Short answer: Jordan Metzner demoed a flow where a web form triggers AI intent analysis; high-confidence leads get an automated SMS and low-confidence ones an email drip. Anthony Monteiro says vibe coding has changed the game versus big scrum teams.
Watch this answer from Episode 25 (starts at 1:57) · Episode 25 notes and transcript
From Episode 25: Car Dealers Are Losing Inventory — AI Is the Only Fix, starting at 1:57.
Sam Nadler [1:56]: Amazing. This is super top of mind for me. I'm actually selling two cars this week, so excited to learn more. With that in mind, Jordan, we built something with AutoAcquire in mind. Give us the quick demo.
Jordan Metzner [2:09]: Yeah. So it was really great to meet earlier this week with Anthony and Chris. Awesome to hear the story about AutoAcquire and some of its features and tool sets. So based on that, I did a little vibe coding, mostly front end type tool, which, based on your feedback was like a messaging bus. So I'd like to show that to you now. So essentially, what you do is you have some type of event. In this case, it's a webhook. So you can customize it. Basically, someone fills out a web form. So once they fill out a web form, then we send it over to AI, and we do some analysis. So, kind of what is the person's intent by their form. If we have a high confidence score, then we'll actually score it and then send them an SMS, like, hey, why don't you come on by or something like that based on a template type system. And if we have low confidence score, then we might do something like an email drip campaign if we realize, like, maybe the candidate isn't, optimal to maybe buy a vehicle at this time. And then, obviously, all of this is kind of totally customizable. You can edit each of these parts of these flows, of, if this case you wanted the confidence score to be 90, etcetera, and then, you can, edit all these templates.
Jordan Metzner [3:08]: And I'll show you a quick example of what this looks like. What you'll do is you'll execute the flow. And so here we have the webhook, the analysis, the text, and then here, it looks like my CSS is a little off. But he says, is the twenty one F-150 still available? Hi, Alex. Yes. The F-150 is still available. It's on a lot now. Would you like to schedule a test drive? And so, this message is just completely automated directly based on kind of the web form, the intent analysis, and then the scoring, and then send off that SMS automatically. So, yeah, that's that's what I built for you guys, Anthony. I would love to hear some feedback what you think.
Anthony Monteiro [3:41]: I think it's awesome. I love the fact, people using Vibe in other AI type generative platforms today has changed the game for us. Right? I've been doing this stuff since, I don't know, 30, I guess. And the days of having a massive room full of developers and scrum teams of 10 people and review and peer reviews and all of the things that really slow down code development, I think, are in the past. Now that's not to say that it's completely gone and we don't need humans that are gonna prompt these engines and or when the code gets generated, that has to be tested and connect to all the other APIs and fun things that we do as we build software. But, super cool. And it can you could easily, I'm sure, take this into anything. Right? Instead of, hey. Is the v F-150 available? It would be is your 2021 Porsche Cayenne still available, and would you like to sell it to us? Right? Mhmm. Here's an offer. I'll pay you cash today and just kinda work that, build this work flow down a intelligent conversation path via text with the customer, and it doesn't tie up staff.
Jordan Metzner [4:58]: Yeah. Absolutely. And, I just built a basic format, but, you could probably include this analysis later on down the pipeline again. Right? So you're kind of keep analyzing the conversation, decide, hey, do I keep responding back with SMS? Maybe I should call the person or even maybe bridge a phone call where you maybe call the dealer first and then have the bridge call the customer and say, hey, this is Johnny. You could even payload the data to the dealer side. So, his phone rings. He knows it's AutoAcquire. He picks it up, and he's like, hey, Johnny's got a F-150 to sell. Just try to get him down to the dealership. And then Johnny picks up the phone, and then, they can kind of bridge that relationship. So this was just a simple easy one. But what I like about the interface is I think it's it's very easily understood by a non technical user.
Anthony Monteiro [5:40]: Mhmm.
Jordan Metzner [5:41]: And I think this is is cool because dealers inherently are non technical users, and so, they may say, hey, you know what? This SMS template, like, it's it's not converting that well. Maybe we should, change out this template to a different template to see if we get better conversion or maybe AB test or, hey, know, sending them a text when they have high intent is not as good as just calling them on the phone. Maybe we wanna change this to a phone call example. So, I think dealerships can start to learn, and obviously, with your software understanding success rates, like, which flows are working the best, you might be able to do some analysis to see, how to optimize these flows for each dealer kind of thing. Because to your point, Anthony, these dealerships are some are small, some are large, some are in, small, Midwestern towns, and some are in, large, metropolitan cities. And I'm guessing that based on these geolocations, their customers have, completely just different behavior. Small dealership in a small town, you're kind of looking for a phone call versus a text, big dealership, big city, maybe a text is, like, more appropriate, for example, in, like, Manhattan or something.
