All right. Good afternoon and welcome to Samsara's Investor Day. My name is Mike Chang, and I'm the Vice President of Corporate Development and Investor Relations here at Samsara. Thank you all for joining us in Chicago today. It is amazing to see so many familiar faces. And for those who are joining us virtually, it's great to have you on as well. Before we get the show started, I have two quick disclaimers. The first is that we'll be making forward-looking statements during today's presentation. These statements are subject to risks and uncertainties described more fully in our SEC filings, our Investor Relations website. And the second is that we will be discussing non-GAAP financial measures today. These measures should be viewed in addition to and not as a substitute for GAAP. And a reconciliation can be found in the appendix of today's presentation. All right.
So at Samsara, we are digitizing the world of physical operations, and it is pretty incredible to think about the progress we've made in the last decade since we were founded in 2015. To give you a better understanding of the Samsara vision and where we plan to take the platform, I'm excited to bring on Sanjit, our co-founder and CEO, Kiren, our Chief Product Officer, Lara, our President, two of our very own customers with DHL Nutrien, and Dominic, our CFO. We hope that through this process, you can get a better understanding of the multi-decade journey we're embarking upon. We have about 2.5 hours together today, so please reserve all your questions for the very end so we can keep on track. With that, I'll pass it over to Sanjit to lead us through our vision for the future of connected operations.
All right. Thanks, Mike, and thanks, everyone, for making it out. I'm going to get things started, share a little bit about the big picture, but we're going to try to mix things up a little bit and do some live demos as we go. I'll get things kicked off. Kiren will do a few demos as well. But that way, you get a sense for what our customers see in the platform and where the value really is. Let me start with the very big picture, which is our mission. As a company, we're focused on helping our customers, the world of physical operations, improve their safety, their efficiency, and sustainability. These are big, asset-heavy, labor-intensive industries, and they're very large industries. We'll talk about that in a second.
But in terms of how we found our way to this, it's a bit of an interesting story. So my co-founder, John, and I, we started this company about nine years ago. Our previous venture had been bought by Cisco Systems, a big networking company. I think everyone knows Cisco pretty well. We were trying to figure out what we want to work on next. As we looked at the marketplace and we thought about where we could go have impact in the world because we still had a lot of energy, we still wanted to build, we noticed that a lot of the other folks in the industry were focused in two places: either the world of the consumer, helping revolutionize consumer experiences. By 2015, Amazon, Apple, Google had all shown what can be done with modern technology in the world of consumer.
So we knew a lot of people working in that area. We also knew a lot of people who had also been in enterprise. So we knew people who had worked at ServiceNow or Salesforce or Workday, building up those systems and transforming the enterprise user experience. But in our time at Cisco, we'd also gotten exposed to a different part of the world, which is the world of infrastructure and operations, the infrastructure of our planet. And we didn't know a single other technology team, at least in the Bay Area, that had been focused on that buyer, that customer. And so we saw it as an opportunity to have tremendous impact. Because if you think about it, the infrastructure of our planet is absolutely enormous. It's 40% of our world's GDP.
You've got billions of people who work frontline jobs, and we didn't know anyone building modern technologies for them. So we really started market first. We said, "What can we do to go have an impact in that world of physical operations?" And as we got closer, it became very obvious that this was an underserved market that was about to go through the digital transformation wave that we've seen in the other parts of the industry. And so when you go and visit physical operations, especially a decade ago in 2015, you would see pen-and-paper clipboards. You'd see people planning their routes using big maps on the walls, see them making phone calls to figure out where everyone was. And that, to us, was the opportunity. We said, "There is a better way. We know the technology can help.
We just need to build it." So that's what we set out to do, was basically build this platform for the customer. We always started with this idea of a platform, not a single application, but multiple applications and a big platform underneath. So let's kind of double-click a little bit. I'm talking about physical operations. When I go deeper and talk about the industries, these are the pictures you should have in your mind. We're talking about the companies that do waste management for all of us in our communities, the construction companies that build our roads and highways and our buildings, the chemicals carriers who are working behind the scenes, the transportation companies who run our supply chains, but also the food and beverage companies that feed us all. This is what I mean by the infrastructure of the planet and the 40% of global GDP.
Hopefully, this puts it in perspective. When you go and spend time with these customers, many of whom are here with us at the customer conference, you'll hear that they have some real-world problems they're trying to solve. You see a couple of them on the screen. As I mentioned, these are asset-heavy, labor-intensive industries. In some cases, they have thousands or tens of thousands of assets, and they're trying to figure out which ones need maintenance and repair. What order should I maintain them in? How much of my equipment is underutilized? I've met multiple customers that have billion-dollar balance sheets of equipment, and they're trying to figure out, do we need all of those assets? Do we need all that equipment?
Or can we get rid of 5% or 10% of it, maybe rent it instead of own it, and use that capital for a more productive purpose? They're also trying to figure out how to serve their customers. And it's not just in transportation. In field services, for example, they're trying to figure out, what order should I visit my customers in to keep them as happy as possible? Frontline workforce and frontline labor is a big front-of-mind concern as well. Because a lot of these industries, these are hard jobs, and they have a lot of labor turnover. So they're trying to figure out, well, how do I train everyone for this upcoming job that we have and get them retrained and up-to-date on the equipment we've got? How do they reduce risk? And more recently, how do we transition to electric and stay current?
All of these are problems that can be answered with data, but you need good data. You need a system of record for physical operations, and that's what we're building. I mentioned a couple of IT names earlier. These are companies you're all familiar with, and they've gotten to be significant players because they have built respective systems of records in their regions, whether it's ServiceNow building a system of record in IT, Workday building one for human capital management, Salesforce building a CRM. These are all big systems of record. But none of these companies, and really no one that we'd encountered, had built a system of record for the companies I mentioned earlier for physical operations. So that's what we're building. As we kind of go a little deeper and look at the architecture of our connected operations cloud, it looks something like this.
The idea is that frontline operations and physical operations have a lot of assets, and they have a lot of people. You want to get data about everything that's going on out in the field. That's what's coming in the bottom here. It's data coming in from those sources as well as third-party sources. In many cases, they have other systems that are already online that are siloed. We're breaking into those silos and exchanging data via APIs, bring all that data into the Samsara platform, which is where we can go and analyze it. That's where we find insights in the data. Rather than presenting a customer with a big, long report or an Excel spreadsheet, we want to go and find the insight in that data for them, surface it, and we can use modern AI technologies to do that.
We unlock the value in that data through those insights and help our customers take action. In some cases, that action's happening in the moment, in the cab. We can train an AI model to detect things like distracted driving or not wearing their seat belt or tailgating. In other cases, we're helping surface long-term behaviors, eco-driving, for example. How can they help their drivers save fuel when they're out on the road? These are all multiple different applications. We'll talk through those today. The idea here is not just to think about the frontline, but the back office as well. Really, the entire operation, unlock a ton of value for the customer, which is, in many cases, worth millions and millions of dollars.
And that gets them to want to invest more and more in the platform, get more data about their operations, and get this flywheel really spinning. So as I mentioned, we're about a decade in. And the first application that we started with, the first use case, was around tracking these vehicles out on the road as telematics. And as we built that real-time GPS tracking platform or application, we started building out the platform. We expanded in concentric circles to safety, which is now our number one application, often sold right next to telematics. And then from there, we added equipment to the platform, workforce. And now we're thinking about the overall future of connected operations.
For those of you who saw our main stage announcement, you can see how we're expanding every single year with this platform, doing more and more for our customers, tracking more assets, digitizing more workflows, and really trying to bring their entire operations into this modern cloud age. So what I'd like to do next is go from the abstract to the real. I want to just kind of open up the dashboard, show you the system that our customers use, and then hopefully that gives you some context of how they're seeing value. I'll click around to a few different use cases, but this is kind of the background. We'll come back to this demo over the course of the day and show you a little bit more. All right. We can flip over. Okay.
Just to orient everyone here, this is the dashboard system that our customers typically log into. We have about a half dozen customers who let us demo their operations live. The amazing part about this is it's a very powerful selling tool because operations companies can recognize these are the real-world challenges they're facing. I'll flip through a few different customer examples. The first one I'll start with is a food distributor who primarily operates out on the West Coast of the U.S. They're actually up and running right now. They're basically moving food from distribution warehouses to retail locations like restaurants. For them, they have fairly complex operations. They have the distribution centers. They have hundreds of trucks. They have hundreds of drivers. They also have the trailers. They want to make sure that everything is showing up on time.
It's being kept at the right temperature, which is important for food safety. In this business, they really care about customer service. In other words, they want to show up exactly when they're supposed to because someone needs to be there to receive the food, unload it, and so on. What we're looking at here is the real-time tracking. I'm just going to click in on a vehicle here. This was the first use case or application that we started with, which is simply understanding where this distribution truck is. Now, this seems like a pretty straightforward use case to, I'm sure, most people in the room. If you've used an Uber or Lyft, right, you know that you can see things in real time. Well, when we started, this was not the case.
Most people were making phone calls to figure out where their drivers were because the state of the art back in 2015 was to get breadcrumbs, a little dot on a map that would update every couple of minutes. That's still, in fact, what a lot of our customers use. So they may have telematics or prospects use. They may have telematics, but it's not real-time telematics. It's not as data-rich as what we're showing you. So we can see on the right-hand side the map of where this vehicle is. We can see the traffic conditions, and we can zoom out a little bit. These folks are out in California. There's not typically a lot of weather, but if there was, we could do a weather overlay and see if there was a big storm overhead.
This is all important from an operations perspective because this is what causes disruptions. On the left-hand side, we can see who's driving, the trailer that they're pulling. That's where all the food is, where they are, a view out the front. We can even see that they're currently on a job. They're on a route. So we can see some details. I'll come back to this in a second. We can see details about the engine itself as well as the vehicle. So in this case, we can see, is this driver's seat belt buckled? The engine computer knows that. The dash knows that. We now know it too. You're in the cloud. You can see the fuel level, the odometer, and a bunch of other diagnostic information. I'll come back to this as well when we talk about maintenance. But we can go deeper.
This is what makes Samsara different and unique. Most of the competitors in this space have been point solutions. They would give you GPS tracking, or they might give you a compliance solution. We brought it all together in one place because that's what the customer cares about, their overall operation. I'm just going to click in here on the trailer that was being pulled. Again, we can see that. This is the little square that you see. You'll see here on the left, it says reefer controls. This is a refrigerated unit that's running on the highway. You can see that it has a set point of 34 degrees. They're running food. They care a lot about the food safety, and they want to make sure everything's kept at temperature.
We can click in, and we can actually see that the trailer is full. We can see the actual temperature. We can even view some additional details. It turns out it's about 110 degrees outside. So it's very important that this thing is working. If it wasn't working, if the driver had forgotten to set it, which was a manual process before Samsara, that load would be ruined, right? They'd have to write it off. That would be a huge expense for them. So this is the kind of visibility that wasn't possible until Samsara showed up. We can even show them that the door is closed in the back. Sometimes what happens, people are making these food deliveries, they forget to shut the roll-up door. The hot air comes in, ruins everything in the back. So again, this is all about operational visibility and context.
By the way, everything that I'm showing here is also available via the APIs. I mentioned that they're running routes. They know when they're going to pick up something that's refrigerated. They know when they're going to pick up something that's frozen. They can set the temperature set point via the API, have everything happen automatically behind the scenes, and the driver no longer has to remember. They can even set the set point in advance. So if they're going to go pick up a load of ice cream, it might take two hours to come down to temperature. If it's 110 degrees outside, it takes a while to cool off those trailers. They can set it in advance so when they show up at the distribution center, they're ready to go.
If they had to do it manually, they might have had to sit in what's called detention, sit in the yard for a couple of hours, lose that productive time during the day just because they hadn't kind of orchestrated everything. This is the fleet side. This is also the trailer telematics and equipment monitoring side. I'm going to flip around here and show you a couple of other things. As I mentioned here, this individual driver is on a route. We can click in. We can see route details. This is a pretty short route, but you can see when they were supposed to be somewhere and when they actually showed up. These routes can run hundreds of miles. Like I said, sometimes there's weather, there's construction, there's a road closure. They might have a case where they're interrupted.
They don't make their stop on time. Well, in this business, you really want to show up on time. And if you're not going to show up on time, you want to automatically notify the end customer. You can do all of that from the Samsara system. You can live share a link. You can do it programmatically. You can send emails. And we've seen that where our customers will wire this up. So the distribution center automatically sends an email, a notification to the end customer that we're on our way and here's what's coming. We can see plan versus actual. We can see how much time was spent at the stop. And in the case of a multi-stop route, you can see this stuff cascade over time. We can aggregate all this as well because it's not about one truck, one driver, one route.
It's actually about how's the system operating. Our customers can see, okay, are we seeing some seasonal differences? Sometimes there's more traffic or less traffic in a certain season. We can see it by driver, which ones need the most attention. Maybe these routes need to be replanned. All of this information is now in one system. This is completely different than what customers were used to. Okay. I'm going to flip to a little bit of a different context here. Sorry, I have a multi-stop route here in case we want to take a look. I think you guys get the idea. I'm going to flip to a different context, which is around safety. Now, we have all kinds of different customers, many, many different industries, but they all want to be safe.
Whether you're in construction or logistics or food and beverage, everyone is trying to minimize the risk of accidents out on the road. At scale, these fleets will spend $millions and millions dealing with those accident claims. The best thing that they can do is train their drivers to be safe on the road and break bad habits, basically prevent people from developing bad habits, like looking at their mobile phone, which reduces their or increases their reaction time. This is another company that operates ready-mix construction equipment. We're talking the big cement mixers you see on the road. For them, safety is really important, first of all, because they don't want to get into accidents, but second of all, because if you tip one of those mixers, it's devastating. The cement dries. You need to get jackhammers out.
It's a very expensive operation, and it's not a good look for their brand. So they have a lot of reasons they want to be safe on the roads. And in this case, this is a regional company. They have a few different operations. So we've got Nebraska and Iowa, it looks like. And they have a few cement mixers per plant. So they're regionalized, in other words. They have different general managers. They make independent hiring decisions, but they want to be safe across the board. So in this case, they're able to see the safety of their entire operations. They can see what the relative safety score is. And as you can see, this is a very safe fleet. They really take this seriously. We can look over the last 90 days if we want. And they're maintaining a 99 or 100.
This is exactly where they want to be. In case anything is changing, they can see it all here. They can see the trending information. They can do this not just for the entire fleet, but the way they operate, the way they're organized is by plant. They can go click in and say, "Okay, looks like distracted driving is getting better, but maybe we have a problem with harsh driving." They can go in, click on that. They can see for a given plant how they're performing over time. This is the level of visibility into risk and safety that customers like this simply didn't have before. This is what I mean by insights. We're taking in all of that GPS data. We're taking in the video data from the cameras that are out in the cab.
We're running the AI models on it and looking for that trending information. Like I was showing you earlier on the previous screen, we can see which driver was behind the wheel because the risk isn't actually tied to the asset. It's tied to the operator of the asset. So in this case, we can see, okay, are folks in the Ames plant doing well? We can actually see their score and their trending. We can see if there's a dip. We can go down and dig into, well, where are the people that maybe need a little more attention? So a lot of detailed information there. Then if there is an incident, which is statistically likely to happen when you have hundreds or thousands or tens of thousands of vehicles in your fleet, you can gain visibility into exactly what happened.
Let me show you what this looks like. This is an event that happened on one of their mixers. I'm going to play it for a second just to give you all some context, and we'll talk about it. If we can turn on the audio, this specific one has a little bit of in-cab audio. Some great tunes going. Harshness detected. So I think there's a little delay in the audio here. But what you can see is that this is a guy driving a big cement mixer. Someone cut him off, right? They went ahead and went through that stop sign. There's no way that this guy could have stopped in time had the guy been a little bit closer. This driver has probably been doing this for 20 or 30 years. He did all the right things.
You can actually see, was his foot on the accelerator or the brake? We can see he took his foot off the brake. We can see when it was in the video. We can see he had to do it. He had to hit the brakes pretty hard, and that created a harsh event. Everything is okay, right? This was not a big deal. Maybe a quick public service announcement. If you do see a heavy vehicle on the road, do not cut it off. They cannot stop. This is the kind of thing that in previous systems would have shown up as a harsh event. Harsh events typically are used as a demerit, like you ding the driver's score. In this case, the driver did a really nice job, and you want to recognize him.
So what we're able to do is run a whole workflow here. We know who is driving. Most of these companies, they have safety managers, safety coaches. This video got automatically routed to a coach. You can mark it as a recognition opportunity. So usually, they want to go recognize their drivers every Monday on the safety call. You can see that. We have the whole workflow here in the system. You can see when it happened. So it was just a couple of days ago that it was a harsh, but also defensive driving. So this is the incredible part about AI is no one's watching these cameras nonstop. It's the AI models running at the edge, and we're able to show them all of that. And so you can see the whole workflow here on the right-hand side. You can see the context of where it happened.
And then you can even see all of the workflow details of, okay, it was marked as, or it was first viewed, it got marked, needs coaching, and who all saw it, and so on. You can even share this with the driver because some of these drivers don't come back to an office or a home terminal. They're out in the field. And this is something that is, again, changing the game because it wasn't possible before. It used to be all the coaching happened in the back office. Now you can send it to their tablet or their mobile device with the video and recognize them, or in some cases, say, "Hey, we need to work on your defensive driving," or, "We need to work on mobile phone distraction or wearing a seatbelt." We've got courses for that built in. It's an automatic workflow.
We talked about this in the main stage, but we're now rolling out training, and it's going incredibly well with our customers. We're just saying this is making a difference to our risk. By closing the loop, by actually breaking these habits as they're happening, we really do change risk out in the field. So I'm kind of going fast here. I want to show you the entirety of the platform, which is pretty broad at this point. But flipping gears one more time, we've also got a lot of information, not just about drivers and risk, but about what's going on with the assets themselves. And we're tied in at a pretty deep level. So when we are on a truck, for example, or a different kind of vehicle, we get information about the engine, how well it's performing.
You saw the diagnostics codes and the seatbelt codes and all that kind of stuff. We can even see how much fuel they're consuming. And for some of our customers, their fuel bills are expensive. We're talking about $30 million-$40 million a year. So if they can find a way to reduce their fuel consumption, that might be reducing engine idling or coaching the drivers on eco-driving or even swapping out the asset, basically buying a newer vehicle because they have better miles per gallon, it makes a huge difference to their bottom line, their profitability. So they're all looking for ways to drive up their efficiency. And some of these customers are even transitioning to electric, especially in the light-duty side of things.
So if you have a van or a work truck and it's only driving 40-100 miles to see customers, you might want to move that over to electric because it costs less to run. So we can see all of that here. And we can see it, again, organized by tag. This is a customer that has a couple hundred, maybe 1,000 vehicles. We can see all of the fuel efficiency across their fleet. And you'll notice here it shows you exactly how much fuel they've used. And in the cases of internal combustion engines, we'll show you the carbon emissions. We can even do this for multiple fuel types. So in some cases, they're running biodiesel, has a different carbon footprint than traditional. You can see the cost. And in some of these cases, we're talking just about a month of use.
They're spending thousands of dollars. This is helping our customers be much more data-driven. One of the things that is also important to recognize is that we are sitting on an aggregate platform. We can see data for millions and millions of assets in aggregate. That helps us provide things like benchmarks. Many of our customers are wondering, "Hey, should I replace this vehicle or am I driving it wrong? Do I need to coach my drivers?" We can see all of that. We can say for a given vehicle, this is a GMC Sierra. It's a 2017. The benchmark for other folks in this industry is 13.75 miles per gallon. We're only getting about 10. Maybe we should do something about this. All of this is around driving action, giving them all this data at their fingertips.
So bit of a whirlwind tour. There's a lot more that we can show you here. We can do things like, well, if I'm trying to figure out which vehicle should I transition to electric, we can do the data analysis for that. This isn't a project for an analyst. This is just something the system does for you. In this case, we can tell you based on the mileage, how many miles they typically drive in a day, what their maximum range is, and the fuel costs, which ones make sense. In this case, you can see vehicle 1548, it's driven 500,000 miles. It gets 5 miles per gallon, not the best. It only goes about 84 miles on average. And the max it's driven over the last couple of months is 140 miles. So it could definitely be electrified.
They would save somewhere in the order of $23,000 on that vehicle. They'd save a lot of carbon emissions. So it's excellent from a transition perspective. This is the kind of data-driven approach that our customers love because it helps them make smarter decisions. Okay. So I covered a lot of the platform. This is a bit of an introduction to Samsara. We'll come back to this in a little bit. But hopefully, that gives you a sense of how our customers use all of this data in an operational setting. So Mike mentioned we're on a multi-decade journey. That's how we view things at Samsara. We're building for the long term.
The first roughly decade or so of our journey so far has been about establishing that product-market fit, establishing that foothold, not just with one application or in one industry, but across all these different physical operations, industries. Our entry point was real-time telematics, that GPS tracking. Now we've started expanding the platform, as you all know. As we work on phase two, this is really about that concept of a platform in a single system. It's an open platform. If you all have a chance, we have an expo hall. There's 290 partners that are now part of our ecosystem. We're able to exchange data with lots of different systems and proprietary enterprise systems. But the idea is now you can have multiple apps running on it, get multiple sources of value for that data.
Over the long term, we see some exciting phenomena that are coming. We see AI continuing to develop. These models are getting more and more powerful. We'll talk about that. We also see autonomy being deployed. In operations, this is vitally important because they're always shorthanded. They always could use a little bit more labor. They're always trying to operate more efficiently and around the clock. Autonomy enables that. We're starting to see the beginnings of that. We're very much aligned with that vision. I've mentioned data a few times today. That is really the core of what we do. It's what powers our flywheel. We measure a bunch of different statistics. A few of them are shown on the screen here, but we look at how many miles are being driven on our platform.
In the last year, we saw about 60 billion miles driven on the Samsara platform. That's the equivalent of crossing the country every second or two. It's a tremendous amount of visibility over our roads and the behaviors that come along with that. We saw about 75 billion API calls between systems. That also counts as data that comes in. You saw the routes, for example, that were in the food and beverage distributors system. We're able to pull that in. We're able to pull in the plan versus actual times. These are all data points. And that totals up to about 9 trillion different data points. And again, this is about solving real-world problems. How do we help our customers reduce risk? How do we help them utilize their assets more efficiently?
If you have a bunch of trailers, but 20% of them aren't being used, you probably don't need all those trailers. You can even figure out how to transition to EVs. So these are the kinds of real-world problems we want to solve. And when it comes to AI, AI is becoming more and more powerful. It's helping us find deeper insights in all that structured data that we have. And our data is not available online. These customers aren't posting all of their routes and their fuel efficiency and so on onto online sources. So we have a proprietary data asset behind the scenes, but we can use a lot of these new models and train them and fine-tune them on the data we've got. And this is kind of where we're going. We are looking at ways to go and measure risk in a deeper way.
We see all these roads that are being driven. Can we understand which intersections are the riskiest? Absolutely. We see that behind the scenes. We're looking at labor and asset utilization. Can we help you operate smarter by maybe replanning how you're running out in the field? Can we improve your energy efficiency, maintain compliance? All of these are data problems, and they're solvable using all of these new techniques. So what I want to do next is give you a little bit of a preview of how we're thinking about data in our system. And sorry, how we're thinking about applying some of these latest AI technologies in our system. And so let me go back to the front here. I'll just think we have a tab. So I think at this point, everyone is familiar with generative AI.
You've all kind of used ChatGPT or one of those new products. And so I won't start there. And we'll just assume that you're familiar with voice recognition and so on. What I do want to do, though, is highlight how those technologies can be applied in an operational context. This is a technology preview. This is not something that we are shipping in a general availability context, but we do have it in front of a number of our customers and gotten some pretty good feedback. So what I'm going to show you is a different way of interacting with the reporting interface and the data that you saw earlier, which is through natural language. Now, we can do some of the basic stuff you're probably familiar with, like querying our knowledge base.
If you want to know how to use our product or how to do compliance, you can do all of that. But we can also go into some of the operational data. I'm going to click in here. This is a fleet, by the way, that's another one of our reference customers. They're local to Chicago, so I thought it would be kind of fun to show them based out here. We're going to take a look at some of their operational data using this interface. Let's give it a shot. Create a report of weekly engine faults over the last six weeks. Okay. It went out, queried a bunch of the data behind the scenes. We can see what sorts of maintenance codes are being thrown off of these engines. We're talking about 2,000 assets out in the field.
And you can see trending data and so on. And our customers love to be able to explore this data, and they like to be able to interact with it and ask questions of it. So I'll do a little bit of that. Show me the top five vehicles that need servicing today. Okay. So here, it went out to the kind of maintenance backend and was able to look through all the different vehicles that they have. And it looks like there's about five vehicles here that might need some attention. And we can see some of them are Freightliner, some of them are Max. And I will go a little bit deeper. I think the top one is HST 245. So let's ask about that. Create a report. Oh, sorry. For a vehicle HST 245, which fault code should my mechanic look at?
And by the way, you can do this in Spanish. I'm not so great at Spanish, so. Now it's going really deep. It's going and looking for the fault codes for that individual vehicle. It's pulling them all off from the last time it was seen. And it's looking up for that manufacturer what those fault codes mean. So it's not just like a number, but it's actually decoding. This is an aftertreatment SCR, which is a kind of emissions fault. It's giving us a lot of details there. It's telling you the severity. So again, these were the five that really need maintenance. Some of them are informational. Some of them are moderate. One of them is severe. They all affect emissions. This is the kind of data the mechanics in the shop love because it tells them which parts they need to order, which ones need attention.
A lot of the times, drivers don't report a check engine light because they just want to get on with their day. But this is a vehicle that would break if they didn't do anything about it. So it gives them a lot of visibility into that maintenance and diagnostic information. Fleets like this, as you can see, they're pretty large scale. And let me zoom in here. And you can see they've got thousands and thousands of vehicles. And what am I doing? There we go. So they're also every morning doing an inspection report. They do a walk-around inspection before they start the day. They do an end-of-day inspection, just called a post-trip. So we can also ask the chatbot about those kinds of things. So let's try that out. List unsafe inspection reports in the past week.
This is a pretty big query because there are thousands and thousands of these that pile up every single day. It's going through all of them, figuring out which ones were unsafe, and organizing them for us. We can see all of those details. This is great because you can actually start to click in. I'll just open these up. You can see some image details. You can see the walk-around photos. This is all at sort of a mouse click away. We'll come back to this as well. But in some cases, you don't want to see a report. You actually heard about something. You want to go dig in. We can actually get multimodal models running on this as well. There's a bunch of text from the workflows, of course. There's images being taken by the phones.
You saw some of those earlier. So we might want to know. Hang on one second. Show me inspection reports with images of broken mirrors. This is the kind of thing that gets reported to the back office pretty often. So let's see how many of those there are. It's running that same kind of query, but going through all the images. And let's see. Here's one. That's definitely broken. So click on that. Okay. So this guy, this got resolved back in May, but he hit a trailer parked on the side of the road with a mirror, and it shattered. Not good. Mechanic now knows I need to go order a mirror assembly. Really useful, really practical, and the kind of stuff that our customers are always trying to find out more about. Okay. So these were the images from the walk-around inspections, from the pre-trips.
Let me click here. And we can also look at all different kinds of images that are in the system. Show me images of driving in heavy rain. And this is related to the dash camera, so we're able to look out the front. And again, semantic understanding of all the scenes. And it's pulling a bunch of rainy pictures. And you can see it's actually been pretty nice weather here, but we can click in and see more details there. And it's not just image data because, as you know, our cameras record video. So if we take a look, let me do a different one. Show me videos of cars braking hard in front of my driver. I told you all not to do this, so you can see how often it happens. So this is now going out and looking at the dash camera footage.
And you can see it happens in all different scenarios. We can mouse over. And what we're doing here is actually pulling data that's related to the Ford collision warning that's running in the camera, analyzing that. And you can see here we've got lots of different videos. The hotel network's a little bit slow, it looks like. But you can see this white car pulls over in front. He's obviously trying to take an exit or something, hits the brakes. So again, this is exactly what we asked of the video. We didn't have to do this through a reporting interface. Basically, anyone can pick this interface up and run with it. And then our customers are finding new and creative ways to look at this stuff too. So they want to know things like, let me try another query here. Show me videos with road construction.
So construction is very commonly something that disrupts operations, right? It slows you down. So we can see videos of that. It's going to pull here. And this isn't just looking for orange cones, right? So you see lots of different kinds of construction here. Let's mouse over for a second. Let me give it a sec. You can see construction equipment off to the side of the road. We've got orange cones. We've got all kinds of different environments. So what's amazing about this is the power of these models is getting more and more powerful every single year. We're able to apply it in the operational context to all this proprietary data that wouldn't be out there. So hopefully that gives you a sense of different ways that AI can be used to really turbocharge connected operations.
But for us, this is exciting because it takes advantage of this incredible data asset that we've been building up over the years. Okay. If we can flip back to the slides. So I've mentioned a bunch of different use cases, and that's what the platform was designed to do. Really align with the world of physical operations and solve their real-world problems. And in operations, operations industries have been around forever, hundreds, if not thousands of years. And over that time, they're always trying to find ways to be safer in their operations. They want to basically make sure no one's getting hurt on the job. So they want to go identify operational risk, train their teams, coach them. They always want to be more efficient. That makes them more competitive. And they also have limited labor, so they have to be very efficient with the people they have.
We're doing that through workflows, through chatbots, all that kind of copilot technology that we're just demoing now. They want to be more sustainable. Most of these industries are energy intensive. That could be in the form of fuel, for example. They're spending $tens of millions. How can they consume less fuel in their operations? That could be from reducing idling, could be from coaching their drivers on eco-driving. They just want to limit the amount of energy they're wasting because it hits their bottom line. Now many of them are reporting carbon emissions. They want to reduce those numbers across the board. For us, we're excited because this is an evergreen set of challenges. Customers in operation will always want to be safer, always want to be more efficient, always want to be more sustainable.
We think data and these modern technologies will be able to help them. From our perspective, we want to be that platform. We want to build everything out so they can run on Samsara, make smarter data-driven decisions. To do that, they're going to need a system of record for physical operations. Okay. I would love to show you all a site visit video of our customers. So because we can't bring everyone on site, we thought we'd bring a site to you. This is one of our customers, Liberty Energy. They're in the oil and gas industry. Watch a short video, then we'll turn it over to Kiren to tell you more about products.
We run fleets throughout North America, and each one of those fleets is the horsepower of a 747 jet engine. Yet we've got to drive it across bumpy dirt roads to remote locations, assemble all those components together under stringent operating conditions in all sorts of outdoor physical environments. To choreograph that dance in real time with robust data is essential to us. It's always been a mission of Liberty's to become better. How do we bring the industry, not just Liberty, to the next level?
When we can see everything in one place, our entire operation comes together. When we were looking for a new telematics provider, we were hoping for a reliable, easy-to-use platform that our drivers and managers liked. We also wanted a company that we could grow with. We quickly learned that we could fully integrate our payroll systems, our maintenance software, our tax structure, monitoring our third-party drivers, all in a simple, easy-to-use platform.
The Samsara integrations into our maintenance management system have freed up a massive amount of labor for all of our different managers. We can have our field mechanics focus only on those 40 units that are out on those locations and not get distracted by the 3,200 units that are all over the country. Using Samsara's location data, we can easily pull the sales tax rates in any given spot where our trucks or tractors are located. We can get us a very, very accurate tax rate. This data from Samsara is going to save probably $10+ million in savings for the bottom line of Liberty. Our relationship with Samsara has only grown. So where do you think that might go in the future? We'll certainly grow and deepen that partnership between us. As trust builds and knowledge builds, that's the ingredients to innovation going forward. There are partners in building and making a great business.
All right. Good afternoon, everyone. My name is Kiren Sekar, Chief Product Officer at Samsara. And you just saw Sanjit demo the Samsara platform, and you saw the breadth of the products on that platform. You heard Sanjit talk about the scale of data that we're operating on. And now you just heard from one of our customers share how that turns into impact in their business: $10 million in savings. What I'd like to do now is share with you three of the newest products on our platform that we announced earlier today.
All of these products were made possible by the flywheel that you heard Sanjit talk about, this flywheel of data enabling insights with AI and translating that to action through our software, which in turn makes customers want to bring more and more of their data into our platform. This flywheel actually underpins our broad product expansion strategy. We're focused on building new products that bring more operations data into our platform. Our customers operate diverse operations with many different types of data, and they're looking to bring more and more of that into our platform and rebuild new products to enable them to do that. We also build products that surface new insights from that data using technologies like AI, taking all of that data and finding opportunities for customers to operate smarter.
Then we enable our customers to act on those insights by delivering new products into a single pane of glass that our customers are already working out of every day. Our customers are operating large teams of frontline workers, and having that simple integrated experience is important to them. It's how they can adopt technology at scale, and our product strategy enables them to do that. So let's talk about what's new, starting with a product that's oriented around data. If you visit one of our customers' operations, you'll see that they often look something like this. They're very asset-intensive operations with a broad, diverse set of assets that they rely on. So we started working with customers by helping them have visibility into their vehicles, giving them real-time GPS for the first time.
And from there, we expanded into video safety, into connected equipment, enabling them to have visibility in more and more of their operations. And for many of these customers, they were getting visibility into their equipment with Samsara for the very first time so they could understand what was going on with their heavy construction equipment or even unpowered equipment like an intermodal container or a roll-off dumpster. And we heard from customers that the impact that this data was having on their businesses. For example, we would hear of customers who would be able to successfully recover lost or stolen assets. And we'd hear about customers who were able to use this data to identify unutilized equipment and move it to a site that needs it. Whereas before, without data, they'd go buy more units.
In the process, they'd be able to save $tens of millions of CapEx. But we also heard of a lot of cases where customers struggled to manage assets in their operations that just wouldn't be practical to connect over the cellular network. These would be assets like a toolbox or maybe specialized equipment like a fiber splicer or maybe even a drum of chemicals. We hear stories like a customer who operates a large crane rental organization and crane servicing, and they use Samsara to track their cranes, their trucks, and make sure a crane gets on site where it belongs. But sometimes that crane is missing an attachment, and this one piece of missing equipment means that an entire work crew is sitting idle. That one instance, that piece of missing attachment can cost them $100,000.
We heard about a customer who had specialized engine cowlings that they were moving from one site to another. These things cost $500,000 each. They shipped six of them, and only five arrived. We hear about customers who lose $tens of millions of dollars of equipment out in the field every year. We wanted to enable customers to be able to have visibility into all of these types of assets in Samsara too. We asked ourselves, how could we bring visibility to these smaller assets? We'd have to build a device that was small enough and efficient enough that our customers could put it on practically anything.
So when you look at the IoT devices that would connect larger assets into a platform like Samsara's, you'll see that the biggest part of them, the biggest component in these trackers, is actually the battery that's used to power the cellular radio and the GPS system. So we started thinking, how could we get rid of the GPS? How could we get rid of the cellular radio? How could we get rid of the big battery that powers them? And the aha moment came when we realized that there are already millions of connected IoT devices that are connected into our platform and are collectively reporting millions of GPS points every single minute. These happen to be on our customers' vehicles, on our customers' trailers, on their powered equipment, and they are everywhere that our customers operate.
So we thought, if we could build a device that could piggyback off of this network, it could connect to a nearby Samsara IoT device. It wouldn't need to talk over the cellular network. It wouldn't need GPS. And that means we could get rid of these components and the battery that powers them. And that means that we could build a device that's small enough and efficient enough that our customers could put Samsara on just about everything. Well, we built that device. It's called the Samsara Asset Tag, and it looks like this. As you can see, this thing is really tiny, but it's really tough. It's ruggedized for our customers' harsh operating environments, and it communicates to any of the nearby Samsara gateways using an industrial-grade Bluetooth. This is an efficient technology that's far more powerful than consumer Bluetooth that you'd have in your personal devices.
So that means it can work through concrete and metal and all the things in our customers' operations. We think this could be really impactful to customers as it can help them improve their asset utilization, recover lost or stolen assets, and help their teams be more efficient by making sure they have the right equipment at the right job site at the right time. We also think it could be a really great business. By eliminating the cellular radio, the GPS, and data connectivity, we're able to offer this in a really efficient package with great margins. From a competitive perspective, this takes advantage of our network of connected devices.
So if someone else wanted to offer the same product experience, they'd not only have to build this device, but they'd need the millions of connected IoT devices with secure, modern, real-time connectivity, which gives us a really unique differentiated offering for our customers. We've been beta testing the Asset Tag with some really early customers, and it's been incredible to see what they've been putting these on, everything from specialized tools to safety equipment to high-value parts and materials. I'd like to show you now what this looks like from our customer's perspective with a quick live demo. So if we can move over to the dashboard. And here, we're going to look at the exact same system that Sanjit showed you, and we're going to see the Asset Tag. But first, we're going to take a look at the network that powers it.
Here we can see where we are in North America, and this is a map of all the connected IoT devices. Each of these dots represents a single IoT gateway on one of our customers' assets over just a one-hour period. So as you can see, these look like they're on virtually every road and everywhere that operations happens. And that's pretty much where they are. Let's zoom in to where we are right here in Chicago and take a closer look. And as you can see, these things are practically everywhere. And that means that customers using the Asset Tag can find their assets in so many different places quickly and easily. So now, to test this out, we've been testing our own or we've been monitoring our own assets and equipment. And it turns out that putting on this event requires a lot of supplies.
So we have a number of toolboxes that our events and operations team use to make sure that all of this can come together. We sent them out from our headquarters in San Francisco here to Chicago and tracked them with Asset Tag. And we can see one of these toolboxes right here. So we can see that it's right here in the convention center in the hotel. And this is amazing for our team. Now they're not worrying, did it arrive on site? Is it where it needs to be? They can see through the map exactly where it is. Now, we can also zoom out and see the path that it took to get here. So let's take a look at the week that we sent it out. And we can see the path that it took from our headquarters in California.
We can zoom in here, and we see that each of these dots represents a time that the Asset Tag was picked up by another vehicle. Let's drill into its path through Salt Lake City, for example. We can see that it was detected here at 8:21 A.M., again at 8:53 A.M., again at 8:58 A.M. This is pretty incredible. Every few minutes, this tag, which is inside a toolbox, inside a shipping container, inside a truck moving 60 miles an hour down the road, is getting picked up by another Samsara device moving the other direction on the highway at 60 miles an hour. This visibility for customers is game-changing. It helps their teams keep track of their assets, make sure they have what they need when they need it.
Because this is all part of the Samsara platform, our customers have access to all of their alerts, all of their reporting, all of their API integrations, so they can seamlessly bring this data into their operations. That's the Asset Tag. When we look at the Asset Tag and the data that's coming from it, we can see that this data ultimately exists to help enable people. All of the Samsara data ultimately helps our customers operate more effectively in these frontline operations. We've seen examples like DHL, where putting technology in their workers' hands helps them cut their labor turnover by half while also helping them cut their accident costs in half. So that's why we focus on making easy-to-use, elegant, simple tools for frontline operations.
Last year, we previewed a new product called Connected Forms to allow our customers to digitize all of the tedious paperwork that their frontline workers need to perform. We started shipping Connected Forms in Q1, and the adoption has been pretty incredible. We have seen customers digitize everything from safety audits to equipment inspections to incident reports. Now customers have these processes and these forms all digitized in Samsara in the same system where they also have all of their other operational data and context. This enables some really powerful workflow automations. We thought, wouldn't it be powerful if form fields could be automatically filled in based on asset data from our cloud?
We thought it would be great for customers if when a worker entered a hazardous environment, they could automatically be prompted to go through a risk assessment workflow to make sure that they're safe to operate on site. We thought it would be incredible if when our AI detects an accident out on the road, we could automatically trigger an incident investigation workflow to find the root cause and make that less likely to happen again. So we have built a new product to enable all of these workflow automations and more called Connected Workflows. Connected Workflows combines the simplicity of Connected Forms with the power of the data and automation in our platform. And it allows customers to automate and digitize all different kinds of workflows in their operations, from maintenance to safety to workforce management.
This helps customers ultimately have more productive workforces, higher worker engagement, and lower risk to their frontline operations. Let's take a look at a few examples of how this can be used out in the field. Say, for example, we have a customer that operates in a hazardous environment. They've got a job site with heavy equipment moving all over the place. They've got hazardous materials on site. They might have exposed high-voltage lines. And that's all mapped in Samsara with geofences. Now, one of their workers is arriving on site. Our telematics device can detect this in real time and can actually prompt the worker to complete a risk assessment workflow. The risk assessment workflow will guide the worker through a series of questions to make sure they have the protective equipment and the training to operate safely on this site.
If they do, they can go ahead and proceed with their work. But if there's something that they're missing, the workflow will automatically notify their supervisor who can help and make sure that that worker is safe to operate. The worker starts their work on site, and the first thing they need to do is inspect the equipment that they're about to use to make sure that it's in proper working order. Well, Connected Workflows can make sure that they have the right inspection for the equipment they're going to use and can even pre-populate data based on the sensors and diagnostics data that's in the Samsara platform so they don't have to type that data in, so they don't make mistakes. And most importantly, they're guided through the right inspection steps for the right assets. Now, they continue their process.
They continue going through their site inspection, and they find that their equipment's working properly, but there is a damaged ladder. They can be prompted to take a photo of this. This can automatically trigger a workflow for their dispatcher to send a technician who happens to be nearby to bring a replacement on site. These are just a few examples of the types of workflows that can be digitized and automated with software. Our customers have workflows that span safety to maintenance to workforce management to routing to field services. Connected Workflows gives them a tool to digitize all of these and more. That's Connected Workflows. Now, when we spend time out with our customers and their frontline teams, we find two things. One is a lot of these processes that we can digitize.
The other is a lot of tasks and skills that these teams need to be trained on how to do. Our customers are often doing specialized work. It's often in hazardous environments, and they operate with extremely high labor turnover, sometimes 90%-100% annualized turnover. So that means that they're constantly onboarding and training new workers into their operations. We realize that all of the data in our platform and the AI insights on top of that data could enable a much better way to train frontline workers out in the field. And so we built a new product called Connected Training that does just that. Connected Training starts with the data in Samsara and uses it to understand what an individual worker's risk factors and skills are that they need to be trained on.
Connected Training can then automatically deliver targeted training out to a worker in the field based on the skills that they need to be trained on. This training is optimized for frontline workers who are on the go, so they can be delivered training content seamlessly through their mobile apps integrated into their daily workflows. Let's take a look at an example. Let's say we've got a driver, and they have developed a tailgating problem. Our AI detects that they're falling too close, putting them at a risk of an accident. We see that this is happening again and again. The in-cab alerts are helping them in the moment, but then they're falling back into bad habits. We want to deliver them training. Well, the AI can detect this opportunity and automatically push out a training module specifically for that driver.
It's going to target the training for that driver with defensive driving for heavy vehicles, targeted for the specific areas this driver needs to skill up. This training is delivered in their mobile app. And so at the start of their next shift, they can be prompted to complete the training before they start their day. They can view short-form video and even complete quizzes to make sure that they've retained the information. We have customers who have been beta testing Connected Training, and they've seen that this training can actually help reduce repeat behavior by up to 35%, which can translate to massive reductions in accidents and insurance costs. So now driver safety training is something that applies to virtually all of our customers. So Connected Training includes professionally produced commercial driving training content right out of the box.
But our customers' operations are very broad, and they often need to train on more niche or specialized skills. So Connected Training also includes the ability for customers to create their own content specific to their operations. Say, for example, there's a construction company, and they operate concrete mixers. They might have a driver who's very skilled in operating a heavy commercial vehicle, but has never used a concrete mixer before. Now, they can be prompted to complete their concrete mixer training before the start of their very first shift. They can be taught how to use the control panel, how to properly deliver concrete on site, how to properly wash down the drum at the end of the shift. And all of this makes sure that their equipment stays in proper working order, they deliver high-quality service to their customers, and they do this all safely.
So that is three of the newest products on our platform: Asset Tags, Connected Workflows, and Connected Training. All three of these are made possible by our flywheel. They start with the data coming into our platform. They're benefited by AI applied to this data to find insights, and they're delivered alongside all of our other products that our customers use every day through a single pane of glass. And finally, they're connected to our ecosystem of partners who have integrated with Samsara. It's been awesome hearing the early feedback from customers in the last few hours since we've announced these new products, and I'm glad to have had a chance to share them with you all today. Thank you. We're now going to take a 10-minute break and come back with Lara.
Okay. Can you guys hear me? Great. I got the thumbs up. All right. Well, I'm Lara Caimi. I'm the President of Worldwide Field Operations. So I look after the whole customer journey from presales all the way to post-sales. And when I saw you all here a year ago, I think I was two weeks into the job. And I shared a bunch of why I decided to come to Samsara at the time: the market opportunity, the platform vision, our customer impact, and our incredible purpose. And what I can share a year later is that all of it is true and so much more. I really do believe that we are at the beginning stages of digital transformation for our customers. And Samsara is the platform that they're choosing to go on that journey with, which is really, really exciting.
I am super optimistic about the road ahead and wanted to take a few minutes to share a bit about our go-to-market strategy. We should start with the most important thing: our customers. As you know, we sell to the world of physical operations, which is 40% of the GDP. Our customers span the largest and most complex physical operations companies that are out there. It's not just one vertical. It's many, many verticals. Manufacturing, utilities, construction, airlines - truly the companies that are powering our planet. The logos that you see here are evidence of our momentum and our relevance. We resonate across all sorts of companies and verticals because we have a horizontal solution that truly does solve our customers' most pressing needs. We can manage their fleet safely and efficiently.
We can optimize high-value assets, and we can help them deploy hundreds, if not thousands, of frontline workers. The list goes on and on. So I want to start to share a few things about our go-to-market strategy. First, let's start with the left. We divide our customers by size with our largest customers in enterprise and a powerful foundation of customers in mid-market, which are supported by go-to-market motions. As you know, we are on a committed strategy to continue to grow up-market. So our larger customers tend to have more complex operations and therefore more opportunity for us to expand and drive real value for them. And so we're seeing a lot of traction in that. Dominic, in his section next, is going to share a number of statistics on how we're measuring that, and we feel really great about it.
Geographically, on the right, we continue to grow our non-US portion of our revenue. So we actually see white space everywhere in our core, but especially internationally. So we're in really early days of penetration there. Our growth rates are really high, and we have a ton of white space. And as you know, Q1 was our largest international quarter to date. We had most of our net new ACV we've ever had coming from our non-US countries. And I feel like we're really just getting started there. We've been making a lot of investments. We're seeing real success. So you land a big customer that's relevant in the region, the flywheel kind of starts to go in terms of being known. And we're continuing to strengthen our value proposition in those markets. So I truly believe there's a huge amount of opportunity for us.
So the other part of our go-to-market strategy that I want to spend a few minutes on is our land and expand strategy. So we land our new customers with an optimized top-of-funnel, a sales team that's focused on leveraging our products and our platform to solve real business value. And we often show quick time to value with a real ROI through a trial. And I think this is a really, really differentiating part of how we go to market. So as many of you know, I came from ServiceNow, and you could not show you could not trial ServiceNow. It took a complicated 6-month implementation before you could see any value. Well, what's really cool about Samsara is it was designed to be plug-and-play. So in just a 30-day trial, you can baseline customers' metrics and then show real value.
And this is often hard ROI, which is what's really great. So these tend to be in areas that are evergreen, right? Things that our customers are always interested in: safety, efficiency, sustainability. These are C-suite topics. They're topics that they talk about with their board. And we can show hard and real ROI. So for me, this is super, super exciting and a huge differentiator. We tend to land with our connected fleet applications, so telematics and safety, where we, like I said, get to ROI quickly. And then more and more, we're landing with both of these, and then we can leverage our platform vision to talk about where we can go together over time. So then from there, we expand. And we do this by truly partnering with our customers around business value. So we'll start with one program. Let's call it safety.
What they want to do is decrease accidents. So we'll focus on behaviors that tend to lead to that. Maybe we'll start with a quarter or a month of speeding events, and then we'll look at distracted driving, and we'll go after it. And so over time, we're partnering with them to change their culture and change their behavior. And then, of course, we'll discover things about their operations. We'll see assets that are out there that are underutilized. Or maybe we'll see a bunch of paper processes that are there that need digitizing. So you can see that this partnership allows us to expand the platform pretty materially. And so that's why we've really focused on our post-sale experience. And I think this is a true differentiator of our partnership.
You'll actually hear that, I think, as part of the message from our customers who come up in the panel later today. So Nutrien, which is one of the customers you'll hear from, we actually collaborated with them to get the buy-in of 12,000 drivers. And so what we did was we created custom change management videos and other collateral with them. And I think in just 90 days, we were able to decrease distracted driving events by 50%, which is pretty impressive. So let me share an example of a customer that we've been growing with over time. So NexTier is a leading publicly traded provider of oilfield services. They have 10,000 employees, and they offer a full suite of services to help customers service their oilfields.
For those of you who heard the Beyond Keynote this morning, Amy was one of the customers on stage sharing her story. We landed in 2021 with telematics and safety displacing a legacy telematics provider. We worked hand in hand with them, got them to quick ROI, and then they've been growing through acquisitions over time. They bought a bunch of companies. We got them to value. They standardized on Samsara, so we expand that way. We also expanded into assets. We're tracking a number of their oil field assets. That went from 2 to 3 products.
Then as our new product forms was in beta, which Amy talked about this morning, NexTier saw the potential to digitize what they call journey management, which is just a standard oil and gas workflow to drive safety, making sure that the trip is safe for the drivers. They immediately saw value and decided to roll it out. I met with Amy and her team yesterday, and what's really great about them is Amy's like, "Sign me up. I want to sign up for every beta." So I think we are well on our way to making NexTier a 4, a 5, a 6, and more product customer. So it's just a great example of how we land and expand. Okay.
So one of the themes that I mentioned about why our companies buy and then expand is our commitment to get them to business outcomes in the terms that they define, right? So we sponsored a study with IDC to really understand the business value that our customers see with Samsara. And it basically came back with what we regularly hear from our customers. So customers that participated saw an 8x return on investment. And for this particular cohort of customers, which was mid-market all the way through our enterprise, so the average savings was $2 million a year from Samsara. And this was driven by an improvement in key behaviors around safety, efficiency, and sustainability. So decrease in idling time, decrease in accidents, and an increase in vehicle lifespan. But those behaviors translate into real dollar savings.
This slide kind of enumerates the different buckets of savings that were out there. Obviously, safer driving, reduced crash costs by 14%, fewer accidents, reduced insurance costs by 24%, less idling, fewer miles driven, that reduced fuel costs by 5%. I'm not going to go through all of it. You guys can see it, but the list goes on and on. Again, these are those tangible, hard dollar savings that are really C-suite topics for our customers, which is great, right? We love being able to really impact metrics that matter in a real meaningful way in a short period of time. It doesn't stop there. This slide has a lot of stuff going on. I promise I will not read everything on it. We see lots of ways that we're driving ROI from our customers.
The center circle in blue are the areas that I just talked about. That's what IDC validated. That's what really we see across the board for all of our customers. We're calling that our universal ROI. But because customers are integrating with OMO, I think we have 300 integrations on the platform. Every time I say this statistic, every month I have to look it up again because it continues to grow and grow. They're driving additional value. I thought the best way to do this would actually be to tell you a few stories about customers because I think it actually helps you understand their operations a little bit, and it brings it to life. I'm not going to go through every number on this slide, but I'll go through a few.
So let's start with the $30 million one, excuse me, on the bottom left. Certarus is a distributor of compressed natural gas and hydrogen via fleet instead of pipeline. And they've been able to use Samsara to automate 95% of their invoices. They send a truck to deliver natural gas to a customer, and then Samsara has an integration with their transportation management system. And that gives them all the data that they need to automatically create an invoice. They have start-stop time. They have chain-of-custody data. They have customer data that's tied to the location of the vehicle. This automation has allowed them to bring in payments 23% faster and free thousands of hours of time for folks that were processing that in the back office. That was $30 million of savings, which was pretty incredible for this company.
Another example, so the top left, the $8 million. So this is one of the largest LTL, less-than-truckload carriers in North America. And a key inefficiency in that business is something that they call detention. So it's basically when a customer's holding them up. A customer's not ready to either put the delivery on or accept the delivery. And now they have data to prove that they're being detained by the customer, so they can share this transparently with the customer and now bill for that time. So that's led to $8 million in revenue, which for that business drops straight to the bottom line, which is a pretty big deal for that industry. So switch over to the top right. Liberty Energy. You saw the video of Liberty Energy earlier. They're a provider of oil field services, and I think they operate across 60 different wells.
All of those are different tax jurisdictions. So now they can really accurately define where their assets are being used, file their taxes with more accuracy, which saved them $10 million a year in taxes. Interesting. And then the last one I'll talk about, and I like this one actually because it's very visual for me. So the bottom right, Aegion. So it's a multinational provider of pipeline services. So one of their main lines of business is actually to manufacture and install these liners that go inside existing pipelines. And so what's interesting about that is the liners have to be kept at a certain temperature. And if they go too high, they basically turn into solid rock inside the truck. And so they literally need to be chain-sawed out, which costs a ton of money and a bunch of time, etc. It's not good when that happens.
So they use temperature monitoring with Samsara to make sure that never happens, which has saved them $6 million a year. So I love all these stories because for me, it really brings to life not just the universal value that you see with Samsara, but also the more bespoke value that is specific to all these different kinds of businesses by being able to just operate with more transparency and more data. Okay. So as I close, I want you to take away a few things today. First, physical operations really is in the early innings of digital transformation. So I talked about all the white space that we see in the market, but I see opportunity everywhere, right? We have new geographies. We have new products. We have new integrations. We see new use cases.
So as I think about the ability to create an enduring growth story, which is why I came, and I think what we're all here to create, there's a ton of opportunity ahead of us. The second, our customers experience real ROI in mission-critical areas that are board-level priorities for them: sustainability, safety, efficiency. It's pretty incredible to be able to sell a product that effectively pays for itself in a really short period of time. And I love that our products and our selling motion allow us to do that. And then finally, we are really just at the beginning of a journey with our customers. What we find is that when customers start their Samsara journey, it can lead to a culture change. And I'm looking at Fred because we're going to talk about that. They start to rethink how they're going to approach things.
So at DHL, you may have heard the story of they wanted to reduce accidents and reduce accident costs. But the surprising, amazing statistic that happened was they reduced driver turnover because they introduced a really powerful culture of safety, which led to this incredible benefit. And then at Nutrien, we'll talk a little bit about it, but we started with one project, and now Nutrien's on this amazing digitization journey, and we're co-innovating together, which is really, really exciting. So our customers, they want to be innovative.
They want to be more efficient, and they really want to partner with Samsara to help them do that. So speaking of our customers, I am thrilled to be joined by Kyle and Fred from Nutrien and DHL. So we're going to talk a bit more about Samsara, the platform, and the journey together. So come join me on the stage. All right. Okay. So let's kick off with introductions. Fred, maybe introduce yourself. Tell us a bit about DHL and maybe your safety culture since I mentioned it.
Yeah. Thank you, Lara, for having me. I'm a senior project manager for DHL Express, which is one of multiple divisions at DHL. And with that, I oversee technology initiatives and compliance initiatives across our globe, regions, and U.S., so that way we can continuously grow our platform of safety-integrated platforms and systems. And with that, DHL is the top logistics company of the world. With that, we have about 600,000 employees. In that, we service 220 countries and territories, about 92,000 on-road vehicles, and a gabillion trailers across the world, right?
So with that, I would say the military has a pretty good logistics program, but we can get it there much faster, much quicker. We literally deliver things from one side of the world to the other overnight. And with that, we have our customers, which are Ford, Boeing, Amazon, the UN, embassies, militaries from around the world, and we can get it there faster than any other company in the world.
That's great. So your turn, Kyle.
All right. Thank you. So Kyle Springs, I'm the senior director of safety and health for Nutrien Ag Solutions. At Nutrien, retail is really a quarter of our pie. We're also into nitrogen production, phosphate mining, and potash mining in Canada. So our retail operation is the largest in the world from an agricultural inputs provider direct to the farmer. So think of everything a farmer would need to grow a crop from the seed going into the ground until the time it's harvested. About 1,500 locations, retail locations servicing the customer.
About 14,000 employees in our direct retail business, 24,000 overall. We operate around 14,000 vehicles with another approximately 75,000 assets. Really, safety is a core value for us. It is one of our four core values for the company. Our goal is to send people home safe every day, but also with the goal of feeding the world and feeding the future. With population growth approaching 9.8 billion by 2050, our goal is to grow and to have agriculture modernized and advanced enough to keep up with the growing population, knowing that our farmland is shrinking really every day, especially in North America. Thank you for being here.
Yeah. Well, thank you for being here with us today and for being great customers. So I want to start by painting a little bit of a picture for the group who might not work in your industries every day about what some of the operational challenges are that you faced before implementing Samsara. So Kyle, maybe we'll start with you on this one.
Yeah. So logistics challenges, obviously, which we're going to discuss a little later, but we really came to Samsara from a safety perspective. Humans make 2,000 decisions an hour. With our workforce, we're making about 460,000 decisions every second. And we don't always get those decisions right. So humans make mistakes. We can't make the right decision 100% of the time. And with our previous telematics provider, we could see things like speeding, seatbelt, harsh braking, cornering, but we weren't really seeing an improvement in our accident rate. And we wanted to get behind what's going on in the cab.
How are we going to improve driver behavior if we don't really know what that behavior is that's causing the ultimate outcome? So we came to Samsara with a problem. The problem was we needed visibility into our cab, a way to communicate and coach with our drivers. The solution and the partnership that we formed has given us the ability to have that visibility in the cab and to move forward with giving our drivers an opportunity to fail safely. So they may become drowsy. They may accidentally pick up their cell phone. But with that Samsara device, we're given the opportunity to correct that behavior real-time.
Yeah. That's great. Well, we're happy to partner with you on that. So Fred, what about you? What were some of the operational challenges you were facing that brought you to Samsara?
Yeah. So the reality of this is that we had way too many gadgets and gizmos vendors and systems in our trucks, right? And the bigger the company, the more systems it has, right? So we wanted one platform. And with that, we were able to just right off the bat reduce by 7 different systems. That's the cab cams, DVIRs, ELDs, and so on. And then the hundreds of spreadsheets, massaging and messaging the data from all kinds of other systems, right? So that was really a huge driver for us is just a simplified system, right? And then with that, it's the integration piece, right? I mentioned all those systems. And one of the probably big key integrations for us was what we did with the Kronos integration. So ELDs, right? DOT, hours of service, regulatory requirements.
The DOT says, "Hey, you got to keep track of your driver's time from clock in, clock out, and everything in between." So Samsara was able to pull off a miracle for us by putting in that integration, and it lowered so many errors for us, bringing down the regulatory risk here in the United States. And I have really wonderful drivers that take to this technology like a duck in water, but I also have some that don't do so well. So I had a lot of errors when they were fixing their times. And kudos to Samsara and the NASA scientists, engineers that they have that were able to fix this problem for us.
Yeah. Yeah. No, that's great. I love the stories of, like I said earlier, all the stories of how different integrations solve a ton of problems. I know you have some OEM integrations with John Deere and others. Maybe talk a little bit about how you're leveraging integrations.
Yeah. Absolutely. So we mentioned the 75,000 other assets in addition to vehicles. We have a lot of applicators on the road. We have a lot of fertilizer spreaders, chemical applicators, a lot of things running on green paint. And the OEM connection with John Deere really gives us that full picture. So from the delivery vehicle taking the product to the field to the equipment that's going to be putting the product on the field, determining efficiencies, making sure we get the customer what they need, when they need it, and how they need it. And OEM integrations on some of our light-duty vehicles.
And really what that's done for us is cut down on installation of hardware and software, which is cost savings, but it's also saved us about $1 million in just OEM integration versus going with physical hardware. So it's given us a better view into our equipment and our assets and the data we need while also providing us an opportunity to save some costs.
Yeah. That's great. Well, I'm going to stick with you because before you guys heard me talk about the journey that we go on together. We started, obviously, with safety and giving you visibility in the cabs, but now we're on a much broader journey. So maybe tell us a little bit about your digital supply chain vision and the partnership with Samsara and what you'd like that to be.
Yeah. So we would like to partner with Samsara for an end-to-end process or operations where we have visibility into all of our fleet, all of our fleet, all of our fleet, all of our assets, whether powered, unpowered. We have operations across all of North America, and they differ very greatly. But at the same time, there's assets we could utilize sometimes that are down in areas where we could bring assets from other areas. The logistics of providing the products to our retail branches as well as to our growers when they need it. Something very unique about agriculture is you can't plan too far ahead. So you may have a grower that tells you they're going to plant 10,000 acres of corn and they don't get rain in time. They're going to be planting 10,000 acres of soybeans. And that's completely different products, completely different things they may need.
If they're in a drought, they may have bug pressure. And we were planning on rain, and we had fungicide in place where we may need now pest control. So the logistics of being able to see an end-to-end operation of where are our trucks, where are our trailers, where's our product, how are we going to move that, get it to the customers when they need it, how they need it, as well as have the equipment there. And so the opportunity to provide that holistic view with Samsara is going to give us that end-to-end integration that we've been lacking.
Yeah. Well, I love it. It's really cool. We sat in EBC a couple of months ago, I think, in San Francisco with your team. And one of my favorite stories was our team's solving this with them. And I can't remember which guy it was, but he was allergic to peanuts and deathly afraid of snakes. And so they're in Georgia or something at a peanut field. And your guy's in the truck and our guy's climbing on top of the tanker or something.
Yes yes.
It was an awesome visual. Again, I love a visual. It was a good one.
Yeah. That was Ryan Sherlong from Samsara climbing in through everything and one of our people hiding in the trucks. Yes.
Yeah. But what's really cool, I think, is some of the tank monitoring stuff we're innovating on, which I like the idea of being able to decide when you need to not have to waste time refilling tanks or checking on tanks, etc.
Yeah. So from a logistics perspective, being able—we're in remote areas. People say, "You get to travel for work. It must be fun." I fly into an airport and I go as far away from civilization as possible. That's where we operate and that's where our tanks are. The ability to know how much product is needed in a tank prior to dispatching a driver, maybe a tank's half full and you have another one half full nearby, you can send one load to both as opposed to sending a full load to a tank that only needs a half and then having to come back to refill to go to the next. So it's really going to help us be more efficient in our operations.
Yeah. That's great. Well, I'm super excited and we're privileged, obviously, to be on this journey with you. One thing I talked about before you guys heard me was ROI. And I think both of you have experienced a lot of fantastic ROI. So maybe let's pivot over to you, Fred, and talk about some of the return on your investment that you've seen with Samsara.
Yeah. So for us, it was a little bit eye-opener, right, when we looked at just the cameras themselves for a minute because we had kind of bits and parts of everything else. And what we found really crazy was the three-sided ROI sword. And what I mean by that is, number one, you put the cameras in the truck and you cut down on all those shenanigans that people aren't doing if there was a camera there or wasn't, right? So that cut down a whole lot of stuff.
Then you got the second part, which is the harsh events and the coaching opportunities, which we shared org-wide as a learning, right, all the way through to change our culture. Then number 2 is that we're able to address the individual on their behaviors. Then number 3, when we were wrong, I cut so much of that claims cost on the back end. He said, she said, reconstruction, lawyers, and everything else. I'm able to close my claims very, very quickly and settle very, very quickly. So we saw a massive savings right there, which led to those statistics you heard earlier this morning or even up here, which is 69% or so of harsh events and the triggered events. Then you got, what, about 29% in accidents altogether reduction and then about a 45%-49% cost savings in just our back-end claims management of that. So remarkable.
Yeah. Well, and then there's the retention issue, which is amazing. It's an incredible outcome.
Yeah. So we've been able to exonerate, if not dozens, hundreds of drivers. Then we have our retention piece, right, which is extremely high. And that's in part to Samsara because now we got their back, right? We're protecting them. And drivers want that long-term longevity inside of their career. They sign up for something. They're looking for a long haul, right? Then we got the backside of ROIs on EVs. And the EV is huge for us. We've got something like 22,000 in our network, 22,000 vehicles around the globe that are now electrified. So think about this for a second. The EVs, number one, you're going to have to reroute everything. Can't just pull over, gas station. So that ROI on all the rerouting and everything that we have to do.
Then number two, the EV suitability report. If I have a vehicle that's got a lot of waste, a lot of mileage, a lot of CO emissions, this EV suitability report pinpoints my biggest losers like that. I can switch them out, send them to auction, whatever it is that I need to do with them. So then everything there is one massive ROI for us.
Yeah. That's awesome. The other story that you shared when we were prepping for this is how Samsara has actually helped you win business. So maybe talk about the revenue implications that you see.
Yeah. So that's on our supply chain side. One of the big wins we had was in their life sciences division. One of the customers with drugs and medicines, very high-valued. One of their particular requirements, very strict requirements, is that when they're delivered, they can't open the trucks until they get to that end point. So wrapping those geofences around where it's at and an alarm if the doors open up before it's supposed to happen, that immediately shot off to our security firm that was handling those, and they can respond immediately. And then that there, they were very, very happy with that as well. So DHL, we deal with everybody. Winning business because of the different technologies and functions that we can have is just an A-plus add to everything we're already d oing.
Yeah. That's great. Well, so Kyle, why don't you share your ROI stories?
Okay. Thanks. We're really getting into that journey, and we've already seen lots of benefits like the $1 million I talked about, which is an annual savings on integrations with the OEM.
We were able to reduce our unsafe driving events by 50%. And then if you just look at unsafe driving that was distracted driving and speeding, it was 51% within just a 3-month period. We haven't actually put pen to paper on what that cost savings is on gas just by not speeding the gas cost savings. But we have recovered, I think, close to 10 stolen vehicles now here in just under 15 months of really only about 6 months of full deployment because it took time to deploy to 12,000, 14,000 vehicles spread out over all of North America. But we've also had the ability to not just recover stolen vehicles, but the same as Fred mentioned, exonerate a lot of our drivers.
So we've had at least seven incidents that would have previously probably required a lawyer, probably required court, probably required some type of settlement that we were never even approached by a lawyer from the other team because we had video proving that our driver was in the right and did everything absolutely possible. And I heard you talk earlier about the change management and the change management for drivers and getting them on board of big brothers watching me with a camera in the cab to my job was just saved. I'm not sitting on a stand.
That spread like wildfire. So all it took was one really good story, one really good incident of how we put those in the cabs to protect our drivers. And the story spread. And now drivers are happy that those are in the car, and they're using those to become better drivers and not just protect themselves, but protect our communities. That's exactly same with all my thousands and thousands of Teamsters. Once it goes, it's wildfire, and they really like that.
Yeah. That's great. Well, changing gears a little bit. You all saw the keynote this morning, and I think you just saw the demonstration earlier. We announced our exciting new product, our asset tags today. I think each of you probably has one in your pocket, your gift from your Oprah gift. So tell me about it's brand new, but what use cases can you imagine for that product? Whoever wants to go first.
Yeah. So for us, we deal with a lot of airplanes, ships, right, trains, trucks all over the world. And for us, I think for me, right off the cuff when Kiren and them had mentioned it, is it's all the stuff behind the scenes that the customer doesn't see, the end customer. So we got a shipment, $ millions, computer chips, and it's not on the right plane. It's not on the right truck. For companies to be able to solve that problem before it becomes a big problem, and now you have an upset customer, they're no longer loyal, you no longer have your loyal customer there, right? For us, things like that, then the gap in visibility, a lot of trucking companies work in checkpoints, right? It's going to ship to truck to warehouse back on another plane. So to be able to see that stuff in between, I can see limitless, limitless openings for those kinds of products.
I think one of the bigger things is when we make an error, right? And Houston, we got a problem. It's going one way, and it should be going over there with that truck. And I think that's going to resolve a lot of problems and make the world faster. I can think of some brides back in my past that the dress was on the wrong plane.
Oh, I didn't know where you're going with that.
That was going to.
You got me nervous.
Yeah. That product right there probably would have saved some angst that would solve the world's problems. So good move, the entire Samsara team. Amen to you.
All right. Well, that's awesome. So Kyle, any thoughts?
Yes, absolutely. About a million. So with all of those assets, we have things called fertilizer buggies that farmers borrow. We have bubble trailers farmers borrow. We have seed tenders farmers borrow. They usually come back in a lot worse shape than when we lend them out, but sometimes they don't come back at all. And so the ability to track high-value assets that we're providing to our customers as a customer service. We have about a $150 million investment in seed boxes.
One, from a safety aspect, the bags would fall on people. So we invested in hard-sided seed boxes. They're $900-something a seed box and have about a 30-year lifespan on them, 20-30-year lifespan. And so being able to protect that investment and make sure we get those boxes returned to us. I could probably go on and on and on and on about retail uses, but one that popped in my head just as I was walking in is our potash rail cars. They get lost. A 25-ton rail car with 100 tons of product in it is quite frequently lost, and it takes weeks to determine what rail yard is sitting in and when it's going to get to us. I'm going to go straight to my potash business partners and say, "Look, here's a tag. You can stick in our rail cars, and you can quickly identify where it is on any rail yard, anywhere in North America, and send it to the rail company and not lose weeks of waiting on your product." So it's a game changer, really.
That's awesome. Well, so now I want to talk about the other new product. You heard us talk about our Connected Workflows, which I know Kyle, you guys were, I think, a Forms pilot customer at Nutrien. So talk a little bit about what value you see coming out of Forms.
Sure. So we have a lot of IoT devices to do different things, such as audit, such as vehicle inspections, probably over 30 out there. What Samsara is building in their IoT is going to allow us to eliminate nearly all of them. Forms being one of the things that we actually didn't have a provider for. It was still a very manual process. 1,200 locations in North America that I'm directly responsible for. The rest are in Australia and South America. For us to do a pre-job hazard assessment, it was a written card. So you had to go to the office, get a card, write it, take it back, have it signed, find somebody to sign it, and then you had to file it.
Well, when my CEO comes to me after an incident and says, "We're doing our pre-job hazard assessments," or, "We're doing our process safety reviews prior to restarting the process," or, "We're doing our life-critical permit cards prior to doing those jobs," right? And I have to look at him and say, "We've trained them to, and we hope they are, but I'll have to go to all 900 locations and look in their filing cabinet to answer that correctly." I'm going to be able to open an app and say, "Yes, we are. No, we aren't."
And if we aren't, then I'm going to be able to work with those locations and coach them and train them to make sure that we're doing it. And so we were looking for a digital service provider to be able to do that for probably two years. Samsara listened to us, and they developed and built the product and partnered with us and given us the opportunity to really share our needs and have a product developed that's going to work for everyone but meets our needs right out of the gate. So really excited about that and looking forward to it.
Well, we're proud to partner with you on that. Well, so Fred, what about you with our workflows? What do you see for DHL?
Yeah. So very similar, but we've just barely touched the iceberg with the workflows. We have them integrated into our DVIRs, going to outside mechanic vendors and stuff like that, triggers for unsafe, for defaults, and all that stuff. So we got that kind of building all out. Then ours is service is super important to us. We're a trucking company.
So we got a lot of different workflows going with that. And then our U.S. side of supply chain is actually piloting a whole lot of hosted Connected Forms with yard checks and gate checks and performance dialogues, which are employee discussions and stuff like that. So they've got a whole host that they're also piloting right now. And I think we've just barely touched the iceberg, and I think we've got like 40-50 different workflows that are working like every hour now, so.
That's awesome. All right. So one of the things that I shared with these guys, I claimed that our customer service and our post-sales support is a differentiator, but I'd actually rather ask you that question. What do you think?
So we went shopping for a telematics provider to replace our existing one and thinking that someone was going to sell us a device and give us a platform to use it in. And very quickly, it was beyond evident that Samsara didn't want to sell us a device. They wanted to sell us a solution. They wanted to engage us in a partnership. From the pilot to the change management to the implementation to now the customer support from being fully on the system has just been nothing short of miraculous. And it's truly been a partnership and just couldn't be more happy with the decision to go with Samsara.
That's great. You all make me feel proud. What about you?
So for us, right, there's that honeymoon period, right, with any vendor. We're really going back to weddings. Come through the door, and they're going to sell me the world, and I spend a lot of energy batting these vendors away and actually literally getting rid of them. And with that, I was extremely surprised when Samsara that never ended. We kept going. We kept going. I still now have 2 weeks. Every 2 weeks, I'm on with my project teams with Samsara, and we're working through those things for long-term, for actionable items to improvement, gaps, what's on the future, what's on the horizon.
And honestly, from me personally, Samsara is an amazing company, and their people are incredible. They fit in really, really well with DHL. They match our culture and our values. And it's pretty simple. Motivated employees to great customer service to quality customers that you keep forever, right, loyal customers, and then a profitable network. You guys broke down that recipe very, very well. I'm very, very pleased. And honestly, that's their niche.
Oh, well, thank you so much. Well, it really is an honor for us to be able to partner with you and serve you. And I look forward to many years of great stories and great value that we can deliver together. So thank you so much for taking the time to be here with all of our investors and teach us all about your industries and our partnership. So appreciate it very much.
Absolutely.
Thank you.
Thank you. All right. Well, thanks for attending our investor day. I'm Dominic Phillips, our CFO. I've got about 20 minutes of content. I'll bring everyone back up. We'll do a little bit of Q&A before wrapping things up. And with that, the theme of my section this year is sustaining durable and efficient growth. Before I get into the reasons why we expect our growth to remain durable and efficient going forward, I first just want to level set with where things stand today. We had our Q1 earnings call a few weeks ago, and on it, we said we're doing roughly $1.2 billion of ARR, growing 37% year-over-year. We've been able to sustain this level of kind of scale and growth in just a little more than eight years of selling for a couple of reasons.
The first is we're operating in really large markets that are very underserved by technology. We've got a lot of greenfield opportunities still in front of us. And then secondly, we're exceeding well across a number of really important growth factors. We're selling into many different customer segments. Both mid-market and enterprise are growing quickly.
We're selling into many different geographies, both North America and Western European countries. We're selling into many different industries and end markets, and we have multiple products at scale all on one unified platform. And because of that execution, we've been able to demonstrate this scale while still growing very quickly in a short period of time. And in addition to large scale and fast growth, we've also dramatically improved our profitability over time. We've driven a lot of leverage across all of our major business functions. We've greatly improved our working capital efficiency over the past few years. On the right-hand side, you can see we've been free cash flow positive now for the last 4 quarters. And down at the bottom, you can see we've greatly improved our free cash flow margin over the last few years.
We're set up really well to be able to continue to deliver more leverage as we scale the business. And as a result of that consistent execution, we are really now operating with a unique combination of scale growth and profitability. There's roughly 300 U.S.-listed software companies that have some part of software in their business model, of which approximately 70 have achieved our level of scale or greater that are doing more than $1 billion of ARR. And of those 70 at-scale U.S.-listed software companies, there's only two, including Samsara, that are expected to grow more than 30% this year and to be free cash flow positive. And so we really are now operating in rarefied air in terms of scale growth and profitability. All right. That's a look at where things stand today.
Now I want to talk about why we think that growth rate can continue to be durable going forward, and it's for these few reasons. The first is we're operating in really large markets that are growing very quickly with a fragmented competitive landscape and a lot of greenfield opportunity in front of us. Second, we have a horizontal platform of applications that's applicable to many different end markets. Third, while we sell into many different customer segments, Samsara is really purpose-built for large enterprise customers with complex operations, and we're seeing continued momentum within that segment. Fourth, the majority of our customers subscribe to multiple products, which opens up a lot of opportunity for future expansions and cross-sells. And then lastly, we have a very nice balance of both land and expand, which creates a nice floor for long-term durable growth.
So double-clicking into each of those in a little more detail. Again, our products today address a really large market opportunity that's growing very quickly. Connected operations, which encompasses all of the products that we sell today, is estimated to be $122 billion, growing at an 18% CAGR to $200 billion over the next few years. Within that, connected fleet, which encompasses our two largest products, video-based safety and vehicle telematics, is estimated to be more than $50 billion, growing at a 25% CAGR to almost $100 billion over the next few years. So not only are these really large markets, but they're also growing really quickly. And the key driver of that fast growth is digital transformation.
Similar to what we've experienced in our consumer lives over the last couple of decades and more recently within the workplace with business applications, the world of physical operations is really in the early innings of digital transformation, and that's creating a tremendous amount of growth within these markets. Not only are these markets very large and growing very quickly, but historically, they've been very underserved by technology. On the left, if you just look at the North America vehicle telematics market, this is a market that's operated in some shape or form over the last few decades. You can see on this pie chart that just under 50% of all commercial vehicles in North America are using a vehicle telematics solution today. That market is made up of more than 35 different vendors, of which we have a market-leading position.
More than 50% of the commercial vehicles are not yet using technology and have not adopted that. That is greenfield opportunity. We see that shrinking rapidly over time as more and more of these customers are adopting technology. On the right-hand side, it's a little bit of a different market. This is a look at the North America video-based safety market. Now, this is a market that's much newer. Some of these recent technology tailwinds like AI and 4G, 5G connectivity, and HD video have only allowed this market to operate in its current form for the last 5-10 years. As a result, roughly 10% of commercial vehicles are using a video-based safety solution, again, of which we have a market-leading position, but the other 90% of commercial vehicles have yet to adopt this technology.
So while these markets have different dynamics, in both of them, we still have a lot of opportunity to capture more market share over time, both by replacing legacy incumbents, particularly within vehicle telematics, and also introducing technology into new markets where they didn't exist before. All right. The second driver of durable growth is that we have a horizontal platform of applications that's applicable to many different end markets. As Sanjit mentioned earlier, physical operations makes it more than 40% of global GDP, and it spans across more than a dozen industries. On the left, you can see a pie chart of our ARR as of Q1, broken out by the industries that we sell into. And you can see there's not one end market that really dominates our ARR. Our leading markets: transportation, construction, wholesale retail, field services, logistics.
All of these make up roughly 10%-20% of our ARR. On the right-hand side, you can see a visual representation of some of the industries and customers that we're selling into, and you can see that we're also often selling into the market leaders within these markets. Third driver of durable growth is that we're seeing more and more momentum within our large customer segment. Now, this is a space where we have made a number of multi-year investments. On the go-to-market side, on the pre-sale side, we've added a lot of investments into our sales engineer team, our business value strategy team to help align with customers on ROI before they sign the contract. We've added a lot of field sales reps. We've invested a lot into our field marketing team. This customer conference is a great example of that.
We've added into our partner ecosystem, into our customer success organization. Similarly, on the R&D side, we're releasing more than 150 software feature releases a year. We have three product categories at scale to meet a number of different use cases for these large customers. We've invested in security and scalability. We have almost 300 integrations to other technology vendors, which is important as these large customers increasingly use Samsara as their system of record. The result of all of those multi-year investments has driven a lot of momentum within this segment. Now, there's really two key inputs to driving large customer momentum. The first is we can increase the average ARR per large customer, and the second is that we can increase the overall number of large customers, 100,000-plus ARR customers.
Both of those contribute to the ARR mix coming from large customers as a percentage of our total ARR. That's down in the bottom boxed in green. That is the truest indicator of success for us with large customers. We want to see that ARR mix increasing over time. That implies that this customer segment is growing faster than our overall ARR. In Q1, we had almost 2,000 customers paying more than $100K of ARR. On average, those customers paid $316K per customer, and you can see how that has increased over the last couple of quarters. Those two combined to result in 53% of our overall ARR coming from large customers. That's up from 49% one year ago and 45% two years ago. Not only is this our fastest-growing customer segment, but these large customers are getting larger over time.
So on the left-hand side, this is a look at our 10th largest customer over time. In FY 2020, our 10th largest customer paid us roughly $900,000 of ARR. Today, our 10th largest customer pays us roughly $3.3 million of ARR, or an increase of 3.5x over the last four years. In the middle, our 25th largest customer today pays approximately $2.1 million of ARR. That's also up 3.5x over the last four years. And on the right-hand side, our 100th largest customer's ARR has increased by 4.6x over the last four years. Or said another way, our 100th largest customer's ARR today of $900,000 is the same as our 10th largest customer four years ago in FY 2020. So this is our fastest-growing segment, and these customers are getting larger over time. Fourth driver of durable growth is that we're seeing an increase in multi-product adoption across our customer base.
On the left-hand side, 83% of our core customers and 94% of our large customers subscribe to two or more products. Those percentages are up over the last 1 year and over the last 2 years. Similarly, on the right-hand side, 34% of core customers, 58% of large customers subscribe to three or more products, which is also up over the last 1 and 2 years. Our multi-product adoption is increasing year-over-year. That is leading to more products at scale. If you look at just our two vehicle-based applications, video-based safety and vehicle telematics, each of those products are now doing more than $450 million of ARR. If you look at our non-vehicle-based applications, primarily equipment monitoring, and you add in all of the other emerging products, collectively, that product set is doing more than $125 million of ARR.
Each of those three product categories are still growing north of 30%. So they're all contributing meaningfully to our overall ARR growth. As a reminder, I said 37% in Q1. What's also interesting is that we're seeing broader multi-product adoption within our largest customers. This is a look at our top 25 customers by ARR. In each row, you can see checkmarks for all of the different products that these customers are subscribing to. A couple of really interesting takeaways. The first is that 25 out of 25 subscribe to vehicle telematics, and 23 out of 25 subscribe to video-based safety. So obviously, a lot of adoption within our core vehicle-based applications. But what's also interesting is that 20 out of 25, or 80% of these customers, subscribe to a non-vehicle-based application. So that can be equipment monitoring or site visibility or some of our workforce applications.
These customers are subscribing to more and more products over time. 19 out of 25, or just under 80%, subscribe to three or more products. Lastly, if you take a look at the second to left-hand column, you can see, again, the diversification of the industries that we're selling into: construction, field services, logistics, wholesale retail, etc. Another example of the diversity within the end markets we sell to. The fifth and last driver of durable growth is that we have this really nice balance of net new ACV mix coming both from new logos as well as expansions to our existing customers. Over the last few years, roughly 50% of our net new ACV has come from new customer acquisition, and the other 50% has come from expansions to our existing customers.
Within that expansions piece, the majority of that is coming from upsells as our customers are doing phased rollouts over time to a broader set of physical operations assets or frontline workers. Double-clicking just on the new customer opportunity. We have a tremendous opportunity to continue to go out and land more and more new logos every quarter. When we look at our CRM, we've identified more than 325,000 physical operations customers that can be on the Samsara platform. Today, we've only licensed roughly 5% of them. So we have a tremendous opportunity to land more and more new logos every quarter, and we're doing a really good job of that. In Q1, the quarter that we just announced, it was our second highest quarter ever in terms of new core customers added onto the platform. You can see the callout.
We added more than 1,000 in the quarter, and a lot of opportunity remains. We also have a really big opportunity to expand the relationships with our existing customers. So if you take our current customers and everything that they're licensing today and call that 1x, if we were to go wall to wall across all of their physical operations assets and all of their mobile devices and all of their frontline workers, and we were to license all of the products that we have: connected vehicle, equipment monitoring, site visibility, mobile experience management, connected forms and workflows, and connected training wall to wall, we estimate that there's more than a 6x uplift on their current ARR.
If you just look at the connected vehicle opportunity, our two core products, video-based safety and vehicle telematics, we estimate that we've only monetized roughly one-third of our existing customers' commercial vehicles, and we have an opportunity still to monetize the remaining two-thirds. A tremendous opportunity to continue to expand with our existing customers. What's really exciting is that the expansions are happening very consistently over time. This is, again, a look at our top 25 customers by ARR. In each row, you can see the light blue box represents when that customer first landed with Samsara, and the dark blue boxes represent each quarter that those customers expanded with us. You'll notice all of the dark blue that takes place subsequent to the light blue, the customers are expanding very consistently over time.
Just within Q1, the quarter that we just reported, 18 of our top 25 customers did an expansion with us. If you look at the last two quarters combined, Q4 and Q1, 24 of our top 25 customers did an expansion with us. We've got a major opportunity to continue to land more new logos over time and to build deeper relationships with our customers, and those expansions are happening very consistently. All right. Those are the reasons that we think the growth rate can continue to be durable, but we also think we can get more efficient over time as we scale for these three reasons. The first is we have a subscription business model, which has really strong customer unit economics that will lead to a lot of long-term profitability.
The second, we've demonstrated a lot of operating leverage as we've scaled, and we think we can deliver even more leverage as we scale going forward. Then lastly, we're really focused on effectively managing our equity dilution. Double-clicking in a little bit more detail. Again, Samsara has primarily a subscription business model. More than 98% of our revenue comes from subscriptions that we recognize ratably over the customer's life, the customer contract lifetime, which is roughly between three to five years. Like most subscription businesses, we spend money upfront to acquire a new customer, and then as the customer pays a subscription fee and renews their service over time, the customers become very profitable.
So much so that our customer lifetime value to customer acquisition cost ratio has been more than 8x, which means that for every $1 that we spend upfront to acquire a new customer, we're driving $8 over that customer's lifetime, which creates a lot of long-term profitability. Secondly, we've demonstrated a lot of leverage over the last couple of years, and we believe that we can demonstrate even more leverage as we scale our business. On the left-hand side, our non-GAAP operating margin improved by 27 percentage points over the last two fiscal years and by 12 percentage points over the last 12 months. And the middle, our adjusted free cash flow margin improved by 45 percentage points over the last two fiscal years and 8 percentage points over the last 12 months. And we've listed out some near-term leverage points going forward.
Now, in terms of our cost of goods sold or gross margin, I don't expect a lot of near-term leverage, but I do expect more medium opportunities for more medium and long-term leverage as we continue to optimize our cost of goods sold and as we scale some of these newer products. So last year, we announced Mobile Experience Management and Connected Forms. We talked about Asset Tags today and Connected Training. All of those products are gross margin accretive. So as we scale those businesses, there is more opportunity for medium and longer-term gross margin leverage. Sales and marketing is an area where I expect near medium and long-term leverage. The cost of sale on a renewed $1 of ACV is dramatically lower than on a new $1 of ACV, whether that's for a new logo or for an expansion to an existing customer.
Now, remember, we've only been selling for a little more than 8 years, and we sell 3- to 5-year contracts. So still, less than 50% of our overall ARR has gone through a renewal cycle. As more and more of that ARR renews over time, that will drive more leverage in sales and marketing. R&D is an area I don't expect a lot of near-term leverage as we continue to invest in innovation to drive durable growth. I do expect opportunities for more medium- and long-term leverage there. G&A is an area where I expect near-, medium-, and longer-term leverage as we benefit from economies of scale, more automation, and lower-cost hiring strategy. Then finally, we're very focused on continuing to effectively manage our equity dilution.
Our largest cost as a company is personnel, and that comes in the form of compensation in these three buckets: salary, some form of variable cash, and equity. As a business, we view all three of these as real costs of the business, and we manage them as such. We budget them, we forecast them, we report on our actuals versus forecasts, and we're actively pulling levers to manage these three buckets. Double-clicking on equity, if you look at our annual equity dilution, it was 4.4% in FY23. We lowered it last year to 2.7%. This year, we're on track to be below 2.7%, and we've got a path to getting it to less than 2% in the future, and we've listed out a few levers that we can pull. I think what we want investors to appreciate is that we're shareholders of the business.
We view all of these as real costs of the business, and we're very focused on managing them to be below 2% in the future. Okay. Those are all the reasons that we expect the growth can continue to be durable and efficient going forward, but I'll leave you with these highlights that I've covered. Again, the first, physical operations: large, fast-growing market, fragmented competitive landscape, a lot of greenfield opportunity. Second, we're selling into a diverse set of end markets and industries. Third, we continue to see more and more momentum out of our large customers, and that will continue to drive durable growth. The fourth, we're seeing more multi-product adoption year-over-year, and we're seeing broader adoption within our largest customers. Fifth, we have this really nice balance of land and expand.
We've got a lot of opportunity to continue to land more new logos and to expand the relationships with our current customers. And then underpinning all of those, we've driven a lot of leverage in the business over the last couple of years, and we expect to be able to do more of that as we scale the business going forward. So with that, I'll welcome everyone back up, and we'll get into a little bit of Q&A. Okay. I'm past the mark.
Great. So we're now starting the question-and-answer portion. If you have a question, please raise your hand. All right. Let's go. Sure. Keith, first.
Excellent. Thank you, guys. Great presentation, and beyond is always very inspiring. It's such a cool view into operational technology. I wanted to ask about the generative AI side of the equation. And a couple of questions in there, but they're tight questions, so it's not going to be too much. So you have an incredible data set. You have all the miles driven. You have all the transactions. Do you have access to that data? Do the customers allow you to use all that data, number one? And number two, how are you thinking about sort of charging for that on a go-forward basis? That was a great demonstration that you showed. I can also imagine that this is the ultimate Copilot application. The drivers can't be using an app in the cab or whatnot. Copilot and a voice-activated, it would seem like a really cool interface. How do you guys think about monetizing that functionality or that access to the data over time?
Yeah. Let me take care of the first part and kind of tag team a little bit. So technologically, we do have access to the data, and that's how customers got involved with us in the first place. Because if you think about what we did for distracted driving, we were looking at the video data to understand risk, and that's very specific to a customer. So we've always been upfront and transparent about this, but it's what the customer is offering us, which is, "Hey, we have all of this data." They don't want it sold. They don't want it exposed. They want it to keep it private and secure, and so we have a responsibility there. And then in terms of our perspective, we are training models in an aggregate sense.
So we are taking in all of this customer data, training a model across it to do things like benchmarking, for example, and we think that's going to be tremendously valuable. And then in the third part of your question, you mentioned Copilots. We absolutely see that opportunity. We want to make sure we get the usability right. Because if you think about the frontline workforce, it's got to just work. It's got to be really seamless, really easy to use, and very practical. So we're taking our time to really refine. We've got a lot of customers giving us feedback here. We're doing that on a continuous basis, but we absolutely see that opportunity, and I think it's going to be just tremendously useful to the customer.
I'd say in terms of monetization, our customers in operations, not IT, they're not asking for an AI SKU or an AI license. They're looking to actually have these problems solved for their operations team. We try to understand what are the areas where AI can benefit our new products as well as our existing products. What's the right way to deliver that to a customer? It could be a chatbot. It could be ingraining it into a workflow experience and then making sure that we monetize the innovation appropriately.
Next question with Agzukin, Wolf.
Hey, guys. Thanks for taking the question. It's rare when you have a slide that has 5% penetration, 20x, and then 6x opportunity to increase ARR, and then you launch a product like the tags product. So help us size up the opportunity for that. Was that part of the 6x? Is it 60x? Dimensionalize for us because I just did a quick query of how much stuff is lost in some of those industries, and it's bigger than your TAM.
There's a lot of stuff out there.
Just help us understand.
It is included, and I think you're right. I think that's what's also exciting is where a commercial vehicle is a very, it's one type of asset. When you get outside of vehicles, there's, I think, even one of the customers said limitless opportunities. And so we used a ratio, a simple ratio of potential assets to vehicles. It was included in the 6x. I think it's relatively conservative, and we're just getting started with that product, and we'll see where it goes, but we're really excited about it.
Great. Next question with Matt Hedberg from RBC.
Thanks a lot, guys. This has been great. It feels like an Apple event when you announce stuff, and people in the crowd love it, so it's great to hear. Dominic, for you, you didn't update kind of the IPO margin framework longer term. It feels like there's a lot of leverage in the business. I guess when we think about the gross margin line and all these new products like asset tags, you said they were accretive to gross margins, but I guess I'm wondering, could you kind of outline where maybe gross margins can go longer term? And it feels like the 20% operating margin target longer term is skewed higher, but just give us maybe a little bit more granular about where can we see gross margins 80%+? Can we see margins 30%+? Just a little bit more of a framework there.
I think we need to see how these products scale over time, but if they do add a lot of scale, they're going to be margin accretive, gross margin, and therefore operating margin. So I think we want to see where those ultimately kind of play out. But we did 77% gross margins in Q1. I do think that they will go higher, medium, and longer term, but how high they go, I think, will be determined on how successful these products are and how much they scale. But there's nothing structurally different about our business that I think implies it can't get to similar margins of other at-scale enterprise software businesses. And so we need to execute it, but we do have upside there.
Great. Our next question comes from Junaid, Truist.
Thank you. You've talked about the resiliency of the operational budget compared to IT budgets, which have been challenged recently. What are some of the markers that you look at that could potentially give you pause in that resiliency of that operational budget going forward?
I think from our perspective, we don't see those sort of indicators on the horizon. We serve a number of different industries, whether it's transportation, logistics, or construction, and so on. There's cyclicality in some of those markets, but these are companies that have been around 50 to 100 years. You heard the folks that were on stage with me, Trayecto, for example, and Carlos has said this is a business that they've been running for eight decades. So I think there's a lot of resiliency in operations because they have to operate nonstop. It's not really a choice for them to just pause or stop or pull way back, especially for these larger enterprises.
I think what's also exciting, and a reason I think we've had some consistency through some of the macro volatility over the last few years, is that there's real hard ROI. And so it's pretty easy for customers to identify if they deploy Samsara, how much money they're going to save. You've heard it from customers. You've seen it in the IDC report. And so in any sort of macro cycle, customers are deploying Samsara because it helps them operate their businesses more efficiently and ultimately save more money. And so even in a situation where there potentially is an economic downturn, customers may be more motivated to find additional ways to save money, and we can help them do that.
Our next question from Kirk Materne with Evercore.
Yeah, yeah. Thanks very much, and thanks for having us. Maybe just a slight follow-up to Keith's question just around AI. How far away is the platform from being able to make suggestions to your customers about next best actions? Meaning, it seems really just from listening to the customer examples, there's so many things that your platform can prompt them to do to create better outcomes, and it seems like that's another level of the flywheel. I'm just kind of curious, where are we on that sort of journey? It seems like you have the data. It seems like you have the natural language interface. I'm just kind of curious how far away we might be from something like that.
I can take a stab at this. Well, we're already doing that in many cases in the product today. If you think about our safety product, for example, it will use AI to look at a driver's behavior and risk factors and suggest behavioral coaching. And we talked about their new product, Connected Training, actually suggests training modules based on those AI insights. So you can imagine other types of suggestions or prompts in other areas from maintenance to routing to asset utilization, all the other areas where we have this data. And that's how we think about continuing to build there, but it's very much a core part of the product today and how customers are using it.
Okay. Our next question comes from David Unger with Wells Fargo.
Front door that can help unlock newer segments of the market? Is the go-to-market for those product categories similar, or are those different buyers?
Yeah, I can start with this one. We're actually looking at that as an expansion product. I mean, I'm sure we will land with it. Maybe there's even a few that I don't even know about yet that we have. But once they understand the power of Samsara and the platform, it's sort of an obvious next step when they understand how to leverage the data and use the driver app, etc. So I view it more as expansion. Now, over time, as it becomes a more powerful workflow engine, and we learn from our customers, which is how our innovation flywheel works, there might be targeted plays that we decide to run as land plays. But as of right now, the most efficient motion and the way we're enabling the team is really as an expand motion.
Our next question from Dan at BMO.
Great. Thanks for taking my question. Maybe two ones. On the asset tagging, can you just RFID has been out for a long time. Can you just be explicit about sort of how asset tagging is different than an RFID device? And then on the ROI slide, insurance cost reduction was the biggest single component of the ROI for a lot of your customers. Maybe just expand on anything you can do to sort of further improve that ROI because it seems like with everything you've shown today, there's lots of further opportunities. Thanks.
I can take on the asset tag. So without getting too deep in the technical details, you're right. RFID has been around for a long time, and typically the way that will work is you'll have an RFID in, say, a piece of clothing or something like that, and it's going to be detected by a reader that's nearby. And there are different types of RFID technologies, but there's a dedicated reading device that's in close proximity. With our asset tags, they're actually able to communicate over a pretty wide range, so hundreds of feet or more, and they're connected to this mobile network of Samsara gateways that are across all the roads and operational sites and already moving around where our customers operate.
So without having to go deploy reader hardware, which is cost-prohibitive in these areas, these asset tags can be detected and reported, which makes it a fundamentally different experience. And in fact, we had Pike on stage with us this morning. They had this problem, and they'd tried using RFID as a solution and found that it just wasn't cost-effective or feasible to put RFID readers everywhere they operated. But now, because the asset tag uses the network that's already in place, they're able to have visibility across their assets.
If I can add one other piece on the technology front. So RFID works really well in retail locations, for example. You've got an enclosed setting. It's a store. You've got people doing inventory, and they can do it faster with RFID, but it is a passive technology, as Kiren was mentioning. So you need a dedicated proprietary—not proprietary, but you need a dedicated reader to read that tag, which for someone like Pike doesn't make sense if they left their tools behind somewhere in the dirt in the middle of the night. So the Bluetooth being active really has a pretty fundamental shift, and then it can be read by any number of devices, including any mobile phone. And so the usability, I think, of this technology is way greater, but it's a different price point, and so it's the right technology for the right place.
Our next question comes from Jim Fish.
Hey, thanks for the time here, guys. Maybe just following up there on tags. How are you guys looking to potentially expand? I think it got up to 1,000 feet that it had to be within a point of reference to. What can we do to make it so that, especially in those more remote areas, that you could get kind of that greater visibility? And then just on the go-to-market side of things, are we going to have to start moving to a kind of new packaging or a good, better, best model? Because unfortunately for you, these guys are creating more and more products for you.
That's a problem.
How are you guys thinking about that? Thank you.
Yeah, I'd say on the range side, we've been testing the product with early customers, and they've been blown away by how well these devices are being picked up out in the real world and in cases where they didn't expect them to be seen, and they're getting picked up every few minutes. So I think we and our customers are really happy with the range and the ability to handle all the things that come with physical operations: concrete, steel, etc. So we'll get the technology out there, and we'll see how do we want to keep improving it. But right now, customers have been really happy with the ability to read these products over to the ability to find their assets out in the world with the technology, and it's pretty game-changing for them.
Yeah, and I'll take the second one. Our sales guys are thrilled to have more things in their bag, and customers are really excited about the asset tags, for sure. But look, on pricing, we're not announcing anything in particular. We continue to look for how do we evolve our pricing to have a more fair exchange of value as we drive more and more innovation from the platform, and we will look to continue to evolve that over time.
Our next question comes from Dylan Becker.
Thanks. Maybe for Dom and Lara, maybe how should we think about the timeline of adoption within connected equipment? Going back to that slide, it's as big, if not bigger, than kind of the core products set today. Maybe the ROI use cases are a bit more nuanced, but you also can unlock incremental use cases or value tied to that. So how do you think about kind of those dynamics within connected equipment?
Well, I think we're starting to see more and more adoption. Again, if you go back to even just within the top 25 customers, 80% of them are using something outside of vehicles, and in most cases, that tends to be equipment monitors. So the adoption is quite high. I think there's an opportunity for us to go much, much deeper and find more and more use cases, and things like asset tags only increase the number of things that we can ultimately go to. But what's also really exciting is I showed you, even just within our core customers, the connected vehicle, the two applications, we've only saturated one-third on average of the commercial vehicles.
So there's so many more opportunities for the go-to-market team to go back into our existing customers and sell more of these vehicle-based applications. That's why you're seeing all three of these product categories growing at such high rates and so consistently because there's still so much opportunity across the board. I expect that to be pretty consistent.
Our next question comes from Matt Bullock with BFA.
Thank you. You mentioned that only about one-third of existing customers are penetrated right now with connected fleets. Can you talk about some of the gating factors preventing more rapid penetration of those customers and what Samsara is doing to accelerate that?
Yeah, I would go back to the slide that I showed with the top 25 customers and how they expand and how consistent it is over time. That's very typical for most of our customers. Even within the, I said, roughly 50% of net new ACV comes from expansions, and that tends to be upsells as opposed to cross-sells. So customers very consistently do these phased rollouts over time, and they may roll us out across a segment, a subsidiary, an operating company, and those will happen over time. If you also think about it, it's practical for customers to make changes maybe when the vehicle is at the depot for maintenance as opposed to pulling everything off the road at once, which would really zap their productivity. And so these things tend to happen in a phased approach.
Yeah, and I think also because we serve the largest customers, they grow through acquisition. I mean, you hear almost all of our customers talk about that, and so they will tend to standardize technology as they continue to acquire.
Okay, let's go to Alex.
I want to hit you with two more. So one of the things that a lot of investors consistently ask is, when are you going to raise price? How are you going to—how do you think about price? And from the just sheer volume of products you're launching, it doesn't seem like it's anytime soon. So as you think about maybe pricing the AI functionality that you demonstrated, if maybe it's the AirTag—not AirTag, sorry—the Asset Tag that you launched, how do we think about pricing from the lens of new product or new innovations versus just price increases? And then I got a quick follow-up.
Yeah, well, I think it's important that we make sure that we're monetizing the innovation, that there's a fair exchange of value as we add all this new functionality. And at the same time, we want to make sure that the pricing is simple for our customers to understand. As Dominic said, we still have a lot of net new market to go after, and we want to make sure that it is easy for our customers to understand. They see rapid ROI, and we can bring them onto the platform and continue selling them more products over time. So we're constantly iterating on our pricing and looking at ways to evolve it over time, but those are the two factors that we keep in mind: that fair exchange of value and simplicity and balancing those as we're growing.
It's a portfolio strategy. One of the things that I demoed live was refrigeration monitoring for the reefer trailer. That has a lot of value for our customers, especially if they've got tens of thousands of dollars of product in the back, and they're trying to monitor temperature they need in real time. They need it to work wherever kind of all the time. So it's cellular-based. The asset tag hits a different price point, and it's a different kind of asset typically. So we really do think of it as a portfolio as opposed to this one product will cover everything. Yeah. Then maybe the last one for me. If I think about a lot of the use cases, the federal verticals specifically and public sector jump out as a large opportunity. How are you doing with the timing and kind of the levels of FedRAMP certification? What do you need, and kind of how much do you unlock with each of those different levels of certification?
You want to?
I can talk a bit about the certification. So we are (this is public now) we're in process for State RAMP, which is the state and local equivalent of FedRAMP and federal certifications. And interestingly, in the operations market, the state and local markets are actually much, much larger than federal. So that's where we've been really having a lot of growth. We've been talking about it on earnings recently, and we're seeing that these certifications are going to be helpful down the road as state and local governments get more and more focused on cybersecurity. So we want to be ahead of that. So that's where we're focused now, and we'll continue to look at what customers are asking for across the different public sector agencies.
Let's go with Matt.
Thanks, Sanjit. One of the questions that we often get from investors is we see these as bullish arguments, but we sometimes get a question: is the OEM opportunity? We tend to see it as maybe gross margin accretive and helps you get scale, but some others worry that Ford or somebody like that could compress pricing, or there could be some headwinds to the OEM opportunity. Could you talk a little bit more about that? It didn't really come up today, but.
Yeah, absolutely. So a couple of thoughts there. First, Ford is a partner of ours. They're part of our app marketplace. They're an OEM partner. I think they were even here. I saw a few folks from Ford. So there, what we're able to do is integrate the data from your Ford trucks. They come with SIM cards and cell modems in them already. You can see the telematics in the cloud. Our customer, Kyle from Nutrien, also mentioned they do this with Deere. So we've now partnered with a number of different OEMs. We view it as new ways to get data into the cloud, and we want to do whatever is easiest for the customer. Our customers, they have heterogeneous assets. So they don't just buy Ford. They don't just buy GM. They don't just buy Deere. They have all of these things.
So they need a system of record that ties it all together. They need that orchestration there, which is what we provide. And I would say from a margin perspective, it's very similar to us. We're happy to provide it in either way. What we're finding is a lot of customers like the breadth of data we're able to provide through the vehicle gateway, but we give them a choice, and they're kind of voting with their feet. I don't know if you guys want to add anything.
No, that's fine.
All right. I think this concludes our question-and-answer portion. Thank you all for spending the day with us, and I hope you learned a lot more about the Samsara story. Bye, and we hope to see you again soon.