The Future of Conversation with Josselyn Boudett

September 1, 2022
The computer on the Starship Enterprise that crew members could talk to. R2-D2. Jarvis. Science fiction has been showing us AI and digital assistants for years. And Josselyn Boudett and the team at Stella Automotive AI are bringing them to life. Stella is a phone answering AI that books service appointments while helping customers get loaners or talk to specific employees. It’s unlike anything on the market today. Josselyn explains how Stella came to be and how digital voice assistants will become more prevalent in the future.
Listen On
Apple Podcasts IconSpotify IconGoogle Podcasts Icon

What we talk about in this episode:
0:00
Intro with Michael Cirillo, Paul J Daly and Kyle Mountsier.

4:59 Josselyn’s company, Telepathy Labs, isn’t specifically automotive focused, but it has a lot of background in automotive. The voice team’s background was building onboard speaking systems for cars in the early 2000s.

6:52 Stella AI started off talking about parts for heavy machinery. But then Josselyn took Stella to NADA and everyone started asking if it would work on a dealership phone system.

12:47 Stella is a true artificial intelligence. Most of the telephone systems that we have are IVR systems, or Interactive Voice Response. Paul and Josselyn talk about the differences between IVR and AI.

21:01 Josselyn thinks there will come a time when you won’t need a keyboard anymore. Instead, you’ll use a digital voice assistant.

⭐️ Love the podcast? Please leave us a review hereeven one sentence helps! Consider including your LinkedIn or Instagram handle so we can thank you personally!

We have a daily email!

✉️ Sign up for our free and fun-to-read daily email for a quick shot of relevant news in automotive retail, media, and pop culture.

🎧 Like and follow the podcast:

Connect with Josselyn Boudett on LinkedIn
Learn about Stella Automotive AI

Michael Cirillo: 0:00Hey, want to play a fun game?Kyle Mountsier: 0:01

No

Unknown: 0:11

this is Auto Collabs.

Paul Daly: 0:13

Ah, thanks for listening to Auto Collabs.

Michael Cirillo: 0:17

I'm gonna say what? What starts with the letter S and rhymes with tele.

Paul Daly: 0:25

That was a really terrible way to start a podcast but I think it's entertaining. So

Kyle Mountsier: 0:29

it's perfect. You know? This is we're not talking about Stella Artois that that will actually beverage right now. We're talking about Stella AI which is very, very close Automotive AI you know, the first introduction that I had to Stella Automotive AI was on a Clubhouse call.

Paul Daly: 0:47

I was there joke.

Kyle Mountsier: 0:49

I was there. You were there. It was unbelievable. So at that point, gentlemen name Hunter Swift is on was Stella, Stella AI. And he comes on this call and goes. So I'm not sure if I'm allowed to do this. Because I know it's supposed to be a pitch but can I just like show everyone the product? And I kind of you know, moderating the thing. I was like, Sure, but it's audio only he goes hold please. Home of like, you hear this like ruffling and all of a sudden, like what in the world is happening? He literally completes a servicing at some local dealership that he's caught up that has Stella AI. And at like, on the three minute drop, he doesn't even say a word after it. And everybody just loses their mind.

Paul Daly: 1:36

That was a mic drop moment.

Kyle Mountsier: 1:37

It was Oh, wow. So yeah, that's, that's, that's the introduction that you need today. Yeah, I mean, I don't know. I don't know what else to say. But we're gonna hang out with Josselyn, who's the CEO and founder of Stella Automotive AI, and can't wait for you to hear the conversation that we get to have with her.

Paul Daly: 2:06

Josselyn, I'm so excited to have you on the show today, primarily because we don't know each other right. A lot of times I get to talk with and Kyle and Michael, we have people that we know that we spent some time with. But I'm so excited to to get to know you a little bit better. I've heard from multiple people that you're like one of the coolest executives in the industry. So we'll see how you do.

Josselyn Boudett: 2:27

All right. We'll see how I do. But I just wanted to say we we you interviewed one of my staff at a conference and you guys just showed up with cameras. Boom, mic, boom, pow. 3 minutes it's super exciting. Love that energy.

Paul Daly: 2:41

Thank you. Yeah, I think we took a couple of people by surprise when we did that. They just showed up with the Lights Camera Action. Great. You know how it is? I feel like like the real the star players they perform better in that environment, don't you think?

Josselyn Boudett: 2:56

Yeah. That's more fun. That's for sure. Okay, so

Paul Daly: 3:00

Stella, Stella, Auto Stella.ai is the website. But here in waves about it, you all made an awesome impression at ASOTU's Pitch Tank. And really kind of had a mic drop moment with with an AI product. And it really, it's one of the few Pitch Tank's that I was at. I've been out over the last three or four months. And it really like shocked me, I was like, dang, like, that was one of the best pitches I've ever heard. Did you did you get a chance to listen back to that? Because I know you're gonna be pitching on Pitch Tank.

Josselyn Boudett: 3:30

Oh, yeah. And we're armed for Bear, we'll be with you in a few days on your next Pitch Tank. product does change. You know, one of the indications of AI is the product changes. So, products, you know, we have AI, adapt, learn grow. So we're always excited to show the product and it's different month to month.

Paul Daly: 3:51

Yeah, well, I guess I guess that's just kind of indicative of the auto industry in general, and what's been going on anyone who thinks, right, anyone who thinks that like you're gonna get to, like just rest on your laurels and like, get comfortable and like, Hey, this is what we built. Now we get to enjoy it. Right? That just kind of doesn't exist anymore. Well, tell me this. Let's do just a little bit of background, before we get into some of the amazing things that you are doing with the tech and how you're freeing people up. Give us just a little bit of background because I know you didn't come from inside automotive, which is in our opinion here at ASOTU is a major benefit. Because when you come up in an industry, you're just kind of trained to grow up and look inside the box that you're raised in. Right so you don't realize other things can even happen. Just give us a little background. You know how you got here? Why Your experience is relevant to auto in this moment that we're gonna get into Stella?

Josselyn Boudett: 4:42

Yeah, well, I'd say we, you know, we got into auto by request from auto and we're here with a product that we surveyed and people said they wanted so we were really excited to build it for auto we love solving problems. And then you know, how did we get here? My company is coming up on seven years old. We have over 150 employees I have 1/3 of my staff is PhDs in AI. It's an a pretty academic and advanced technology type company. That's why we call the route technology comes from a lab. We hold over 70 patents, we are experts in a lot of things, but strangely of a large portion of my org board is is has a background in automotive, my voice team's background, was building onboard speaking systems for cars in the early 2000s. And my team built the I think Toyota and BMW and Volkswagen onboard voices before they were acquired by nuance and then built voice for car back at nuance. So a lot of experience and affinity for this industry.

Paul Daly: 5:52

It's interesting how now the work you're doing is no longer inside the vehicle. Now it sounds like Yeah, the other side, right? You're you're actually you're speaking with customers speaking with an air quotes, right? Trying to try to free people up. Okay, so you have Telepathy? Well, maybe you have telepathy, but Telepathy Labs is what we're talking about. We did mention that, and a lot of surveying a lot of requests, and you realize like, hey, automotive is ripe for this technology. It's ready. And then you decide to get into automotive and like, go all in what was like, what was the primary target? Number one, ifyou could sum it up? This is yeah, do

Josselyn Boudett: 6:33

you have a party I like our product platform is called Stella. And Stella is kind of like a Watson. You know what I mean? You have IBM has Watson and AWS have had the Alexa platform. You have Cortana at Microsoft. So ours is ours. This is Stella and we're using Stella in heavy equipment to configure parts. And Stella is fully conversational and heavy equipment to talk about parts and repairs, maintenance, does, reports, looks look ups gives people advice on what part is needed, understands configurations, and helps with configuring. So let me stop you right there. So around, and I brought them to auto show them at NADA. And the only thing people said was that thing just talked. So I'm telling them parts. And they're just blown away that it's fully conversational and 100% wanted to know if we work on phones, and it was like the pain in the area of phones was so loud that we just zeroed in and listened and found out dealers need help with their phones.

Paul Daly: 7:40

So people were so impressed. And when you say you know, we brought it into show automotive and you're just saying like, hey, like we work in this major equipment of the parts. This is just what it does. Let's give you an example. Right. And the AI is making recommendations. And pairing part numbers up and connecting all the dots behind the scenes. And everyone was so impressed be like can you do that same thing? On the phones, right? Just want to make sure that we're clear on that's exactly what happened.

Josselyn Boudett: 8:09

Yeah, I was telling I'd wandering around in NADA and 2020 with the ZAP meeting a bunch of people shaking hands. And I was chased down. And I have over nonstop meetings with everybody. And I had brands asking me I had software companies asking me and like the most and then as I zeroed in on the conversational aspect of the system was more interesting than the content and you know, I think then it was really zeroed in on there's a problem in the phones. So we built a product for automotive Stella answers the telephone and helps book appointments, answers questions. She's building skills every day. And I would just say Stella is for dealers with telephones, she helps if you're looking at like you're not I mean, there's analyzing a telephone, trying to get people to move off of the telephone onto chat. But this is the first product that answers the telephone. In many cases better than a human. Because she gets the branding, right? Every time she in she can help people and defer some of the low hanging fruit of repetitive questions that a computer would be suitable for.

Paul Daly: 9:17

Can you remember? Can you remember your first experience with with like a phone? voice assistant? I'm trying to remember mine right now. Because it's come so far. And in the very beginning, I guess it was probably if I'm thinking about it. Probably the movie theater. But that was the

Josselyn Boudett: 9:35

I'm gonna say hands down first experience I ever had was on Star Trek. But I'm an older person.

Paul Daly: 9:42

Explain that. unpack that for a second. Well,

Josselyn Boudett: 9:44

I mean, I think in a part of the purpose of our company is to do something like you see where depending on if you're new Next Generation or Original or, or even something like Iron Man where you have or Luke Skywalker, where you have a droid helping you. And I think that's what we set out to build, we really ignored the Chatbot narrative. We never tried to build one, nor do we want to, because we, we don't think they're very effective. And compared to what you're seeing what we all expect what we all grow up knowing we're going to have, which is a true digital assistant, you have it at home. Our purpose is to bring it into your work.

Paul Daly: 10:28

So there you so focusing in on voice right voice is fast, right? And that's the nice thing about it. It's fast, it's hands off

Josselyn Boudett: 10:37

time. Yeah, you don't care about the software, you just saying like today, you can say, hey, what movies playing tonight, you don't care if it went to Fandango or AMC or I couldn't care less where it went? I just I just want the answer. I think, you know, when you have customer calls your your dealership, they don't want to worry whether you're on what scheduler you're on or whether they just want the appointment so that you know, you get to just the digital assistant really takes the burden off of the staff and the customer of the technology and just, you know, puts it into a dialogue and then arbitrators back to the back end or the staff as it should. So you can say, hey, R2 what's wrong with the hyperdrive? You know, hey, you know, Stella, what do you got open on Friday? Like, I don't want to worry about what she had to do to answer that question.

Paul Daly: 11:31

You have, obviously a deep connection to the film industry. I can see. Yeah. Is this Star Trek your number one US Star Trek Star Wars Marvel? I don't know. That's just the references as they came in? Yeah, of course.

Josselyn Boudett: 11:45

And I, you know, I'm kind of a gamer. So I'm a technophile. For sure.

Paul Daly: 11:49

You're a gamer. Okay. What are the games?

Josselyn Boudett: 11:51

Oh, no, no, no. Oh, we can't go there. Come on. My favorite game is Stella.

Paul Daly: 11:57

Right, did the real world game

Josselyn Boudett: 11:59

in any TV show or game?

Paul Daly: 12:01

So I think that, let's talk about some of the kind of the pushback that people might give to AI. You know, people might say it's slow. It's frustrating. It's annoying, you know, depending on and there's just a, such a wide variety of experiences out there, right? No, I'm sure No, two, you know, voice assistants are the same. Like I even experienced that. But let

Josselyn Boudett: 12:26

me put it this way. I'm gonna interrupt you and say that nobody's ever called an AI before. Really? I mean, if you're talking daily, that is, all American Airlines, call Apple call Ritz Carlton, call Procter and Gamble. I mean, I don't know who you're gonna call. But the thing we built for automotive is better than all of them. Wow. Because you're not talking to an IVR. You're talking to an AI?

Paul Daly: 12:50

So break that down for a second. Yeah. Talking to a what? An IVR.

Josselyn Boudett: 12:55

Yeah. And interactive voice response system explained the difference between

Paul Daly: 12:58

the two things and IVR. And what you're doing?

Josselyn Boudett: 13:01

Well, I think, you know, back in the day was the horrifying press one, press two.

Paul Daly: 13:09

And that it's just trying to identify what the category

Josselyn Boudett: 13:13

Yeah, all it can do is receive a number and then that move just somewhat voice where you could say, sales or service, or tell me in a few words, but really the key word, I want to check my balance, right? If you have sales service parts and used cars, and you go, Okay, well, one for sales, two for service three for us parts, and then you replace that with a language system. It just knows those four words. It's not very impressive, right? Because you go parts. All right, you're still transferring to.

Paul Daly: 13:40

So you're saying that Apple, I'm talking about? All IVRS

Josselyn Boudett: 13:45

I'm talking about on the telephone, that you say, hey, I want to come in Thursday morning, and I want a loaner car and it goes, Okay, what year what year making model that's frickin smart. And you'll you don't see that anywhere. That that is what we expected. But you can't take for granted that you don't see it anywhere. And I believe this is a leading edge product. Market wide not just for auto. You know,

Paul Daly: 14:14

there's so I'm on the borderline between a Gen X er and and a millennial. I was born in 1979. Right? So it's like, I'm kind of straddling both worlds here right in the middle. And I do know that the Gen X, I mean, sorry, millennials and Gen Z a very low tolerance for slow tech. Right? Like, I've just literally watched people be on the phone like with an I would be now I know it's an IVR and just be like, Oh, I can't do it. I can't do

Josselyn Boudett: 14:40

it. How embarrassing and you know, on top of it take it's like, as a customer, absolutely. And then but also as an employee, they don't want to work somewhere where they look that stupid. Like, I don't want to meaning because like I have to log in and do some linear process online. So just I gotta I gotta read to script, look up a system get an answer. It's not all there, I'm not really sure what's going on,

Paul Daly: 15:05

we're saying that they don't want to, they don't even want a job, not

Josselyn Boudett: 15:10

only do they not want to buy there, they don't want to work there. And you'll see it reflected in, we have dealers paying $25 An hour and struggling to get people to take the job. Just because

Paul Daly: 15:20

it's feels like it's like one of those soul crushing jobs where it feels a little robotic and pointless, maybe. So you're saying, they don't want to do that? It's slow. And it's also I never thought about it from the employment angle.

Josselyn Boudett: 15:37

Yeah, I mean, it goes both sides. I mean, the real pain that we're going to help try to solve is taking some of the load off of the repetitive processes that are mundane. And, and leaving people to solve problems, which is the best, you know, people are good at solving problems. Computers aren't. But you know, repetitive things that can be solved would be really good for an AI. And, you know, you're growing in the in the consumer side, I think there's no case where we don't fight I think 94% of the planet Earth has adopted Siri, and Google. And you know, there's just no question. It's hyper functional and helpful. Why? Because you use it all the time in preference to go into a website or looking something up. So I think, you know, putting that on the phones is exciting. Most of our appointments get booked in a minute, minute and a half, we can answer in one ring. So it's like boom, one ring done. booked, scheduled, confirm, done, in and out. That's that's what I would call a brand aligning experience. I mean, the the brands are selling, the dealerships are selling top technology, look at a car and how loaded it is, even on a simple low price vehicle. You're getting amazing technology, and you have your, and then you call this dealer, and it's like you're calling an IVR from 1985. So I think, you know, we're just I think we're helping dealers, we're helping their employees. And we're cutting costs, and we're giving and providing a better brand. Like, I think our average score on on customer satisfaction on a Stella appointment is is 85%. Awesome.

Paul Daly: 17:19

You're just bringing up all these memories, recent memories, and there's one dealer that I won't mention, but when I dial in, because I go there for service. It's like it sounds something like this. Thank you for calling blankety blank blank for a world and it's like just slow. The cadence is slow. Like I listen to podcasts on like 1.5 or 2x speed. So I'm already like, Come on already

Josselyn Boudett: 17:43

spit it out. But you don't get nervous hysteria. Yes,

Paul Daly: 17:46

yes. And then it sounds like something like this. For a world class sales experience. Press one I'm like, it's already not right. Like it's already not. So for a world and it really just drives the point home that this is slow. This is annoying, right? And then if you happen to press that property here, hold music. And so you're saying Stella fixes that?

Josselyn Boudett: 18:10

Yeah, it fixes that. And there's no reason why you can't pick up and book an appointment in one ring. Now for something else you might need to hold, but at least we peeled some of the work off. I mean, these guys in the call centers are just getting shelled, you know, and the other strange. Just, you know, I feel bad for them. I'm like so but the other strange thing is, you think because of COVID the chat lines would be on fire, and they're not the primary. The primary channel is still telephones. Why?

Paul Daly: 18:41

Why do you think the chat isn't taking off? Because I try to chat whenever I can. Is it lack of availability? Or is it like just behind the times, and it's typically slow

Josselyn Boudett: 18:49

as well. Most of the chat bots aren't intelligent.

Paul Daly: 18:53

So it's the same thing. Like it's just like, you

Josselyn Boudett: 18:57

know, I got a chatbot maybe your first thought is going to be how do I get the chat bot to get me to a person very bright, and you know, like, you go, Oh, I'm interested in your, you know, your, your vehicle and it shows you a picture and then the very next question, it's, it fails. So, you know, I was like, alright, well, I saw a car now how do I you know, so I want to compare it or look at it or find out more about it or it's really hard for the bots to track those conversations,

Paul Daly: 19:29

you know, in a best case scenario, and a best case scenario. Somebody deploys Stella in sales and service right? Is it both like when the phone comes in or is it primarily services the primary deployment now

Josselyn Boudett: 19:44

we're just handling appointment like look, we're your appointment line and concurrent calls. You have like 20 concurrent calls going on. No problem. scalable. 24/7 answers the phone anytime a day. And then the system also takes voicemails So, you know, you could say during the day, I really want to talk to Paul and Paul's busy and tell us that can I take a message? And then we'll we'll email that to you. Well, that saved you some time as well. I mean, you come in Monday morning, you got 20 messages, that's at least 40 minutes of noodling around listening to those voicemails for sure. So we our goals are time saving, I think, you know, we'll take on more topics in the next year, we're, you know, we're about 120 days in market and just crushing it, and we're growing so fast. So what's your

Paul Daly: 20:30

like? Okay, so your 120 days in market? Primarily taking calls to make appointments? You said, you know, there's some other things over the horizon, you know, that you'll you'll lean into? What is your vision? Like? What is your five year vision for what Stella could actually do? For an auto dealership? Thorough, like, I mean, throw the ball over the horizon for a second and just kind of give me a picture and give the audience a picture of what could be possible with this type of technology understanding, it's got to be developed with this type of technology to deploy throughout the store.

Josselyn Boudett: 21:01

Yeah, well, I think, you know, the future is these digital assistants, I think you can, again, go back to the movies to see what it's like to have an intelligent digital assistant in your area, we do not think will be the only voice company, we think there'll be a lot of enablement of voice technology across software's, and that you'll see more interaction using voice instead of keyboards and mouse's. And I would expect that across the board, I think we're like, the telephone is a whole other problem set very complex to have a conversation on a telephone with a digital assistant. So I think we'll be cutting edge on phones. So I see our area is really helping on the telephone for a very long time and expanding those capabilities and helping relieve you know, the the employee and staff have repetitive things and helping taking the burden off the organization to get better scores. Now, you're talking about what will be there in five years, true digital assistant technologies coming out everywhere, but unifying it having a single interface that you can ask a multitude of questions, you know, regarding, you know, is somebody working today, operationally, you know, how busy are you? What's going on right now? For the staff members? Do we have this, don't we? And you know, looking at inventory, looking at sales performance, looking at the success of marketing programs, looking all of that accessible on voice all of it. So you're saying nothing, your keyboards will be gone? So I'm

Paul Daly: 22:35

gonna give you some examples. And you tell me if this is what you mean or not? Yeah. Show me every vehicle we have with a red interior.

Josselyn Boudett: 22:43

Yeah. Are this been on a lot more than 30 days or show? Show me all? Show me a car, this car the last 20 times we sold it? Or how does it compare this year to a same time last year? How much did we How many snowstorms did we have last year? And how did it affect sales?

Paul Daly: 22:59

Like? I think that's, that's unbelievable. Yeah, that is amazing. Well, this is Auto Collabs. So we talk a lot about, you know, industry partners working with one another. Is there anyone you're working with right now to kind of like figure out the better product fit or like collaborating with other tools that dealers are already using?

Josselyn Boudett: 23:19

No, I mean, we are a member of star and we are definitely a believer and proponent of open standards and data sharing. I think this industry is making some changes that are important that will make it easier to federate, we certainly aren't trying to recreate the wheel. There's some awesome vendors that are figuring out a lot about the service lane or marketing or selling cars. So yeah, I mean, I think in the long run, federating and working with partners, but right now we are so focused on just doing our job, which is booking appointments. We're doing a great job, we are scaling, and I think we took on opcode Matic mapping using natural language we took on, you know, you know, rentalcars, shuttles and million topics using natural language. I think that's pretty exciting. I think, carrying that through into potentially booking appointments for sales, and then answering more questions about car readiness, assisting with doing a better job and leaning in on recalls, and maintenance and things that the OEM are interested in. I mean, I see there's a lot of opportunity here, but we're focused on phone what is the consumer need, you know, what is the dealer need the consumer to get or No, and that's really where we're our lens is not on the data, our lens is on the customer.

Paul Daly: 24:38

I think that anyone who has their lens focused on the customer is actually going to be winning and helping anyone they're working with win as well. Josselyn, thank you so much for giving us some of your time. I know you're you're traveling the country and you're preaching, preaching this Stella Auto Gospel all over the place. I can't wait to meet you in person sometime soon and excited. Good to have you all participating. Yeah, we're

Josselyn Boudett: 25:01

really excited to be a sponsor of your of your conference in the fall. So we'll look forward to seeing you and the dealers there. cannot wait. Thanks again. Thank you.

Michael Cirillo: 25:14

I remember a couple years ago, I think it was a Google presentation, where the CEO of Google demonstrated their AI being able to book an appointment, and they called a Chinese food restaurant, I think, and made a reservation and everyone was like, oh, and I remember being such a nerd when it comes to this stuff. I just ate that up. And I'm like, won't that be cool and 700 years when somebody actually does something with it, but now here we are. And that's what's so intriguing to me, you know, in an industry where we throw around the term, AI, artificial intelligence, oh, yeah, it goes into our AI and this and that. Stella's stepped up. And they're actually distributing something that is not weak AI, probe grammatical AI, like what we see in chat bots, and things of that nature. They're actually stepping up to the plate and innovating and saying, No, here's some strong AI for you.

Kyle Mountsier: 26:08

I didn't Well, and what I appreciate about it is also the fact that, you know, and this is across the board with Stella and I think and I think we're seeing this more and more in companies that are that are saying, like we have AI, that it's not a replacement, that it's supplemental to, to the company and to your people. And it's and it's freeing them up to be doing the things that they can be doing, and leaving to computer technology and an AI and ML machine learning. You know, the things that it is good at, which is solving the problems of a customer needing to schedule an appointment, it can handle that level of conversation and solve enough problems enough, yes, knows, and enough movements around the way that someone's conversing with it, to get to the result, that really is what that customer is desiring, freeing your people up to handle the more complex ideas. And the more you know, the more like the things that are the need that human touch.

Paul Daly: 27:07

I'm so excited about the future. She described where the data sources are all connected. And the AI will be able to pull things from the CRM and from all the different systems and actually give you a cohesive. What was the example? She said in the interview? She said something like, how many snow days did we have last year? And how did it affect sales? And it will be able to spit you out that answer the amount. I mean, I just wish someone would design something in QuickBooks that just says, Can you show me a year over year profit and loss starting in April? Right? That would just be amazing. Right? So but what she describes is like three levels beyond that. So it was great to just kind of hear a little bit about the future from the people who are making it on behalf of Michael Cirillo, Kyle mount seer and myself. Thank you for listening to Audible sign

Kyle Mountsier: 27:58

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@asoto.com That's ASO T u.com 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 Welcome to recording

Get the daily email that makes reading the news actually enjoyable. Stay informed and entertained, for free.