The Data Behind Every Drive: Powerful Insights from Phone Telematics

Mike Chung 0:07

Welcome to Autocare on Air, a candid podcast for a curious industry. I'm Mike Chung, Senior Director of Market Intelligence at the Auto Care Association, and this is Indicators, where we identify and explore data that will help you monitor and forecast industry performance. This includes global economic data, industry indicators, and new data sources. Hi, everybody. Welcome to another episode of Indicators. I'm Mike Chung, and I'm very happy to introduce Anthony Johnson from Eriity. Anthony, welcome to the show. Tell us a little bit about yourself, your role in Aridi plays.

Michael, first and foremost, I'd like to thank you and ACA for having me on. As you mentioned, I'm with Eridi. I am what you would call a solutions engineer. All that means, fancy way of saying technical resource and business development. So, you know, I think we'll talk about this, but Aridi has a number of different data products. And, you know, as we build partnerships and relationships and in selling those data products, someone has to understand it from a technical perspective, right? Solutioning. I'm trying to solve for X. Aerity does your data, or do you have a data solution that can solve for it? I manage that process. I support that process in that's essentially what I do. And um, I've been with Aerity for almost five years now, and I'm based out of uh Humid Houston, Texas.

What Arity Data Products Do

Humid, Houston, Texas. I love it. Thanks for that introduction. And for viewers and listeners who don't know, AutoCare Association has partnered with Aerity to get vehicle miles traveled for five, six years now. And tell us a little bit about, I guess, some of the other data that you and your team are providing to your clients, Anthony.

Real Time Drives And Traffic

Risky Driving Events From Phone Sensors

Retail Visitation And Location Intelligence

Yeah, certainly, certainly. So VMT, um, that's kind of like the the birth of it in the mobility intelligence space, which is where I sit within Ari. And, you know, that space is think of just any of the derivative data products that uh we've we've created from this data that we capture first party, by the way. Uh, we are not a third-party, you know, data provider. We are the ones with the technology that's capturing the data itself. But um outside of VMT, um, we have something that's called real-time drives. Um, and you know, I always like to do this. We say real time. Well, what do you mean by real time? Um for real time for Aerity, there's about a 60-second latency uh in that data, which is a stream of every single driving trip that Aerity sees 24-7 every day. So let's say we're tracking somewhere around 50 million connections. Think of a connection as a driver in the US, and um that's about 150 million trips every single day. Um, so imagine seeing those trips in real time, right? Now, why would I want to see this data in real time? Um many cases um or or reasons for that, and the customers and the partnerships that we have that are leveraging that data source, that is around real-time traffic decision making. So when you're driving in your car and you see those little red, green, yellow lines is telling you, hey, you're gonna be in traffic for six minutes, or it's green. Those entities can then, you know, predict, or not so much predict, but determine what is the traffic on the road. Okay. Um now I mentioned the real time. That data then gets um matched or route matched to road segments. And I'm choosing my words carefully here because I don't want to go too deep in the weeds, I don't want to lose anyone, but just know that if anyone is wanting to understand Aries traffic that we see, which is you know, that's a significant amount of data, historically, um, we go through a process to package the data historically, meaning that so for any given stretch of road, and going back to 2019, January 2019, we can look at that road segment and see things like, well, what was the total volume of trips within that hour? What was the average speed with zero speed stops and without it? What's the speed distribution? Okay. Um we then can move into things like driving events um in aggregate. So um I got to take a quick step back just to explain what I'm saying when I say driving events. So not only is Aridy seeing the trip itself, um, we are also capturing driving behavior. Not just any driving behavior, but it's specific to what we would deem as risky driving behavior. So um, well, how are you getting all this data? It's coming out of the car, the OEM? No. We have an SDK that sits inside of a number of different mobile applications, mobile publishers. This little thing right here that you know we get real nervous about if we can't find it. It has an accelerometer built into it. And this is Android and for iPhones. Uh, it has a gyrometer. There's a number of different sensors installed in it, and we use our algorithms to do these types of detections for that driving behavior, such as harsh braking, rapid acceleration, speeding, which is going over 80 miles an hour, um, distracted driving, which is related to um phone things that we should not be doing on our phones while we're driving. Okay, so like the unlocking and locking of the device, um, picking up the phone and doing things. And I'm not talking about just the rattle, the uh movements of the vehicle naturally. Things like, hey, I pick my phone up. Now, if I'm doing these things, there's probably a good chance my attention's not on the road. Right. Last driving behavior is crash detection. Um, does not guarantee that a crash occurred. But again, that sensor data that I mentioned earlier that we're capturing, we provide a confidence score, a likelihood that when certain thresholds are met, hey, we believe something happened. Now, there might be a 75% likelihood, a 90% likelihood, or maybe it's just a 40% likelihood. And right, you could filter that data out to say, hey, well, if it's not 75% or greater, I don't want to see those detected crashes. Again, it doesn't guarantee that it happened, but likelihood. Um, last thing I'll mention, and then I'll take a pause. Um, something that we call road traffic analytics. Okay. And think of that as the marriage between all of this trip data that we see and locations of interest. Um, retailers, um, brick and mortar, right? We're talking brick mortar, brick and mortar stores here. So retailers, restaurants, quick service, all of that. Um, with the exception of any sensitive locations, such as places of worship, right, healthcare facilities, things like that. Um, we cross-reference our data against those locations because then we're we we're able to do some amazing things in terms of um looking at visitation, right? Now, for Aerity, visitation, and it's it's important to note that um we focus on vehicular movement solely. We don't look at the movement of just individuals. Um and saying that I say this, I'll just throw something out there because I know somebody's thinking it right now. Um so you can pull me up and you can see where I'm driving every every single day. Not you specifically, not Mike specifically, not Anthony specifically. Everything that we report on is de-identified and anonymized. And especially when we're talking about locations of interest, or you may hear some hear some say AOIs, areas of interest. Um, it's all in aggregate. So if I say in Walmart, between the outlets of Walmart from the corner around the corner for me, if I said, hey, I want to look at Erity's data for Walmart visitation between 8 a.m. and 9 a.m., what we're gonna provide you with is a total count. Now, maybe Anthony went, maybe he didn't, but you're not gonna know it was me specifically. I'm just gonna be counted as a number. Hey, there were 654 people that went within that hour, right? And um, so visitation, origination, where did they come from? And that's that's done two ways. It could be from a standpoint of a work location, census block group, or by home location at the census block group level. So we're not going to places of residence and right, this individual at the census block group level, we'll report on and say, hey, so for that 600 saw some 600 some odd visitors that we saw at Walmart or AutoZone, let's say autozone, where did they come from? So then we'll say, hey, these are the dominant block groups of where those individuals came from. Well, why do I care about that? Well, now you can do two things. One, you can marry that up to demographic data, which AirDusn's provide demographic data, but you may already have a source for that. Or, right, we can assist you and support you in that because there's many great resources for demographic data because now I can then pull in the demographics for those dominant block groups. Now I can create a profile for who are my customers, right? Number one. Number two, where are they coming from from a trade area standpoint is critical because you need to understand what is my trade area, where are my customers coming from. So we can do things like to look at what's the top 80% based on drive distance or drive time of where those individuals are coming from. Let's identify those block groups and then, right, and even not even from the block group level, right? We could basically take those locations and imagine if you remember as kids, the connect the dots, you connect the dots in number one, two, three, four, five, and then you got a picture at the end. Similar approach in spatial analytics. We would create a polygon to say, this is the area in which this auto zone services, right? Or I'll pick somebody else, discount tire. And I just throw some different names out there. But that that is the application. Um looking at competition, visitation, I'll say one more thing before I pause around that. I would imagine just about every brand is very familiar with their stores, right? They know the number of transactions, they know the volume of traffic they're seeing at their stores. Awesome. What many don't know though is what are your competitors doing? Right. So the idea is that visitation can be treated as a proxy to understand your market share. How am I how am I stacking up against my competition? Right. Sure.

So one of your customers might say, I have visitation for my store. Let's compare that to companies A, B, and C.

You got it. You got it. And how does that change over time? Right? And you're running campaigns, you you you kick off a new marketing campaign. Um, how did we do?

Right?

Now, sales for sure, but then again, did we take market share away from uh XYZ competitor when we kicked off that marketing campaign? So those are some of the applications in use cases.

And a very interesting piece of the jigsaw puzzle because you might say, I'm store owner for Company A. I'm using print media, I'm using billboards, I'm using ads that go to your phone, and getting a count of visitors to my store versus my competitors can be really informed the effectiveness of my marketing campaigns.

How The SDK Collects Telematics

That's right. And what we're the name of the game, folks, what we're talking about here is understanding the ground truth. We know in this space there's many, many unknowns. And we won't probably won't necessarily get them all, but the idea is um, you know, the type of data that we're providing is, you know, filling in some of those gaps, right? Uncovering some of those unknowns so you can make the most informed decision possible. Um as an end-all be-all that you you get our data and you have all the answers. We we are one of the pieces to the puzzle, if that makes sense.

Absolutely. And everything you highlighted, it's quantitative data, it's spatial data, it's time-based data. So it really adds a lot of color to anything that an analytics or operations or strategic planning team will be doing. So you gave me so much that I can dive into. I'm not gonna hit every single thing, but just let's go over a couple of uh real quick follow-ups because I know that a lot of our listeners may be wondering the same thing. Um, you clarified already the data is coming not from the vehicle, but from the phone. And just tell me a little bit more about that. How is it coming from the phone?

It is coming uh from the phone through mobile applications. So there are a number of different mobile applications that leverage the Aerity SDK, right? SDK's what again? Software development kit. Thank you. Think of it as the the brains of the mobile app. Not all of them, but it's a component. It is the driving engine, it contains the logic to do everything that I described to you. So one says, well, what kind of apps? Aerity is a wholly owned subsidiary of Allstate Corp. Okay. Allstate insurance, which is a part of Allstate Corp., uses the Ari technology. Um, for Allstate Auto Insurance, they have what's called DriveWise. Right? DriveWise is an app that sits on your phone, those that use it and that have opted in and say, yes, Allstate, I'll share with share my data with you so I can uh get a discount on my insurance, and hopefully my insurance now is being priced appropriately. Um, and it's not biased based on my age or income or education, it's on how I actually drive my vehicle, right? Um, but it's not just auto insurance apps. If you have heard of Light 360, which is the family safety app, you got a teenager that's driving around and and the family's all sharing, you can see where everybody's going. That's another example, okay? To name a few. Um, so all of those mobile applications leverage the airity technology. So, right as that data is being captured, it captured, it's then coming to us and our engineers are taking it through a number of different things to filter it, clean it up. Um, you know, airplanes that might be picking up a trip and trains and boats, and hey, that person's going over an ocean. Probably not a vehicular trip. Right, right. We're removing that stuff. Um, but yeah, that that is how we go about it. And as I said, we leverage the sensor um data on the phone um to detect um those driving behaviors as well as the individual's trip from point A to point B.

Coverage Scale And Data Granularity

Sure. So some of those data points are origination point, ending point, time of day, day of week. Um and you uh mentioned that this data has to be de-identified. So things like the person's name, the car that they're driving, the make of the car, things like that are not allowed, so to speak. And so I'm also interested in you mentioned something because I think I did sign up for one of those apps to lower my insurance costs. And it was it would start and it would it realized though that I was on the metro in Washington. And so that type of a trip gets canceled out. But some of the other things that you highlighted, like speed, rapid acceleration, harsh braking, those things do get collected and um is available for you and your um your your data analytics team for your purposes. Am I getting that correct, Anthony?

You are getting that correct.

Yes. And so I think you mentioned 50 million. 50 million is a big number. And I would imagine from a statistical representative standpoint representativeness standpoint, that's robust to characterize what's going on in the United States.

Yes, absolutely. So what we estimate today and uh ballparking here, let's let's say on average, that that those number of connections that Aerity has represents about 19% of the U.S. driving population. Does it vary by state? Absolutely, right? We have coverage, higher coverage in some states, and it's a little lower in other states. But on average, let's call it around about 19%. And to your point, um, that is absolutely a we we feel strongly about that and proud of it, that that is a very strong representative sample of what's happening on the roads.

As well as not just state to state, but I would imagine your data analyst must be so thrilled to have such a robust data set because I think about things like urbanicity, suburban, urban, exurban, rural, or you mentioned states, geographic regions. You might be able to do things on, say, temperature climates, desert versus mountain versus beach area, things like that. So uh that's got to be exciting from a from a data analysis, a data junkie standpoint, right?

Use Cases For Cities And Brands

It absolutely is. Um, right. There's um excuse me, it allows for a lot of flexibility to dive in deep and explore and uncover or you know, attempt to answer those questions again, flush out the ground truth. Hey, we've seen this happening. Why did that happen? Oh, it was um road closure on this particular highway, and folks are being rerouted. And so that has had an impact on the normal volume of visitation that we would normally see. You know, I'm just giving that as an example. But uh, yes, with that flexibility um down to that level of detail with a date timestamp, um, and even think of uh think of it like this too. You know, again, I mentioned those trips, the individual trips that we see 24-7 every single day. If you were to try to envision it in your head, you know, I I think back to the as a kid and like the little fairy tales and stuff, and who is it with the little um the little boy and the girl and they got lost in the woods and they dropped the little crumbs? Thank you. Then they leave some breadcrumbs and some cookie crumbs later. So thank you. So think of that as what's happening when we're seeing a trick in that for every single, every 15 seconds, we're dropping a breadcrumb, which is a pen. That breadcrumb is a GPS point. Latitude, longitudinal coordinate of that trip, date timestamp down to the second. Okay. So going back to what you mentioned in terms of the just what you can do with that, because of that level of granularity, it allows us to really dig in deep and do those things, even beyond just the visitation. What about dwell time? Right? How long did that visitor, or, and again, it's not to the individual, but uh in aggregate, how long are visitors staying at this location? Right? And let's see a distribution of it. So what's the average dwell time? And then, you know, how does that break out over how many visits did we see from zero to five minutes, five to fifteen, right? 15 to 30, and so on. Um, you have the ability to do those types of things, which which is powerful. It's very powerful.

So two things come to mind. One is sort of the use cases, some of which you highlighted. So I'll talk about a couple and get your kind of feedback. And then the second is something you kind of alluded to, I feel like, in terms of the road trip analytics, some of the new recent aha's that you have and your team have had. So let's let's keep those two things in mind. But let's start with the first because some of the um use cases I feel like you touched on a little bit. It could be a municipal department of transportation that is looking at okay, where are we having congestion? Or what is the impact if we do construction at a certain time for this bridge, for this tunnel, et cetera? And another use case I can think of is I I think I've heard heard it called the Taylor Swift effect. When Taylor Swift has a concert or there's some special event, how does that impact traffic? And then what can that municipality and the local authorities do to plan for, alleviate, reroute, and make it a safer and more efficient driving? Experience for people. And then another thing, and I know we can probably talk about a lot of examples, but something you just you highlighted with the discount tire or autozone. If you're a retailer, what is that dwell time for a location? And then putting it in the context of, well, what exactly is that location? Because if it's a shopping mall, that can have implications versus a standalone store that is the only standalone stair store within a six-block radius, for instance. So pretty wide open. But tell me a little bit about the use cases and feel free to expand on any of those.

Yeah, let's see. And I and I've shared some of those use cases already. I'm thinking about some that are outside of kind of what we've already talked about. So let's let's talk about you mentioned the Taylor Swift, right? The reality is it could be any event. It could be an incident, right? It could be a bridge collapse, God forbid, but that happened recently. Our data allows for pre- and post-countermeasure studies. Even think about up in New York, right? Last January, the the toll rollout, right, for charging the um to go into management pricing. Thank you. The congestion-based pricing. So, hey, I'm a traffic engineer and I'm on that project. I want to understand what was the impact. Now, by us having that data historically in aggregate, we can go back in time and we can look and see what did it look like before congestion-based pricing was implemented? What does it look like now? That's critical. And that applies to anything again, be it an event, be it uh um right, that type of change. It could be an incident. Um, it could be um a control um that was changed on the road. I'll give you an example. So we're gonna install a roundabout right now. Now I'm on the traffic side of things, but a roundabout, or we're gonna change the signage here, we're gonna um change the speed limit. What kind of impact did that have? Did speed go down? Did it go up? Again, pre-post-countermeasure studies.

So any of those problems you're trying to solve, if you will, and problems I should probably use air quotes, it sounds to me like you have to define your area. So if it's Manhattan, draw your line around the blocks and avenues of interest, perhaps the time period, let's just say January 1 to April 1. And maybe there's certain times of day. How about things like weather? Um, are is weather being pulled into the data from those SDKs and apps, or is that something you might have to layer in?

No, you would need to layer it in. So think think of that. You know, I mentioned earlier that there's no um single source or um uh silver bullet, if you will, excuse my reference. All that to say, again, Arity is one of many data sources that you bring together to paint the complete picture. To your point. So do Arity does not pull in weather data into what we deliver. However, because again, we have a date timestamp, there are many different weathered data sources that you can marry up and then see, okay, how is traffic or how was traffic impacted because of this storm or right? New York just had the blizzard. Those types of things, you can align them out temporally, right, based on that time period, and then gain your additional insights.

And similarly, if I'm a retailer, I might say before I had a big promotion, after I had a big promotion, and kind of know some of the other X factors, for lack of a better word, that could be perhaps driving traffic to my location, double pun there, had to do it. But you know, I'm sure it's uh pretty standard for what you and your teams do.

Recent Findings On Commutes And Demand

100%, 100%. Um, another thing I wanted to mention, I'll throw one more use case at you. And this is something that we are, let's just kind of call this in beta, because it's it is being developed as we you know roll these products out. So you heard me talk about historical traffic at these roads, stretches of road. The reality is there's two flavors of that right now. We have it daily. What do I mean when I say daily? So you if you came to me and said, hey Mike, um for the Super Bowl, when was the Super Bowl? Was that the 7th? That was the 8th, I think. Was that February 8th? So on the Super Bowl, February 8th, um, in it was in San Francisco or in that area. For the San Francisco area, um, I want to see traffic for that day, February 8th. We can pull it. Okay? No problem. What becomes even more interesting though, in addition to that, is if we also pulled what we would call average historical aggregated traffic. So what do I mean by that? What I mean is so for that same area, for those same roads, but not for February 8th, let's say we go and look back for the last 12 months and take all of that data and we create an average by time of day. So then it's not by calendar date, it is more so a scenario of Sunday. So on Sundays at 6 p.m., what does traffic look like in that area? We've looked over the last year. So now you kind of you you you have a baseline of what's to be expected by day, by time of day. So now when there's outliers or anomalies or right, these different things happen, you can see how right much how how much more congestion there was but against the average, or you know, the opposite. You know, so does that does that make sense there?

It does. It it really defines it what it sounds like to me is what do you what are what is your comparison period, right? What are you comparing it to?

And that's important because again, and even then going back to the retail scenario for the brick and mortar stores uh from the firestones and so on, um, you want to understand what's normal traffic like for this location. So then when those changes happen, now we're getting w we're able to help better explain and understand why we're seeing what we're seeing.

So tell me a little bit about some of the recent aha's or interesting findings that you and your team have had, whether through the um roadway insights, the real-time analytics, any kind of interesting findings that you and your team have happened upon recently.

Yeah, sure. You know, ARD does this thing where we put out an annual report on just you know all of that data that I talked about or what we see. And so there's a number of different interesting analytics and things that we pull from, but I'm gonna share a few with you here. Um so uh one thing we learned last year in 2025 is that trips are getting longer, okay, both by miles and by duration, um, which is is suggesting longer commutes. Um, you know, miles per trip, trip duration increased in late 2024, and that continued going up through the first half of last year, um, higher levels than anything that we've seen in the prior years, um, which to us is an indication that, you know, that's a sustained shift towards longer commutes. Um, as a part of that, we are seeing the trend of, and many even listening to this podcast may be experiencing this. We went from 100% working from home to companies saying, hey, now we're gonna do a hybrid situation. We need you coming in two to three days a week. To now, many have moved from that to 100% in office. Okay. So I think that lines up. Uh, you know, longer trips. Why does anyone care about this, right? Longer trips can mean more wear on your brakes, tires, wipers, fluids, and other maintenance categories, um, shortening service intervals and creating more opportunities for aftermarket engagement. So I think that's a nice nugget there. Um, summer and holidays are peak driving seasons. I don't know if that's really uh a shot there, but nothing has changed. Um summer is still the busiest driving season with peak mileage and trip duration around July 4th. Uh, that has not changed. Um acceleration. We talked about those risky driving behaviors. No, we've we've seen um rapid acceleration, or say sudden acceleration. And if anyone's just curious about that, so think of that as like an increase in speed, eight miles per hour per second or greater.

Okay.

That's something like say mashing the gas.

Sure, sure.

That's that. Um, but uh that's that's signal signaling increased stop and go congestion. Um that increased in 2025, especially in the spring. Um and uh also just even timing around that, which to us that was an indication of rising local congestion uh and a return to the kind of that stop and go driving, which is also tied to school schedules and commuting patterns. And um again, that stop and go. First thing mine goes to again, wear and tear on the brakes, um, cooling systems, transmissions, batteries, et cetera, which could potentially drive up demand for those types of services. Um what else would I share? Um yeah, those those are the main things. I guess the one other thing I'll just kind of mention in observation of just being in this after auto aftermarket space and in talking about collision repairs and things like that. There appears to be, don't quote me on this, I was saying appears to be that um individuals are not getting their vehicles repaired um as much as they used to. Um appears to be a slight downward trend in that space, meaning that are there fewer accidents? No. Those are still happening, but are individuals getting their taking their cars to the shop and getting it fully repaired when they, you know, um when that happens? Not necessarily. They're doing it.

Just after a collision or in general? Like for oil changes, brake pads, things like that?

No, no, no, no. Collisions. Um I'm talking about collisions. Um, yeah, not regular maintenance of a vehicle. This is um collision repair um at any level. A DEM, uh, you know, right, a paint scratch, whatever level. I'm speaking in that context, um, to where individuals are either one, they're not filing claims. They just say, hey, I don't want to mess with it, I don't want my insurance to go up, or um I'm gonna file a claim, um, but I'm not gonna get the repairs done, right? I as as uh I I can decide whether I want to get the repairs done or not. I could pocket the check if I want to. I have a right to do so as a consumer. Um, but um yeah.

And is that is that observation based on your view of I guess uh driving behavior and resident like uh dwell time at a collision repair shop?

No, no, that's more so in context of um we we work with many different partners in the space that are in the collision repair space. And so, you know, we're talking about our data and how our data is used, and I guess since we're talking about that, I even throw another use case out there. Again, data in aggregate and counts. Our data can be used as a predictor. So if you are doing sales forecasting and you're wanting to understand supply and demand, maybe you're doing sales forecasting, maybe you're forecasting from a manufacturing standpoint, well, our data can be used as predictors. Why do you care about doing that? Well, the name of the game is to get lift in your model, right? Um, so I'm I'm speaking in the context, again, of relationships and partners that we have in this space that have shared some of these insights and um, you know, those observations. So sharing it with the masses.

Interesting. Yeah. I mean, there's so much that I feel like we can unpack. We don't have to go into it now because we're coming up on time, but um, even some of the things you highlighted from summer travel. I think about things that might cause somebody to drive versus fly. Maybe airplane tickets are up. And some of the things this covers all the three uh scenarios you highlighted, but um the the general economy, how are consumers doing? Maybe they don't want to fly because it's too expensive and gas prices have been fairly steady, so it's economical to drive, tough economy, a lot of uncertainty. Inflation generally has been higher, or I mean it's it's come down to 3%-ish, but prices are still continuing to lift, and that could perhaps go into the um the collision repair consideration, too. Whereas like food costs a little bit more, you know, I had some unexpected, unforeseen minor fender bender or a little dent or ding. Maybe it's not so critical for me to save or to make that expenditure.

Now, this is not an analysis that was done. Um, this is purely on my personal observation, just as a consumer myself. I can tell you absolutely ticket prices have uh airline ticket prices have gone up for me. Okay. Pre-COVID, I have not seen airline prices return to what they were. I'll give you an example. Um, I like Las Vegas. I go periodically. What used to be a $350 round trip ticket for me from Houston to Vegas, on average now is $700.

That's significantly higher.

And that's from personal experience, and it has not changed. Um, so yes, airline tickets and and really I feel like just travel in general has become more expensive, hotels and so on. So, yeah, it it it's a factor.

What Comes Next In Telematics

And I think the uh first, thank you for sharing that. Um I feel your pain on airline prices. So you you got me there, my friend. Um the but to the uh and and again for uh just for as I as you were talking about the increased commute times, a couple of factors, like you said. It could be the return to work, it could be perhaps people have moved out further away from city centers. And so I can imagine a lot of interested parties looking at this from a housing development standpoint, um promoting uh mass transit. Um you mentioned the automotive aftermarket. So a lot of interesting kind of developments or interested stakeholders for the type of data that you're highlighting. So thank you for those examples. Um one thing I want to touch on here before we start wrapping up is sort of the future of telematics data. What do you expect in the next couple of years? Are there gonna are there new things in the lab, new things that we can look forward to from uh what data are accessible to us for analytical purposes?

Um what I would tell you is that we will continue to enhance the data products that I mentioned. They will continue to get better and better. They will grow in terms of additional insights and attributes that would be made available. I'll give you an example for, like we talked about again, this historical traffic on road segments. What about intersections, right? Specifically these road segments at intersections. That's an additional set of data that we're actively, feverishly working on to provide insights around intersections. So, what do I mean by that? So, like on things like cues, how long is the queue of buildup of traffic at a road segment at a given time? How many count, what what are the turning counts, right? So think of a four-way intersection and for any given time of day, let's say an hour window, and how many went left, straight, or turned right? That's important. It's especially important for those in the um fleet management space, route optimization, right? Um, those entities really care about the number of left turns that their drivers are taking. Right. Because left turns are the most dangerous, especially ones that are unprotected. So they want to minimize them. So that's something that uh we are are looking into. Um we are looking at furthering um unpacking the road traffic analytics that's around the visitation and things like that. And where am I going with that? Where I'm going is that um, again, in aggregate, but it's very interesting to start understanding what are the aggregated patterns of driving. Oh my customers they go to AutoZone, but they also go to Walmart. 50 I'm making this up. This is all hypothetical, but I'm just giving an example. Oh, I've identified that 50% of my customers also go to Walmart, right? And they also go to um trying to think of something else, some some other repair type shop, whatever it may be. Making that connection, it could go one or two ways. It could be treated as competition, co-opetition, right, or or really just a complementary um entity that I'm looking to build in location, or maybe it's a shopping center, and they're looking to write lease up all of the available space that they have. So they're looking for the most ideal co-tenancy opportunities, right? You want to have stores that complement one another. So again, with the type of data that we have, we can help identify, excuse me, those patterns. And it's those types of things that are on the horizon. On the horizon are, you know, I mentioned pulling in demographics from another data source. I believe that eventually we will get to a place to where, along with the visitation counts, we will also be able to share with you a breakdown of demographics on those visitations. So you would say, Hey, I saw um X a thousand visitors at this Walmart, and here's the breakdown of the demographics. Um, X percentage were between the ages of 20 and 24, and uh they have an average household income of this and uh the family size of this. That's powerful again, right? Because now you're really getting to these insights of not only what was the visitation, but what's the profile of the visitors that are coming to our stores?

And that helps you target your messaging and advertising and marketing campaigns much more precisely.

There you go. Bingo.

Anthony, thanks so much for sharing all these great insights. Before I start wrap wrapping up with any fun closeout questions that I have up my sleeve, is there anything else that you wanted to touch on before uh we go to that section?

Lightning Round And Custom Sneakers

I think that's plenty. Um we're in a fantastic space that's moving very, very fast. Um we're we're we are learning. We know a lot, but we're learning, right? Meaning that we're open to continue exploring. And, you know, anyone that's watching this podcast, and you have an idea, right? You have a thought, art of the possible. Could you guys do X reach out? We're we're glad to have a conversation. And you know, uh, you know, I'm the SC, right? So I'm gonna make sure that the the the a the salespeople don't hound you, right? We can keep the main thing the main thing. But yeah, I'll I would I would leave you with that before we uh uh jump into the fun stuff.

Sure. So um I want to just do a fun round of speed round questions here. So the lightning round. So are you ready? Yeah, I'm ready. Let's do it. Okay. Summer Olympics or winter Olympics?

Summer.

Pickleball or tennis?

Tennis.

Coffee or tea?

Coffee.

Beer or wine.

Well if I had to be that's tough.

Cake or pie. Pie. Taylor Swift or Madonna.

Madonna.

Isle seat or window seat? Window. Uber or left Uber iPhone or Android? And last but not least, football or football?

Football or foot football. Football.

Okay. Well, those are all the speed round questions I have. And I understand your shoe aficionado. Tell me a little bit more about that.

Yeah. Yeah, I am. So I'll try to keep it brief, but um I took art all through high school, graduated high school and didn't do any art for many, many years, but it's always been in me and I love art. I also happen to be what you call a sneaker head. For those of that that don't know, tennis shoes, sneakers, some people like me have a lot of them because we just love them. We like retro sneakers, we like new sneakers that come out, all of that. I was at home during COVID like everybody else, combing the web and trying to find something to do and I saw someone doing what they called custom sneakers. And what they were doing was they were painting sneakers. They take a white sneaker that's the canvas and then they paint it. Hand painted airbrushed amazing beautiful designs. Anything you want football team, university team, baseball team, basketball, um Valentine's Day, uh St. Patrick's Day, you get where I'm going with this. But it's custom so you can create whatever you want. So I started doing that and um found out that I'm actually really good at it. So here recently, you know, Airity is celebrating 10 years this year. And so for our CEO, um we I you know the team came up with the idea um to do a custom pair of sneakers for him. So I I did a custom pair and it's actually on my LinkedIn and maybe I'll get you some picks to even share if anybody wants to see but I love it. I'll put it up in the comments with the link to it yeah you know I um I this is something that I do I love it. It's a it's a labor of love. It's very it takes a lot of time.

So on average great art always does.

Closing And How To Support

Yeah that's right. So on average it takes about 20 to 30 hours for me to do one pair. And because the the devil's in the details um my shoes look like they came from the manufacturer. And you got to stop me because you know I'm talking about something I love so I don't want to just go on and on and on. But that that's what I do around the sneakers. If anybody wants to know more I I share my info and you know I love it.

Thanks so much for sharing a little bit of that passion with us. So beyond data, beyond telematics, Anthony is your sneakerhead guru. So Anthony, thank you so much for being part of our program. Thanks to all of our listeners and viewers we hope you've enjoyed and learned a little bit from this episode of indicators and until the next time we hope you have a great day. Thanks for tuning in to another episode of Auto Care on Air. Make sure to subscribe to our podcast so that you never miss an episode. Don't forget to leave us a rating and review. It helps others discover our show. AutoCare On Air is proud to be a production of the Auto Care Association dedicated to advancing the autocare industry and supporting professionals like you. To learn more about the association and its initiatives visit autocare.org