Welcome to one of the first sessions of the morning. We have Itron, I-T-R-I, the CEO, Tom Deitrich. We'll have a fireside chat and some opportunity for questions from you towards the end. I'm Pavel Molchanov, research analyst. Tom, welcome.
Welcome.
Let's start with a topic that you've probably spent way too much time on over the last few years, which is availability of electrical components. This has been a problem throughout the pandemic. It seems like it's largely in the rearview mirror. Is that in fact the case?
The answer to that question is absolutely yes. The availability of electronic components is far better today than it was two years ago or a year ago. I would say that suppliers are showing up with the parts that they had promised on the dates that they had promised. Reliability is much, much better. The lead times are still a bit long in some particular areas. Pricing hasn't gone back to full pre-pandemic levels just yet, but some cracks in those walls starting to show up as well. That difference in component availability is a combination of certainly things getting a bit more balanced on the macroeconomic supply chain side of things and a lot of self-help on our side to multisource more components and find shorter transport routes to make things available for our product line.
We've benefited tremendously in 2023 because of the improvement in the supply chain itself. Just to put some numbers behind it, we came into 2023 with about $400 million of revenue that had been deferred or stuck in backlog. It wasn't going away. Our customers were impatiently waiting for it, but we were able to fulfill about $275 million of that $400 during 2023. The last $125 or so should be fulfilled during the first half of 2024.
I think you mentioned when reporting earnings a few weeks ago that you had $275 million of catch-up revenue this past year, and you still have $100 million of catch-up kind of yet to come in the first half of 2024. Did I get those numbers right?
Exactly. About $125 in the first half of this year is the last bit to be caught up.
That means all of the deferrals of product delivery because of component shortages will be kind of fulfilled at that point.
Correct.
Okay. Let's zoom out for a moment, and I'm sure you'll be talking about this at the Analyst Day next week. Software adoption by utilities has been kind of a big topic of conversation. Is it fair to say that the adoption curve has not been as rapid as many industry players might have expected, let's say, 3-5 years ago?
I also think the answer to that question is absolutely yes. The utility industry being very, very focused on resiliency and reliability and risk avoidance. No one wants the lights not to turn on when the switch is flipped. So things tend to move at a relatively slow pace based on that fundamental societal need of risk avoidance, but also the regulatory model that goes along with utilities. Most of my customer base is regulated utilities, which means that they've got to make a case to their appropriate commission and get permission to spend money and then put that generally into some sort of rate base. How that happens with hardware is a pretty straightforward CapEx discussion, and that is easier for a regulator to see their way through. Software gets to be a little bit more complicated to understand what is the value that you're getting.
So naturally, the combination of those two factors has made it go a little bit slower. That said, the definition of slower, I think, is all relative. The number of regulatory bodies that were willing to accept SaaS in a rate case as part of a CapEx arrangement was relatively small five years ago. Today, I think 37, 38 states out of the 50 in the U.S. have some sort of performance-based rates. So the regulatory model is changing. But even more importantly, the need on the part of the utilities is changing pretty dramatically. No longer can you predict what's going to happen 10 years from now. That old model of let me buy a fixed capability, put it in the ground, and amortize it over 10 years with a certain fixed capability, that doesn't work in today's world. The world changes too fast.
You don't know what you're going to need next year or next quarter even with the rate of change that is happening in the industry and what the technology can do to address that. So there's real pressure on the part of the market forces and us as consumers, which is speeding this whole thing up a little bit more. So more and more optimistic of that rate of change continuing to increase. Perhaps it's never as fast as I would personally like, but we want to make sure we've got a way that we can help our customers adapt to that rapidly changing world, which is a good piece of value for us and for our investors.
Provide a few use cases just specifically. What do utilities win from software adoption?
Right. A couple of examples that I would give you. So first and foremost is managing distributed energy resources. So when someone buys an EV and puts it into their garage, that electrical load is the equivalent of a small house. So when that thing plugs in and turns on, it's like a new house just popped up. The added complexity here is it isn't just like one dishwasher and one light and then one air conditioner. It's just 0 to full charge in a second. So that variable load is very, very difficult to understand. And if you have too many EVs that are charging in the same cul-de-sac, let's say, you're going to overload the distribution transformer. Today, the utility really doesn't know what is happening out in their territory.
They're flying blind, and they're going to end up with popping distribution transformers and creating outages, or they won't know how to work with their consumers to be able to control that rate of charging and roll people in time of use or other mechanisms to be able to balance out supply and demand. The same would be true for a vegetation or a wildfire type of situation. By watching and being able to understand what's happening in the lines, you can see when a tree branch starts to touch a wire. You will be able to detect that in the signals using software, and therefore you can proactively roll the truck to snip the branch and prevent catastrophic wildfires or outages or what have you. So you can use software to improve the ability to engage with your consumers.
You can use software to improve the resiliency and the reliability and reduce the amount of outages you have. All of the information is available. How do you assemble it and be able to take real actions from it? And that's the basis of our Outcomes business is let's make use of all of the wonderful data that's out there and put it to use to solve real-world problems.
How big is Outcomes within the revenue mix?
Today, it is a $280 million a year or so business out of our $2.2 billion last year in round figures. So it's a smaller piece of the total company. It is our highest gross margin business, and it happens to have the highest growth rate in terms of what's going to happen in the marketplace for all the things that I just talked about. The beauty of all of this is these things move at the speed of software. They don't move at the speed of mass-scale hardware deployments. When a customer sets up a network project for us, for example, it might take two or three years from the time you've gotten regulatory approval, which might take a year to get regulatory approval. Then you run that hardware project over two or three years to figure out how to get all that stuff in place.
All of the logistics and all of the work to be able to touch each and every dwelling, to put a meter on the side of a house as an example, that's a lot of hard work. Once you've got that installed, why not be able to use that capability more like a smartphone? I don't mean a consumer device. I mean being able to download an application into that endpoint to be able to look for an EV that's behind the meter and be able to report it back and say, "Hey, this house has one. Do you want to enroll them?" Why not download an application to look for distribution transformers that are going bad so you can proactively upsize them or go fix them before you have an outage? Those types of things are possible at the speed of software.
We find that from the time we've got an idea until we can prototype an application with this downloadable application stuff is measured in months and not in years. That's only going to get faster the more applications we develop with our customers.
When we think about smart meters, so we're kind of turning to hardware now, I think about 70% of the electric meters in this country are smart. Is that about right?
Yeah, it's probably a little higher than that with some type of communication capability, probably closer to 80. But I mean, it's the majority of them. But again, I think if you and maybe I'm anticipating where your question is going, Pavel, I don't look at these things that we're putting into the marketplace as meters any longer. You're not thinking about it in how we see what we are selling to our customers. It is a distributed grid sensor. Okay, it happens to measure water or electricity or gas. So your iPhone happens to also make a phone call. But I think generally when you pick it up, you aren't making a call with it. You are doing mobile banking or you are calling Uber or you are doing some other application. The same is true for the products that we sell today.
They happen to do metrology, but now they have an extra processor capability that you can equip for a lot of different value-added applications. That's really what is the thin edge of the wedge for us. The terrific backlog that we have, $4.5 billion in backlog, is driven by the need for utilities to have a much more agile operating system, the ability to anticipate what problems would be and adapt and grow their network as they need it. The smart meter penetration to the point that you were going down, that penetration is reasonably high in North America for electricity meters. But all of that first-generation infrastructure is now being upgraded to this notion of a much more agile set of capabilities.
I was going to ask, I mean, there are states, and Florida being one of them, in fact, where I think 100% of meters are smart, just like there are countries in Europe, Italy comes to mind, which had smart meters for 20 years. What does the replacement cycle look like now that those 20-year-old products are coming to the end of their useful life?
Right. You could look at Southern California or you could look at Florida where some of those early adopters with the very first-generation smart meter solution was 2008, 2009 kind of timeframe. They're back for refresh now. In general, the way we are approaching our customers on a refresh is you don't have to rip and replace the whole thing. What you can do is just start layering in this new capability where you need it. This isn't some massive forklift upgrade for a lot of our customers. There is a smooth transition. And why that is so important to the customers is what is in their rate base, right? If they have capital that is not yet depreciated, what are they going to do with that? How do they go to the regulator and say, "Yeah, the stuff I bought, I don't want to use that anymore.
So please write that off and get the ratepayer to pay all this, buy all this new stuff." That doesn't work. So how do you work in that kind of environment? Our technology is forward and backward compatible so you can add more capability to it. Okay, maybe the older meter might not have this capability, but you can put a new grid sensor right next to it, a new endpoint right next to it, and now you have that capability there. So you can roll it out over a longer period of time or a shorter period of time, depending on how you want to attack it.
That has the benefit for us of higher software content, of course, but it also starts to create a smoother growth path for the business overall instead of these big projects that kind of come and go, which creates lumpiness in the revenue. That's a fundamental change in our business model, buying pattern, and the technology need that's out there, which is something we've worked the last five years to put in place.
So on both sides of the Atlantic, there is the replacement cycle from those kind of older initial installations. What does it look like in emerging markets? So for example, Itron's had a presence in India for quite a while and in Brazil. Big difference. Power demand's actually growing, right? So is population. So presumably that first adoption has not even taken place yet.
Right. Emerging markets, I think, are split in some ways. I'll call them almost 50/50 where you would have a market that is going to go for the cheapest thing so that they're not looking for value add. They just want to be able to start billing customers. And then you also have the other part of the market, which sees where it's going and it's driven by population growth or by consumption or loss reduction where they're jumping to this next-generation technology. And those are the markets where we are really prioritizing. When you've seen us make portfolio moves over the last couple of years, it has been to prioritize markets that are going where our technology is going and not trying to compete for the very bottom of the market.
That has led to improvements in the devices' gross margin, for example, if you're looking at some of our segments. It also leads us to be able to skate to where the puck will be.
Okay. One theme that Itron, I remember, has talked about going back way pre-COVID is expanding your software capabilities through partnerships, joint ventures, and potentially M&A, kind of full-fledged M&A. At least on the M&A front, it's been kind of slim to none over the last several years. Can you talk about that and sort of hypothetically, what are the attributes of the businesses that you would potentially want to acquire?
Right. Let me talk about that from two perspectives. Set the acquisition piece of it aside for the first part. The ability for our customers to get full visibility into the distribution grid and really understand what's happening in their territory means that they've got to connect up lots and lots of different software solutions in the back office to really be able to make sense of all of this. That is an area that is very complicated for our customers and definitely slows part of the adoption curve for newer and newer technologies. So what you've seen from us over the last several weeks are a flurry of collaboration activities with others in the industry. So GE Vernova and Schneider Electric integrating their advanced distribution management system, the stuff that works from the substation up in the electrical grid and our stuff, which works from the substation down.
Why would we do that? It's glue those two pieces of data together and pre-integrate it on our side. That allows our customers to be able to make use of a much more robust set of data to be able to take action on things. So that type of collaboration speeds up the adoption on the part of the customers, and it's absolutely what is behind those collaborations. That same idea with Microsoft to integrate AI using the OpenAI API on the data itself, enable our customers to get better insight from the data much, much faster. Those types of collaborative efforts, you'll see more of those from us as we try to round things out and speed the pace of adoption and obviously improve our business. That's one half of the answer.
The other half, when it comes to non-organic development activities, the last couple of years, we clearly were more tentative on the acquisition front as we were working through supply chain challenges. With those in the rearview mirror, it opens up the aperture. We've been extremely active in this space, and I expect you'll see some things in the not-too-distant future. What are we looking for? It is something to expand that Outcomes portfolio. We're looking for a platform which is scalable across multiple utilities and across multiple verticals. We don't want something bespoke that we'd have to re-engineer customer by customer. Areas of interest for us clearly are more things in machine learning and AI, more things in grid planning and integration activities. Those are areas clearly that are interesting from our development and rounding out our Outcomes portfolio.
Just conceptually, you have pretty massive R&D capability, and you certainly have the wherewithal to develop kind of new apps in-house. On the spectrum of build versus buy, is it just speed to market that would potentially argue for the buy option?
I certainly think there's a time-to-market component to it. There really is a finding something that is not part of our existing portfolio that we can slot in from a technology standpoint as well. Just a technology capability we don't have would be another. Customer access, if there are customers that have already bought into a particular solution rather than trying to sell into that and displace it, maybe we would take it on. But those are the types of things that would lead us to more of an acquisition or a buy versus a make capability.
On the hardware side, are you pretty well set? I mean, is there anything conceivably that could be interesting there?
I would say I think we're in pretty good shape for the areas that we are playing in today. I think that there will always be things that we will watch for in terms of new types of sensing capabilities that are out there. If the market looks to be very big, then probably that's something that we would want to have internally. If it is more of a partnership opportunity, it's probably something that's lower volume on the hardware side. But think sensor technology might be an area of interest, but certainly prioritize lower than software types of capability.
Last question before we go to Q&A from the audience. On the kind of policy or regulatory front, you already touched on the importance of having software be included in the rate base for utility CapEx. Beyond that, what else are you looking for?
Well, I think I very much want to see all of the stimulus programs around the globe that have been announced really have that money flow through and start to show up. I'll use IIJA and IRA, so the Infrastructure Investment and Jobs Act and Inflation Reduction Act. Those are big policy moves on the part of the United States that show tremendous promise to help grid resiliency and reliability and improve water quality and availability across the U.S., increase the amount of EV chargers that are out there to create a nationwide network. Good policy moves from an infrastructure standpoint that pays dividend. You'll get a return on that capital just in terms of economic growth for society. How you get that money to move through the bureaucratic process is indeed challenging. There's a bunch of rules in there as to how the money flows.
It's taken a bit of time for us to roll through. I would love to see that stuff go faster. I would like to see more product-level waivers for Buy America / Buy American, where you need a certain amount of domestic content, where a lot of semiconductors are outside of the U.S. in terms of the content. Then you've got to go through a painful waiver process. Waivers are absolutely being approved, but I think there's a more efficient way that these things could go faster. Those are some of the policy moves that I think could be helpful. Again, it's an acceleration issue. It's not a question of if these things are happening. They are. It's going to happen. It's a matter of can we make them move a little bit faster?
All right. Questions right up front.
This question has a little bit of a burden of history, but are all our manufacturing issues behind us? Forget about supply sources. We had a lot of high-cost plants in Europe. Do you feel that the infrastructure now is well balanced, moving and humming appropriately, or do we have more of that to come down the road?
Right. The question is, if everyone couldn't hear it, where are we on our manufacturing transition? Is that stuff in the rearview mirror or is there another shoe to drop? Loosely paraphrased. I hope I got it right. The answer to that is that by the end of 2024, the manufacturing footprint that we have, I think, is balanced to where our portfolio will be. So we now have the right number of plants for the product portfolio that we have. Okay, if you could have portfolio changes, well, then clearly something else could happen, but I'm not anticipating that. We have two factories that are in flight for closure now, one in North America, which is primarily in our network segment, and one in France, which is primarily a devices factory. But those transitions are going on now and we'll be out of them by the end of 2024.
And then I think we have the footprint that we want. For, not everyone who may have the historical perspective, about 5 years ago, we had close to 30 factories around the globe. It'll be more like 6 by the end of this year. And that is the right footprint for the portfolio we have in place today.
Other questions in the back?
Driving out the dumb grid, do you need to have the DI- capable meters sufficiently deployed before you can really see the Outcomes?
Certainly helps by having the latest generation of hardware out there. You can build the software on there faster. Certainly, things like DI apps will be required to have different hardware into the field. That said, DI is a small piece of our Outcomes revenue today. It is not required that we have our hardware to be able to have that software base. I'll take a customer like Exelon, multi-state number of different properties that they have, big utility where they have a couple of their OpCos, which are not on our hardware. Our business analytics runs over the top of all of that. So it works better with ours, I will humbly say. But nonetheless, if it is someone else's hardware underneath it, no harm, no foul, we can still generate good recurring revenue in the Outcomes space.
Any other questions? Just a follow-up.
Yeah, I'm going to say, are you able to elaborate or try to help draw the implications of broader charging infrastructure and what that does for your growth rates and TAMs in all of your products so analysts and investors can understand what the long-term opportunity set for you?
Right. Question again, if I loosely paraphrase, if I get it wrong, just redirect. But at Analyst Day coming up in another few days, are we going to talk about the implications for electrification of transportation and what a broader EV charging network would do and what would that mean? So the answer is yes, but more. The Analyst Day, which is coming up, we will have an overview of our platform and what it can do and why it helps and what benefits it creates to enable electrification of transportation, grid resiliency and reliability, sustainability, how we can help our customers reduce their greenhouse gas emissions where they have their sustainability goals, how that works. We'll have a customer talk to you about how they are using the technology, and then we'll lay out some longer-term financial ratios and what the model looks like. So the answer is, Yes.
I think it'll be a good event. You get a pretty comprehensive 360-degree view of the business from a customer technology financial point of view.
Time for one more question. All right. Not seeing any. We will have a breakout session downstairs right after this. Tom, thank you very much.
Thank you very much. Appreciate it.