." At the end of the day, I believe it's the dealer's data,
• 19:08
  not any vendor's. Right. And if we can create an experience to where, one, we put all of that data in the dealer's hands and they can choose how to activate it. And two, you've got that instant connection.
• 19:24
  And you gotta realize, I'm building AI, so what does AI need? It needs- Structured data... the most context- Yeah... structured, one source of truth data.
• 19:34
  That means if you call me aft- and then tell me, "Hey, I'm gonna be there tomorrow to look at the Ford F-150," and then you come onto my website and you put in a lead on the, uh, uh, new Chevy Corvette instead,
• 19:49
  I need to have the context, all of the context, to be able to do a really good job. Yeah. That's why my view is, is, you know, one-stop shop, as you call it.
• 20:01
  I think of it as centralized inter- infrastructure where we're happy to play with others. But in order for me to do that, it's like your lobes in your brain, right?
• 20:12
  You can't have your sight be disconnected from your, you know, uh, your muscle memory and your, you know, the, what causes your limbs to move, right? Because then things are just gonna be a disaster, right? Yeah.
• 20:25
  So we need all of that in one place for it to work, so. And I think, I think we probably see this in the same way a bit more.
• 20:34
  But because, like, my perspective here comes from, like, trying to gather the troops across the industry a bit, is- Mm-hmm... is that,
• 20:45
  I agree, centralized infrastructure, like especially if you're a core tech, you should be able to have a centralized look at the data and the way that you're bringing in it, right?
• 20:57
  Like, if you're a CRM, or if you're a website platform, or if you are r- really, the CRM and the DMS.
• 21:04
  If you're one of those two, you should be able to have a data structure that can consume a lot of data and leverage multiple endpoints to h- to house that data, and then expose it when authentication is, uh, is appropriate, right?
• 21:23
  And this is where, like, I guess my pie-in-the-sky mentality is much more this, what, what you said, like, we wanna play with others.
• 21:31
  However, we haven't culturally, in our industry, built systems and structures in which playing with others is possible, right, really, really well.
• 21:41
  There's a few vendors, there's a few out there that can, that can h- make that happen, right? And, and maybe it takes these integrations. But, like, I think about a HubSpot or a Shopify or a Salesforce.
• 21:53
  Like, the, the core infrastructure is built on this app integration, right? It's built on, "Hey, we're going to, every time we release a new endpoint, it's gonna be available on Make and Zapier," right? Um- Yeah...
• 22:06
  it's, you know, all those types of things.
• 22:08
  And so I think if, if that's able to happen, and I, I agree, like, there are, there are people that are, like you, should probably build more of the things right now because that doesn't exist, right?
• 22:21
  So I appreciate what you're doing. I think that it's, I think that it's, like, it's really hard work also, because when you have to be great at all these disciplines, it's actually really freaking hard, right? But, um,
• 22:35
  I'm ch- I continue want, to wanna challenge our industry to think like the way these enterprise organizations really think, in the way that they build and systematize their structures for data and APIs and endpoints.
• 22:50
  So that, in the future, or now, others can plug into that when authentication is appropriate based on the dealer's wishes, right? Based on the dealer's wishes. And I, I, I... So you and I agree on that point.
• 23:03
  You, so let's talk about ADF XML. [laughs] Right? Well, well, okay, okay. Okay. Standardized. Correct. Everybody agreed on it. There, there are good things about ADF, uh, XML. They're a good thing.
• 23:20
  But do you know what the good thing about ADF XML always was? Is that everyone got together and agreed, "This is the way that we share data." This is the way. Yep, exactly. Nobody's done that for APIs.
• 23:34
  Nobody's done that for, like... Y- you, you familiar with, like, GraphQL? Yep, very. Yeah. Right? Nobody's done that with a organized schema for automotive to where we can all play nice together. Right. Right? Yep.
• 23:50
  Because everyone's got their different way of doing it. And, and I can, I can go into- [laughs]... I can go into that deeper.
• 23:56
  But it's really- You can go, you can go into how one day you'll get a pipe delimited, uh, file, and the next day it'll be comma separated, and the, and the files in one col- column are not in array. No.
• 24:07
  And it's just a, it's just a list item which c- further creates new columns in that comma separated environment. E- Oh, yeah. [laughs] Yeah. And then the whole inventory feed goes down. Goes down. Yeah. Yeah.
• 24:17
  The whole, everything. Um, but no, that is, that's- There's a, there's a therapy session after for [laughs] Thank you. Oh. I think I've got three support tickets in that, right now for that.
• 24:28
  But, um, well, we don't do that. I mean, everything for us is a real-time fetch. Um-And I just, I think it's gonna take, and I wanna be part of that movement, right? Yeah.
• 24:38
  I wanna be part of the movement where, like, if, if, if I had one wish of the legacy that I could leave on Automotive was that I helped rally the troops to create a centralized API standard.
• 24:55
  I mean, people are doing this, but- For vehicles and customers... and deals and- Deals, parts... yeah, vehicles, customers, deals, incentives, packages, F&I products, a, a service appointments.
• 25:08
  This wouldn't take that long, right? It, no, because it's all the same data. It's all the same. It's, it, it's all the same. You know, you know how long a- We should just ask ChatGPT.
• 25:17
  [laughs] No, but we've gotta, we gotta hold everyone to those standards. And every- Right... I think we've, we, we go- gotta get every company together and say, "This is, this is how we all play nice together."
• 25:30
  Just like ADF XML, right? Even though it is something that is, you know, g- relatively dated at this point, the, the- If you send a well-structured ADF XML into any CRM- Someone will receive it... it works.
• 25:46
  And it, and it- It works... parses and it works. Yep. Yes. And that's what Automotive should be doing. Now, in the meantime, I wanna create really badass AI products and be on the precipice of that.
• 26:00
  So I'm gonna consume every piece of da- data I can with companies- [laughs]... that'll let me, right? I love it.
• 26:04
  Because it, 'cause I'm, I'm, I wanna create an awesome customer experience, and a, a awesome dealer experience as well, because I think that is lacking from some of the legacy systems.
• 26:14
  But in order for us to do that, someone has to rally the troops from all of these big CRM and DMS companies and say, "Remember that time that we all got together and ra- and, and did ADF XML?
• 26:28
  Well, this is the, this is the API version of that, and these are our standards." All right, one more question. You're looking at the horizon.
• 26:37
  You're, you're seeing what is, what's on the playing field, where you're moving. You're a visionary. You think about the future. What's, what's that, like,
• 26:48
  what's the shot you're taking for the next six to 12 months that maybe is a bit unexpected or is a l- bit of a long shot that you think is, is on the horizon, either in our industry or outside of it? Hmm.
• 27:04
  Are we under NDA? No. Um, so- [laughs] With all of the podcast listeners. Everyone just signed a friendly A. With all of the podcast, you're now verbally under NDA. Uh, Texas is a one-party state.
• 27:16
  [laughs] Um, no, I'm just [laughs] kidding. So I think the l- the long shot and the thing that I'm, I'm actually betting on is, uh, not as, as sexy as people may think.
• 27:30
  You know, everyone's, like, really, uh, focused on things like, uh, um, you know, AI BDC voice agents and stuff like that right now. I think background agents that just run tasks, clean data.
• 27:47
  You were texting a customer. They said, "Hey, my name's not Nicholas. It's Nick." Background agent's updating that in the CRM. Mm-hmm. You're saying, "I'm gonna be there for an appointment tomorrow."
• 27:58
  Background agent's adding that, you know, to the, to your appointments for tomorrow, sending out the email automatically. You're, uh, looking for reports. Background agent knows that you're gonna need that report.
• 28:11
  It's doing it. I think what the biggest thing that I, I, I think is gonna be unexpected is while everyone's trying to focus on these really cool, uh...
• 28:22
  You know, I'm, I'm even doing it with, like, you know, AI, uh, uh, AI help desks, right, where it's- Right... you're talking to a person, right? And we're all, we're all doing experiment- experimenting with those things.
• 28:34
  I think the silent, uh, and not so obvious, uh, real-life example in the next six to 12 months is fully deployable, uh, agents that are working in the background to, like, do everything from clean up your financial statements to admin work in your CRM to, you know, setting appointments, like I said.
• 29:02
  It's all of the things that we are manually having to do right now, like reconciling bank statements and things like that. Yep.
• 29:12
  Like, like we, we can fix those kind of problems, not just in Automotive, or I'm not even talking about my product here. Just talking broad horizons of- Yep... of AI in general.
• 29:23
  Background agents, that's gonna be the biggest winner. Monotony will be replaced by background agents. I like it. I like it. It's a good call. Yep. Well, Nick, uh, always fun hanging out with you.
• 29:33
  Thanks for- You too, man. Thanks for having a little bit of a chat today. [upbeat music] I just gotta get this off my chest. I love the way Nick says da'a. Like da'a. It's beautiful. It's magical. Yeah. He said the da'a.
• 29:48
  It's kind of frustrating, actually. A'a. Does he say die'a or da'a? I heard, I hear da'a. Da'a. Like, how do you spell that? I spell that- And it's D-A apostrophe A. Well, my question is, in, in school,
• 30:01
  how do they teach the children what the letter T sounds like? They don't. It's- They're like, "Oh, that's just a word-" They're like, "It's this." "...
• 30:10
  that we, that's a letter we put in there when we want you to keep your mouth open-" [laughs] "... for an extra beat." Just a'a. Da'a. A'a. You know? Right? It's just, I sorry, Nick. I love it. I love it. Sorry, Nick.
• 30:20
  We love your accent. Whenever you're around- We love your accent... I, I feel, I feel smarter.
• 30:24
  I do think, and, and this is the conversation that, you know, we obviously had, we went a lot of different places, but I think one of the core tenets of what, what Nick and I both believe is that this data ecosystem in auto needs a home base.
• 30:36
  It needs someone to say, like, "Hey, this is what a customer record looks like. This is what a vehicle record looks like.
• 30:42
  This is what a parts record looks like," so that we all are swim- uh, you know, at least able to plug into that standard. Um, and, and until we do that, like, it's going to continually be hard to integrate. There you go.
• 30:54
  You said it. Well- Nothing gets me going like a standardized API schema. [laughs] Yeah. Oh.
• 31:03
  [laughs] Uh- So we hope you've enjoyed this lecture on API schemas presented by Kyle [laughs] And the pronunciation of the word da'a. And Nick Ascue and the pronunciation of da'a.
• 31:15
  Uh, on behalf of myself, Michael Cerullo, Paul Jay Daly, and Kyle Mountsert, we hope you enjoyed this episode of Auto Collabs. Nerds.
• 31:24
  [laughs] Sign up for our free and fun to read daily email for a free shot of relevant news in automotive, retail, media, and pop culture. You can get it now at asotu.com. That's A-S-O-T-U.com.
• 31:39
  If you love this podcast, please leave us a review and share it with a friend. Thanks again for listening. We'll see you next time.
• 31:46
  [upbeat music] Welcome to Auto Collabs. [laughs] Welcome to Auto Collabs. [laughs] Why are we recording? Are we rolling yet?
• 32:04
  Not yet.