Trimble Inc. (TRMB) Earnings Call Transcript & Summary

Good afternoon, everyone, and thank you for logging in to our next fireside today at the Design Software Conference. My pleasure to introduce Trimble next. And representing Trimble today is their Chief Financial Officer, David Barnes. Dave, David, nice to have you today.

Thank you for having me. Look forward to it.

Awesome. So David, considering the fact that Trimble has been around for more than 40 years now, but it's also one of the most consistent and consistently evolving companies, I'd say. Only had 3 CEOs in the -- has had 3 CEOs in the history of the -- and everyone seems to be staying around 20 years or so. So very, very pragmatic and very well balanced in terms of the execution of the strategy. Can we talk a little bit about the new team that kind of has come in, yourself and Rob, obviously, moving from the CFO to the CEO role? As Rob has done that, obviously, came in and saying, we're going to kick off the next 20 years with a new strategy, which is the Connect & Scale part of it. If we think about it, it's the third year that we're really into it, right, just entering kind of it. How far do you think you have progressed on the journey today and some of the things that you're thinking about that are still ahead of you that are interesting or worth flagging up to us?

Yes. I'd say in pursuing the Connect & Scale vision, we've done a lot, and we still have a lot to do. So the -- what have we done? I think we've galvanized the organization around the idea. We have taken a number of steps that demonstrate the power of the strategy, which is all about workflows connected by industry platforms. We have early examples of the power of that in creating value for our customers and creating growth opportunities for our business. We have changed the way we're organized and the way we go to market to capitalize on that. So that's all good progress, but there's a lot more to be done. We have a history of being a very decentralized business. And our business process and technology architecture makes the Connect & Scale process difficult. And so we're doing a lot of it by kind of working around the way our process and technology infrastructure functions. And we're just in the early stages of a major digital transformation effort, which I think will catalyze our business around this, make it much easier to go to market in this connected way and I think will accelerate our performance. So we're really optimistic, but it's hard work. And we have -- as much as we've done it, I think we have a lot more to do.

Got you. And those things kind of go hand-in-hand, right? Sometimes I feel like there was a little bit of an expectation that, oh, for the first 2, 3 years, we're going to do that part. And then after for the 2 years, we're just going to scale and ramp up. Whereas I guess the way just very operationally things have been playing out the way I've observed it, and obviously, correct me if I'm wrong, but has been really interesting because I've seen some -- especially like in terms of the business model transition hand-in-hand going with kind of replatforming some of their assets and connecting them together when it makes sense. Some of them have been further along, whereas the others are kind of -- may take a little bit more integration efforts in order to get there. So that's going to come kind of 1, 2 years down the line. And because of that, something else is already kind of being scaling mode more than the connecting mode. Is that a fair assessment? And I'm seeing like especially some of the more recent bids, which were made in mind -- some of the acquisitions made in mind fitting with the together portfolio has been easier to kind of connect, and then the scaling part has already been ongoing.

I think it's a fair observation. There's a lot that's in common with our major end markets. And what they have in common is that they're all serving very difficult end-user workflows where productivity is really important and hard to achieve. They're all under-penetrated by digital technology. They all have a characteristic where they're effective point solutions, but the integration of the point solutions to make the whole workflow optimal is not where customers want it to be. So that's true across agriculture and construction and transportation. I will say, though, that, as you said, our businesses are in different stages of this evolution. Where we see the greatest opportunity and the greatest customer demand is in the Buildings and Infrastructure segment, and that's where we're the furthest along. So as we do our digital transformation, that's where we're focused first. We're actually in market now with an offering we call Trimble Construction One, which is revolutionary in some respects. So those who follow Trimble, know we've operated as many divisions and many brands. And we have a Viewpoint brand, which is enterprise software for the construction market. We have Tekla for structural design. We have a broad array of offerings, each of which has had their own go-to-market team, their own commercial terms. And the collaboration has been more informal than formal. So now we're in market. We're going to customers with a integrated offering called Trimble Construction One in which those customers sign one agreement under consistent terms to use a broad array of Trimble solutions. So this is an indication that, of all the variety of our end markets, construction is first on the list. And we've got real examples of customers that -- in some cases that we haven't done any business with in the past. More importantly, big customers that we've had business with as Viewpoint or another product category that are now greatly expanding their business with Trimble. So we're now in the market -- going to market as one company, solving our customers' biggest problems. And that's exciting.

Right. And that kind of shows that there's an element of this cross-sell opportunity between the different areas of the business, right -- that within the area of the business, but different products that previously had disjointed kind of efforts.

That's right. I mean disjointed may be harsh, but I would say, informally coordinated. But we have made it relatively difficult for many of our customers to buy a broad array of Trimble solutions. And we're hard at work to make it easy and transparent. The challenging part for us is that the process and system infrastructure behind that still is designed for this more disparate world, and that's the change we're in the midst of.

I mean, yes, there's a huge amount of will that I've picked up from both yourself and Rob over the last couple of years to do that. And what's also very welcome is that when I do my primary research, and I talk to your customers, general contractors and some of the largest in the U.S., for example, historically, the feedback has been, I cannot tell you how much I spend with Trimble. And it's not because I don't care, because I have 4 contracts between the Exploration Department, between the Design Department, which every GC kind of has as well in terms of the engineering between the Tekla kind of side. So I have so many contracts that I can -- I'd have to prepare myself to answer that question because there's not one contract. And some of the largest customers -- engineering customers have also said it would be much easier for us to consume kind of [ by one ]. So it's very, very nice to see that you've obviously received that feedback and are kind of completely on the same page with the way customers want to do that. So I guess that could be a nice tailwind to relationships that you have.

Yes. Gal, I'll offer this anecdote. When I joined Trimble a little over 2 years ago, my last pre-COVID business meeting was the kickoff meeting to one of our divisions, e-Builder. And I asked some of the people who serve our customers what they -- how they think about this Connect & Scale strategy. And in a decentralized company, it's natural for people to like the control of their separate pieces of the business. What I heard from our salespeople who serve our customers was that they want this to go harder and faster. They see a real need to meet the needs of their customers by leveraging the full scope and scale of Trimble. And they encouraged me in this early meeting to get after it harder. And there's a lot of companies with big strategies that come from the CEO, and there's a lot of selling to be done with the people. It was very encouraging for me to hear reinforcement right from the people who are serving our customers. I think this is a very powerful idea. We see some of our peers and competitors talking in a similar way about moving from point solutions to solving integrated problems. What we think is unique at Trimble is that we have the broadest array of solutions across most of our end markets to meet the most critical needs of customers. So Trimble is not alone in seeing this Connect & Scale opportunity, but I think we are unique in what we can bring to bear for our customers.

Yes. And when you talk about these unique solutions, they're both hardware and software solutions, right?

That's right.

So we can't discount one or the other. And where you come in is the fact that a lot of the hardware is there in order to automatically capture the data that needs to be then fed into the software world.

So Gal, I think that's a critical insight. These connected industry platforms all need as input data from the physical world, whether that's Geospatial data from a scanning solution or in the world of construction, having real-time data from the machine control system to know exactly what was done on the site to compare with the plan. And we have software-only peers that are trying to address similar problems. They're trying to do that through partnerships with companies that are connected to the physical world, and we think it's a unique strength of ours that we have in-house the solutions that meet those needs as well.

What we're finding is that also in the industry just broader, there is a sense of realization that no one will ever be able to solve all the problems themselves, right? Trimble has a great connection between the hardware and the software. But what if someone uses another piece of hardware and stuff. So what's been interesting to me to observe over the last 5 years or so is how technology barriers have been falling. And with the use of APIs and the connected kind of environments where we're not calling kind of this common data environment platforms that are kind of establishing Trimble Connect maybe one or [ others ] and all that stuff. It's interesting that these platforms are now able to speak together so that the whole industry for decades was built on very low interpretability whereas now, I guess, everyone has realized that there is an advantage in being part of that ecosystem. And I think Trimble, correct me if I'm wrong, but you are at the forefront of that, being very easy and open in order to kind of be able to work with everyone.

Absolutely. It's -- maybe the analogy is the Apple ecosystem versus the Android ecosystem. It's our belief that our platforms, the industry platforms that we're building need to be open to solutions from other providers. So some of our software peers or competitors have a massive installed base in, let's say, detailed 3D design. We don't want to in any way exclude those customers from having full access to the capability of the Trimble platform. And in fact, our strategy is open to -- through APIs, opening our platform to many other providers. We think that the Trimble ecosystem, there will be some natural advantages to connecting with Trimble solutions, but we understand and expect and encourage customers to continue to use other technologies, and we think that's absolutely part of the vision.

Got you. That's great. So talking about hardware and software, right? For 20 years, Trimble was predominantly a hardware business, and then for the next 20 years, started really building on the solutions side and adding a lot of software, all the kind of the major brands were added during that period of time, right? We're talking about Tekla, SketchUp, Viewpoint as well and e-Builder, like same thing in some of the other areas. Naturally, software over time does tend to outgrow hardware a little bit, right, especially as we kind of found that there's less commoditization going on in that space, and that's something that naturally then continues to yield higher returns in terms of the top line growth. Software and services have now kind of grown to represent between 55% and 60%, depending on which quarter we're looking at, how well hardware does of the overall business. And it's something that was -- the plan in the first place to kind of software to overtake -- software and services to overtake the hardware just naturally. Is that a trend that should -- we should expect to continue to grow in the future even when we look past '21 and we're looking maybe towards '23, '24, '25? Is that -- and long term, is there a balance that yourself and Rob are thinking where you think software and hardware kind of balance at a level that becomes very sustainable going forward?

Look, yes, I -- I'll say that actually, hardware even outgrew our software. If you look at the last quarter of the year, it's a pretty amazing story in a constrained supply chain situation or hardware...

Same for the backlog, right?

Yes. Backlog grew. So we -- absent supply chain constraints, we could have grown even faster. I'll say this, we think having a meaningful hardware business in offerings that connect to these platforms that optimize the customers' workflow is a good business to be in. And as part of the unique value we can offer, we talk about connecting the physical and digital worlds. And our hardware lets us do that better than anybody. We're not per se managing to a number. We encourage everybody to grow. I think in our hardware segments, the data indicates pretty clearly that we've outgrown the markets. We're gaining share against our competitors, and we like it. Those are high-return businesses. But I think your fundamental premise is correct, which is that more and more of the value is in the connected solutions, which is, by definition, software to optimize end customer workflows. And both organically and through acquisition, it's likely that the software will outgrow the hardware. I'm reluctant to give you a specific number. I actually think we have plenty of potential to continue to grow our hardware businesses, but the percentage will shrink as the software grows faster. That's sort of embedded in our financial plan for this year, and we think that will be a sustained trend going forward. But I guess as we do the math, I would say that given the growth rates in the markets we serve and our demonstrated ability to gain share in hardware, I think it's likely that our hardware business will continue to grow as well.

Got you. Awesome. Maybe if we kind of carry on with the software side and kind of thinking about some of the big changes that have been happening. Because when you buy a software company, in the past, most of them were based on perpetual licenses. I'd say most, except from the major ones -- obviously, e-Builder was fully SaaS when it got acquired. But like a Viewpoint, Tekla, SketchUp, those are some of the big ones, Vico as well in a way and some of those. When I look -- when you have a person looking at your software brands and portfolio, which of the larger brands today would you say kind of have now substantially fully transitioned? And which of them are kind of at more early innings [ of their ] -- or have yet to do so, so we can understand that road map, I guess?

Yes. So we made a major step in the last 2 big product families that had a majority of their business in perpetual software, that's Tekla and our MEP businesses. They are now I say fully converted. There are some situations in customers where continued sale of perpetual software makes sense. Other one I'll point out is our enterprise software business in transportation. We have still sold some perpetual licenses, but the way to think about this is that it's very small. In fact, of our perpetual software revenue, which, in 2021, was just under $0.5 billion, fully 75% of that was sold in a bundle with hardware, and only 25% was straight perpetual software sales. And that will go way down. We had a meaningful last-time perpetual license sale for Tekla, which increased the numbers in the third quarter of 2021, but that's passed us. So the way to think about our business is that the substantial majority of our software business is already converted to a recurring model, either a term license model or more prevalently a subscription model. And what's left to convert is the software that's bundled with hardware.

Got you. It's really helpful. And as you see -- do you see any opportunities, both on the software and hardware bundled, products to move that more towards subscription as well? Over time, we've seen certain areas of the market kind of moving into that. Obviously, the best publicized probably been telematics business that kind of went from upfront purchases to basically completely per month kind of usage model as well, especially some of the competitors. Is that a trend that you're seeing across the kind of that software, hardware integration -- or integrated assets?

Yes. It's a trend we're trying to lead. In fact, I think we're at the vanguard with civil construction, machine control and guidance market. We have an offering, Gal, I know you're aware of, called TPaaS or Trimble Platform-as-a-Service. We've been in the market offering that in the United States. The way that model works is you buy the hardware and the software and technology assurance and other services in a recurring bundle. The uptake on that has been slower than we expected. There -- it's a harder transition to make. Essentially, all of our hardware is sold through independent dealers. And so to convert the business model, you need to figure out how to keep everybody happy, including the dealer and the end customer. I think we're on track to find a way. I will say that when you operate on a global basis, there's some unique requirements. Things like VAT and other regulatory-driven requirements make it particularly difficult to sell Hardware-as-a-Service across border. So I think our -- we will have multiple models going forward for these integrated hardware and software bundles. In a number of cases, particularly outside the United States, our principal model will be we will continue to sell the hardware, but all of the software and technology assurance and services will be sold on a ratable basis, which is not the case today. And I think that's the conversion we'll see the preponderance of. But we've done this even outside our civil construction machine control and guidance. We've done it in some other markets where we're selling hardware and software bundled. And we are seeing very progressive customers open to the idea of less upfront capital outlay and paying over time to ensure that they have the latest technology and they get the best service. So I do believe that the -- this very big perpetual software business we have that's today bundled with hardware, I believe that will convert, but it will take some time, and the pace of that is very difficult to predict at this point.

Right. But overall, what you've seen over the last couple of years as a result of it you said is the top line probably has had a little bit of a headwind because of that ongoing transition.

Yes. It's been -- actually, we had a little less. We projected and guided about a 150 basis points headwind in 2021. We had less than that principally because we had a larger than expected last time buy of a perpetual software in Tekla. So this year -- but so it was about 100 basis points last year. It will be about 100 basis points this year in 2022. All other things being equal, I would predict that to be much lower in later years. But we have this big slog of perpetual software sold with hardware, which we want to convert not just because we and investors like recurring revenue models, but also because getting the data in the cloud enables the connected platforms to solve industry workflow issues. It's very hard to optimize your end users' workflow when the data sits either on the machine or on a laptop or a server somewhere. So that's where this recurring model, getting the data in the cloud, getting the software in the cloud helps the customer make the most of what -- the problems they need to solve.

It's interesting because when you're transitioning to the cloud, a lot of the workload is also transitioning -- sorry, when you're transitioning business model to subscription, it's not just like-for-like business model as top rank, right? And a lot of it is transitioning to the cloud as well, which basically means it's some sort of a new delivery model. If it's single tenant or multi-tenant, the idea is to be fully connected into one data set, right? So it's not -- again, it's not just moving one for one and just changing how you charge people.

Absolutely. In fact, the most effective -- I mean we have a -- successful conversions where the software works similarly, but the computation is in the cloud somewhere. But where the real value is unlocked is when you get the data in a place where it can be used to solve problems. So take the example in civil construction of Earthworks going on to build a road. If you can get that data back into the cloud to compare the as-designed model with the as-constructed model, the contracting firm and potentially even the owner can know real-time how precise the compliance was with the original design and the status of the project. If the data sits on the bulldozer or in the contractor's laptop, then you can't connect -- close that loop. So that's a key component of our desire to move these businesses into the cloud in a recurring revenue model.

That's really helpful. A result of all that is also that, obviously, your recurring revenue is growing, right? And recurring revenue is now across 1/3 of revenue overall.

Yes.

We saw -- I guess the best proof of that was the year you joined as a CFO, you walked into a business that was a lot more sustainable than if that had happened 10 years before that, right, probably be a different result. And the benefit -- the reward of it was the fact that you were basically able to protect the profit pool. You were able to kind of scale the business the way it was, and the recurring revenue continue to grow that year, right?

Right.

Does that mean that Trimble today is a lot cyclical, less exposed to some of these trends that historically it has been, especially around the macro volatility that we're seeing? We've mentioned crop prices in the past. We've mentioned construction activity and architectural billing in this, and all that stuff, but it seems like with a high proportion of recurring revenue kind of decoupling yourself from those direct relations.

Yes. I -- Gal, I joined Trimble at an interesting time, December 2019, right before the pandemic. One of the first things that I had to do as CFO working with my colleague was make projections in the face of this pandemic and try to figure out what our business trends would be. And I think there's probably no business that felt a lot of confidence in their predictions. And we modeled many scenarios. Our business was much more resilient than we expected. Now obviously, we did see a very small downtick in revenue in the worst quarters of 2020. And we saw a slowdown in the rate of growth of recurring revenue. But what's interesting, the point you just made, it kept growing. And I think that is an indication of the power of our recurring revenue business model and how essential our services are to the continued operation of our customers. So our customers, thankfully -- few went out of business. And if you're still in business, you need your enterprise system. If you're a contractor and you're -- you still have the earthmoving machines, you need the technology that allows them to work, you can cut back on a lot of other things. But we -- our solutions are essential to the way our customers work, and it shows the power of the business models. When your business is only transactional, you are definitely more vulnerable to economic cycles. I wouldn't say we are totally immune from them. I'm not sure anybody is totally immune from business cycles. But we are much more resilient, I would say, than equipment-only businesses or OEMs that are highly whipsawed by cyclical factors. We look at -- it's very interesting. In the civil construction area, if you look at some of the data, there was a meaningful slowdown in activity during the worst of the pandemic, and our business continued to grow. And from that, I infer that what we are selling, particularly our recurring revenue models, is needed even in tough times. In fact, in some -- there are some versions of tough times where you need technology more. So if you think about a farmer, and they are all anxious about input cost inflation, they're all suffering from shortages in the labor pool, and that makes them anxious, but it doesn't make them anxious about buying technology that enables them to use the inputs more efficiently or to have less reliance on skilled labor, and that's where our technologies fit in.

Sure. And I mean the good thing is that the only time that I've ever experienced kind of -- that not being fully reflected in your business was when if there is an external shock like a crop prices collapsing and things like that where their top line is at risk, right? But when the bottom line is at risk is great because at that stage, they have some sort of cash, but their cash needs to be spent in order to increase profitability, right?

Yes. Look, I think we were aware enough to know that the economic conditions for our customers do matter. And if a [ crop ] gets bad enough, and then we will not be fully insulated. What we have to do is be focused on solutions that meet their needs even in difficult times. The capital outlay for a contractor or a farmer for Trimble equipment is much less lower than for a new bulldozer or a new tractor. And as a result, if they need more productivity, and they don't want to invest the very big money in new equipment, there's still room to invest in technology to optimize what they already have.

Right. Especially because it immediately increases the yields, right?

Right.

Awesome. David, one thing that we've started talking about a lot over the last couple of years, and now finally, it's going to come into fruition at some stage, and I guess it's going to be very important, especially if the rate starts rising, is the infrastructure bill and the fact that there is a global need to improve the infrastructure. But specifically, especially in the U.S., we now have some framework to kind of look into that we say it's going to provide material amount of cash to support and renew some of that infrastructure that you need. The way I look at your business is -- and obviously, again, correct me if I'm wrong, but if I understand it correctly, I would say that a significant portion of Buildings and Infrastructure business is exposed to the infrastructure side. And then also from -- just from a Geospatial perspective, obviously a lot of the laser scanners are used into infrastructure surveying and other kind of activities. So would you say that you have a decent amount of exposure [ for us ] to that trend? And then secondly, have you seen any sort of early demand or at least a pipeline building on the back of what happened at the end of last year when we finally got that signed?

Yes. I'll talk about it this way. We are very exposed in a positive way to infrastructure investment. So a broad array of our offerings -- not all of what we sell in buildings and infrastructure is impacted. And obviously, we're talking about the U.S. principally, although similar initiatives are happening in many other countries. But the...

It was a year early, wasn't it?

That's right. And Europe, Japan has had some important stimulus efforts. So that business is -- stands to benefit as does, as you say, the Geospatial offering, which is actually likely to be the first directly impacted because you survey before you design, before you build, before you maintain. In terms of the direct impact, I'll say, well, the positive impact on our business is direct when projects are lacked, when projects are in the market, and contractors and engineers need solutions for a specific job. That has not happened. And actually, the wheels grind slowly. Many might remember the shovel-ready concept in the 2008 recession. That's not what this is about. I think this is more shovel-worthy, which takes more time. So these projects are going to really get going in 2023. But I do think there's a sentiment halo that's probably benefiting our business in ways that are difficult to quantify, but we're seeing positive sentiment among surveyors and contractors to make investments in advance of what they see are more positive trends. So I think it's having a immeasurable, not massive, positive impact on our business now and will really kick in next year.

Right. It's interesting because you said it's fairly measurable, right, by the virtue of Geospatial being in special -- separately reported segment as well. And it has been doing well, but it's been doing well for a very easy comp kind of coming out of 2020. So it's not...

Yes. It's been doing well, but I think a lot of our success is unique to us. We look at our competitors and their public reported numbers, which obviously break things up a little differently, so it's hard to make an apples-to-apples comparison. But we do believe in the Geospatial business. We're gaining market share. So some of the uplift we're seeing is not necessarily a result of the overall market being good. We're just getting more of our share of a pretty good market, and we think that momentum will continue.

Sure. That market is relatively better penetrated than some of the other markets you play in, right? It's also one of the older, more mature brands that you have. And I guess at this stage, it's more like iPhone. It works on a replacement cycle, right? So do you think that both innovation at the right time, plus the fact that people will have some sort of money to buy the later version could kind of kickstart the replacement cycle of it? So maybe the useful life has been expanding in the past. I've heard -- when I talk to some of your customers, they said -- they say we should buy every 5 years or so. We're buying every 7 to 10 years, right? And maybe some of those are on its last legs effective, but they work just from a technology perspective, just like my -- if I had an iPhone, it's 10 years old, probably not being able to update it anymore and stuff. So there's an element of kind of natural demand. It's above and beyond just kind of what the industry growth rate is because there's a decent amount of like fairly older assets in the field today?

Yes.

They still work [indiscernible].

I'd say the core -- that's right. It's a replacement model in much of the business. And innovation has to be really compelling to get someone to ditch their -- using your analogy, the 3- or 4-year-old iPhone for a new one. The new one needs to be meaningfully better. We have a product in the market. The product name is R12i, if anyone wants to see a good example of what I'm talking about. But it's a piece of survey equipment that dramatically improves the efficiency of the field crew because it doesn't require that the survey poll be vertical to make the accurate measurements. It seems like a small thing. But if you're a surveyor in the field working in a dangerous situation, you can go much faster and work safer. And we are seeing customers upgrade for innovations like this that make their workflows better. And so technology can induce more rapid replacement than would be just the wait until breaks rates kind of thing. We, though, also are seeing growth in our Geospatial business by cleverly going after markets that were underserved, if not by the industry, by us. So with some of your investors in the conference, I was talking about using Trimble scanners to do crime scene forensics. That's not a classical survey application, but it's one where our equipment takes a difficult job and makes it much more productive. And so I think we're or on a run of really effective product innovation and very solid go-to-market execution in our Geospatial business. That, plus a surprisingly good market has created tailwinds for us. Now I'm not suggesting that we can continue to grow our Geospatial business at the rate that we have in the last year. I think that's too much to expect. But I absolutely think the market will grow, and we will continue to outgrow the market.

Awesome. Yes. Yes, it's definitely a double-digit grower for a while, which in the past, we kind of referred to it as a 4%, 5% grower, right?

Yes. Look, I'll say we were pleasantly surprised by the magnitude of the demand. It helped that we had new products. It helped that we have product supply ordered for the new products. So our supply chain pressure was less in Geospatial than in some of our other areas with older technology in the market, but still it's been a remarkable success that has surprised all of us.

Awesome. We're almost running out of time, but one topic I did want to touch on is just -- one part that we've kind of ignored so far has been transportation side. And whilst you've made tremendous strides in building infrastructure side, Geospatial has really recovered. And I'd say, resources and utilities, you've been the leader in this space and continue to do that. Transportation has had some successful parts, but especially in the telematics business has been kind of under pressure both competitively and also just from a product cycle perspective. Can you give us a brief update on like where you are today on the transportation business? And maybe kind of try to think -- get us to think forward, and in the future, how that business will evolve, how it's going to be a little bit different? It has been, I guess, maybe less focused on ELD and stuff like that, but more focused on the enterprise software solutions, playing in a higher end again.

Sure. So for those who don't follow us closely, I'll categorize our transportation offering in 3 buckets. We provide enterprise software solutions for transportation companies. We provide mapping solutions that -- think of that as the Google Maps for transportation. And we have the telematics solutions to monitor and track the actual trucks. So the mobility or telematics business is about 40% of the total, and that's where our struggles have been. We had a major challenge implementing the ELD or electronic logging device mandate. We have a lot of unsatisfied customers at a time when there are a lot of new entrants in the market, and we lost a bunch of market share. So that's been a real struggle for us. I've been tracking the leading indicators of the business, which are bookings, particularly of our recurring business, customer retention. Those have been improving. And in fact, I think we would be further along, but the supply chain dynamics hit us particularly hard in transportation because we were less able to pass along the cost increase in prices, because product supply constraints were most problematic in preventing us from shipping product to end customers. And our big OEM customer could not -- they had their own supply chain issues, and they were not able to receive product, which hurt not only our hardware sales, but also our recurring revenue [Technical Difficulty] when they sell their machines to the end market. So you asked a good question, how is the market going to be different? First, we are targeting that our mobility or telematics business will improve. We have new products coming out. We will be less reliant on hardware. In fact, we would like hardware to be less and less a source of our revenue. We would like to make our money, and we believe we will make our money more in software and solutions. So that's an important way that our business will be different. We believe that we've solved most of the issues with product functionality. Our customers are happier. Our customer retention has secularly improved. But the other big way the business will be different is that we will more effectively capitalize on the bundle we have. No other competitor has the breadth of offering that I just mentioned with maps and enterprise software and mobility solutions. And we believe that an integrated Trimble offering can meet the productivity needs of our transportation customers, not only the carriers but the shippers as well, that we can be a catalyst to improving the productivity of the transportation industry. So we've got a lot of work to do. The problems are not all behind us, and the supply chain continues to add to our challenges. I think we're not telegraphing that we expect dramatic improvement in the next couple of quarters. But we do believe as the supply chain improves, as we get more effective in selling the Trimble bundle that we will emerge in the back half of the year with better trends. The key metrics we're watching are ARR growth and operating margins. And we think we have a line of sight to improve those, but we know investors are eager to see results, not words. And so we'll have to prove that.

Awesome. Yes, it sounds like a lot -- like the rest of Trimble, right, bundled together, kind of connecting and maybe have a better balanced growth as well, obviously. Throughout the ELD mandate, it's been kind of heavy on that side, and that's where kind of the headwind then came from, right? The fact that...

Yes. We -- I won't say everything is perfect in every business, but the headwinds we've really seen are in the mobility business. And as that improves, as we're less reliant on selling lower-margin hardware, we believe we have a path to improve trends.

Sure. Awesome. David, thank you. We're out of time. So I appreciate your time. As always, thank you for supporting us. And if investors do have follow-up questions, which they definitely do, please reach out to myself or Michael. We've obviously been very responsive to any request we've had. So thank you so much.

All right. Thanks to you, Gal, and thanks to the participants for their interest in our company.