Cerence Inc. (CRNC) Earnings Call Transcript & Summary

Okay. Great. Thank you, everybody, for joining us. My name is Mark Delaney, and I cover Cerence for Goldman Sachs. Very pleased to have with us Tom Beaudoin, the CFO; and Rich Yerganian, VP of IR. Thanks for both coming.

Sure.

Look, I mean I thought maybe to get started, you could level set us some of the key products that Cerence offers in voice technology and what sort of content per car you typically see today?

Sure. We've built up a quite strong franchise in the auto space over a number of years, both as a business unit within Nuance and then post spin as Cerence. We have really 3 revenue streams; we have an edge embedded set of technologies that have moved over the years from kind of traditional command and control, to now conversational AI plus additional technologies around signal enhancement that allows us to move from just the digital cockpit, to more of the in-cabin experience, which I think is quite important as you enhance that user experience. But also as you move to electric vehicles and provide that differentiated solution that each of the OEMs is trying to drive, we have a connected set of solutions, in the cases where we need to go out and get data, in order to provide that answers or information or data that the driver of the passenger is trying to gain, whether it be parking or gas stations or restaurant reservations. We can do that through voice. We can also do it through a gaze technology that we have. And then we have a professional services revenue stream. That's again, another piece of the differentiated offering that we provide, versus the tech companies, where we have a large R&D and PS support, they're located all over the globe. And once we do a booking, they work with the Tier 1s, with the OEM, to then implement that technology stack, whether it be in the embedded technology or in the connected technology. Typically, from a booking to when -- what we call an SOP, start of production, is about 12 to 18 months. On the embedded side, it's a running royalty model, where the OEMs tell us how many cars they produced, and we take -- we get paid for that and we take that revenue upon that production. On the embedded side, our average is about $4 to $5 a car. It can range from kind of $1 or $2, if it's kind of low-end technology, up to $7, $8, $9 a car. We see continued growth in that over the years as they assimilate more and more of our technology into their embedded solutions. On the connected side, again, we get a royalty report around the production of that car. We get paid for that. It's typically $1.50 to $1.75 per year. because many of those contracts have varying lines of service that we provide on those connected services. It could be 5 years, 7 years, whatever. And then, of course, that's -- we amortize that revenue over whatever the length of that service agreement. As I said, the technology is around enhancing that driver passenger experience within the car, is kind of a multimodality set of services and offerings that we provide. We believe the reason we're succeeding in the way that we're winning and continue with -- probably around 2 or 3 things. One is, we've been working with these OEMs for many, many years, and we have kind of deep relationships with them. The OEMs are really conscious about controlling the data, controlling that customer experience, trying to have a differentiated experience and then continuing to drive the ability to create that kind of third monetization model that they want. So they get paid when they sell the car, they have a service capability, but they have an emerging opportunity to collect services through the life of the car, and we help to enable that. And the differentiation between us and some of the big tech companies, is we provide them with that differentiated solution versus a more homogenized. They own the data, and we have the strong services capability to stand up -- whatever they want from a technology standpoint. We coexist though with, say, an Apple because people want to bring their phones in. And in many cases, we're still in the background, enabling that technology to work in the automobile.

You shared some helpful content opportunities, of course, it ranges depending on the customer and the future. But to clarify, when we destack, right? So there's an embedded price and then connected things later on top.

Yes. Connected, we only sell connected on top of our embedded solution, and we try to put as much as we can in the embedded, because then you're not dependent upon an Internet connection. It's a car, who knows where it's going to be on the top of the mountain or in the middle of the city where it's easy. But there is certain activities that need to happen, where you just need to go out and get the data, to provide that information to the driver.

And then I think kind of all in, it's been something like $4 to $5 per car on average. Correct me if I'm wrong, but where do you see that going over time, and what are some of the key features that could potentially drive increased content?

It is about, as I said, about 4% to 5%, and that's just for the embedded. So if there's connected, that's on top of that. As the OEMs continue to add additional content, then that increases, and that's what I've said. We do have contracts that are at a much higher rate, $7, $8, $9. That gets built up over time. We've closed deals that have those higher prices. But as I said, it takes time to implement them, get the cars into production and get them ramped up. So we see an opportunity to increase the ASP, both on the embedded side and on the connected side.

Rich, maybe I can go to you for this one, on attach rates. I mean, how do you think it is at the industry level in terms of penetration of voice and where do you see that going?

Sure. So the -- we're on about 52% of cars on a global basis. So global production each year, about 50% to 52% of them ship with some level of our technology in them. And for the percentage of cars that have a voice assistant in them, because not all cars do, we have about 80% share. So we work with just about every car manufacturer to a varying degree, whether it be a new EV startup or whether it be a well-established company that's been around for a long time. On the connected side of the business, our market share, again, it's a smaller percentage of the overall market that has that capability, connected car capability, because it's a relatively newer technology. Our share is about 25%. It's a more fragmented market, and it's an area that we think we can grow some over time, as more and more cars get connected, and we leverage our strong installed base on the embedded side.

In the competitive landscape, in those different areas embedded versus connected, there is the -- is it the same types of competitors, or does it vary?

Yes. It is pretty much the same set of competitors. The competitors break into 2 categories. We have niche competitors and then we have the consumer tech giants, if you will. From the niche perspective, there's one company in China, in particular, that provides both embedded and connected services. They're a very worthy competitor for the Chinese, indigenous Chinese car companies, being a Chinese company that lends some natural advantage. Although we've done very well there, off late. And part of that is, because as a lot of those Chinese car companies look to go global, we have that global support that other companies don't. And then there's other smaller companies as well. But from a share perspective, no one really rises to that much on the embedded side. Similarly, on the consumer tech side, we have a pretty strong competitive moat, and that moat comes around 3 areas; because there is the potential of Google, for example. They have a product called Google GAS. It has the potential to displace us. But our moat around a product like that falls into 3 areas. #1, first and foremost, is that we create a customized branded experience for the OEM. And at a time where the cars are becoming more digital, it's becoming more and more important for the OEM to own that brand experience inside the car. And we help them do that, as a white label way behind the scenes, but we customize and implement the technology as they see fit. Right behind that, if not equal to it in importance, is data. Our goal is to help our customers put a system in place that generates data from the use of the customer, that allows the car manufacturer to monetize that data over the life of the car. That's ultimately where they want to go. They don't want to just have a point of sale one time, and then really not have much opportunity for revenue from those cars as a follow-on. They want to monetize the experience of that car throughout its life. And our technology allows them to do that, and the key differentiator is with our solution, while we use the data to train our AI, the ownership of that data resides with the OEM and allows them to monetize it. That's in direct conflict with what the consumer tech guys want to do, because they want to be in a car not because they're auto enthusiasts, but because they want their access to the data. And then the third thing is, we're a global supplier, and we're unbiased in terms of the various people that we work with. We'll work with Amazon. We'll work with Google. We'll work with in Apple, whoever it may be. We have no biases, right? Because we see them everyone playing a role inside the car. And by the way, in China, that's Alibaba, it's Tencent, it's Baidu, right, It's a whole different set of people. And so we work on a global basis from that perspective, and we're not in a position where we want to focus any kind of system to favor one or the other, whereas if you have a consumer tech giant, as we all know the history, what they tend to do, is that when they're in control, they want to kind of exclude everyone else. And that's not really the kind of system that our customers want.

You both have brought up that point now, around being able to have access to data and how important that is for the OEMs and the direction they want to drive these businesses. Maybe we could double click a little bit on that, when this data is being generated, where does that live? Is that some Cerence server or Cerence lease server in a big data center, or it's in with the OEMs, just kind of want to understand that?

Yes, it is [ in ] with the OEMs, right? So they're the ones that are responsible for the privacy of that data, the security of that data. It's through the OEM. It goes through our system to, and gets passed through to them.

Okay. There's a lot of interesting features that I think you're building out support for one of -- the ones that resonated with me recently, was an emergency vehicle detection, right? We've heard that EV companies, in particular, are having trouble with that, and you need to hear what's around them and recognize. I think a number of companies have just spoken to that as a challenge. Can you speak a little bit more, on traction with that feature?

Yes. I'll expand on that a little bit. And that's an interesting one, right? Because as we talk about where Cerence would go, we start with voice, and voice is a very key part of our technology. But we're expanding beyond voice and emergency vehicle detection really has nothing to do with voice, but it has to do with AI and sound, and being able to hear the oncoming approach of an emergency vehicle, which, by the way, I think there's around 1,500 unique emergency signal sirens throughout the world. And these cars can end up anywhere. So we have to be able to identify those. Even if you're listening to music in the car, you're engaging conversation, we can hear that and we can -- or the driver. And that's something that is seeing a very strong adoption rate, I think, success rate, and one that I think is just going to continue to be -- it's one of those things that first shows up in the premium models, which is typical. And then as that technology matures, starts waterfalling its way back down. The other exciting part of emergency vehicle detection is that, there's more and more cars that are now starting to have microphones on the outside of the car, and what that allows us to do is, not only hear the signal, but determine the direction that it's coming from. And so, many times, you can hear an emergency and you wonder where is it coming from? You don't see it yet, and we'll be able to alert the driver of what direction it's coming from. Our technology comes around 3 cores. The first core of our business is safety, right? And that's where we came from because, using your voice is the safest way to engage with the car, because you don't have to take your eyes off the wheel to go down and see where the buttons are and take their eyes off the road. The second pillar of our technology is productivity. We've talked about Microsoft Teams as an example, that we're working with Microsoft to implement that inside a car. And then the third one is entertainment because especially as cars become more automated, it gives more opportunities for people to be entertained in the car. So those are the 3 things that we kind of are -- kind of following, as we go down the journey.

And for something like emergency vehicle detection, is that something more being sold and demoed to traditional OEMs or more robo taxi types of companies?

No, more traditional ones as well yes. And again, it gets -- to what Tom said earlier, there's a value that companies assign to that, right? And so they're willing to pay for that, because it's a capability that they all desire to have and offer their customers.

Maybe when we think about the business broadly, help us better understand if you could, please, what kind of exposure you have to some of these newer areas, robotaxi companies, pure-play electric vehicle companies versus some of the more traditional OEMs?

Yes. I mean, to us, it really there's no difference between internal combustion or electric vehicle, other than as cars become more electric and more digital, there's more opportunities, and there's a greater adopter of our technologies because they tend to be leading forward in terms of what they want to provide their customers. So from a robo taxi perspective, it's still a very nascent market, right? So there isn't a whole lot of to go on there. But the idea is part of what we're looking to do, and we'll talk more about this at our Analyst Day, is to expand beyond just the cockpit, meaning the driver into the cabin, right? And expanding into that area of the cabin provides a lot more opportunities, of which things like a robotaxi plays into that, right? Because now you're really just concerned about what's happening inside that cabin, you're not concerned about driving the car.

Tom, I wanted to ask you, I mean you've known this business for a long time, dating into your time at Nuance and then involved -- also right now back with Cerence. It was something -- when you came back to the business that you think the company is doing better than you would have expected? And any areas you're saying, oh, that didn't go as well as I was hoping?

Well, I've also been involved since the spin on the Board. So I've been pretty active in some of the plans and strategies. I would say -- I think the things that we continue to do well at, is this expansion around the experience within the car, and continuing to move and leverage our voice and assistant technology into some of these other areas around some of the signal enhancement. We can create bubble zones from a speaker around the car. That again, gets to some of the things that Rich talked about. So think about a mother in the front seat, the kid's in the third roam is behaving, what does she do today? She turns her head and put the kids in the back seat. With these bubble zones, you could stay focused, keep eyes on the road and have a direct conversation. We have a product called Cerence Sing, any of you that like karaoke, which creates a karaoke type environment within the car, gets to the entertainment aspect that Rich was talking about. So I think continuing to support the OEMs, and that advancement of what's happening within the cockpit to support them in their monetization efforts, to continue to really provide an opportunity to differentiate themselves in that experience, as that becomes much more important. Think about an EV car, they can only really in the future, differentiate themselves between what the car looks like, and that experience inside -- a battery is a battery. There's no more powertrains. There's no more engines. So that's going to become more important, which is why I think we're seeing higher adoption of the technology in some of those applications. What -- as I've got more into the operations as the CFO of the Board, we've done a lot of work the last few months around really our strategic planning in some adjacent markets that the previous management was pursuing, things like fitness and elevators. We're going to defocus those. We think we have a really strong franchise in auto. We need to continue to build that out, because we still have opportunities to grow penetration, particularly in connected. We still have opportunities to grow ASPs. We continue to build out the broad set of technologies. But we also think we have a strong opportunity, and we'll talk more about this at the Analyst Day, around adjacencies within transportation. So think about 2-wheelers. We just made an announcement this morning about a deal CFMOTO in China. They're going to deploy our technology on their 2-wheelers and ATVs. We've signed 6 deals on 2-wheelers. And again, we have the voice and conversational capabilities, but without the signal enhancement capabilities, combining that, this is an open cockpit. So you've got to bring the 2 together, in order to create that opportunity. We signed a good truck deal 2 quarters ago, and I think there's continuing opportunities in that space. We signed an RV. The beauty of those is they have reasonable TAMs to expand onto our auto TAM, but they leverage our core technology stack, right? And so they keep us focused, they keep us within 'our knitting,' and they create good opportunities. And then we're incubating a team. There's a couple of technologies that aren't subject to the FOU. So for those of you who don't know, when we spun the company out, we had a 5-year Field Of Use restriction with Nuance. So we were restricted to the transportation market. Nuance was restricted from competing against us in the transportation space. There's a couple of small technologies that are outside of that. So this group is looking at technology and licensing agreements with some consumer and industrial tech companies, where we can be the technology provider, create a license stream. Some of those have a much shorter bookings to revenue, than the 12 to 18 months it takes in auto. And we don't have to build the supply chain or the go to market. Our customers will do that. And then when the Field Of Use expires in October of 2024, we can hand over some more technologies to that group and say go at it and continue to build that out. So that's really our strategic focus going forward. Continue to leverage and focus on that core auto, which still has very strong capabilities. Attack some adjacencies, very close to leverage the technology and transportation, and then leverage our technology and expanding markets.

And just to follow up on a few of those points. So even when the Field Of Use expires in October 24, your expectation today is that, these adjacent markets is more of a licensing type of model?

Yes, I think just think about it. I've got a lot of experience. Nuance had 3 big divisions that leverage their speech and AI capabilities, healthcare, enterprise call center and what ended up to be auto. And the thing that you learn, if you're going to be a vertical in those, is that the language models are very dependent upon the industry that you're in. So you think about Nuance, they've got 15 to 20 years of healthcare language models. It's a big moat for them, right, because healthcare language is different. And then you also have to understand the workflows in those. And I think, to say that we're going to go recreate a huge vertical and compete against Nuance in healthcare or a compete against the massive vendors there are in the call center space, is probably not the smartest allocation of our capital, particularly when I think we'll lay out some really strong TAM numbers and building out that transportation and then trying to be a technology provider to companies, then we can leverage their vertical capabilities within a broad set of markets, whether it be consumer or industrial.

And you said there's a few areas you can license sooner than that. It sounds like those are pretty small revenue opportunities before 2024, but correct me...

Yes, they will start to build up a little bit next year. We won at least one deal there. So it's more of a medium to longer-term play. But I'm just kind of tease it out a little bit what we're going to present at the Analyst Day around some of our opportunities. So it's a little bit of a teaser.

November 29 in New York.

But like -- it's already in my calendar. Looking forward to it. One of the other opportunities -- you guys haven't really brought up too much, but within the car is in vehicle purchases, right, voice authentication? How does that fit in with your plans?

Yes. So we've had a product called Cerence Pay that's had some adoption, not as much adoption as we originally expected. So we've kind of transitioned it a little bit to more application focused, rather than just a generic pay thing. So for example, and we have a customer that's already using this restaurant reservations, right? So there's a royalty, if you will, for a customer using the voice to make a royalty -- to restaurant reservation in this particular solution. And there are several others like fuel purchases and things like that. So it's more -- we're finding more success tying it to a specific application, than just saying, here's this capability to be able to pay for whatever you want and having a whole pack. So it's a bit of a transition, but we think that's going to be good for us moving forward.

I want to ask some financial questions. One about the fixed contract change that the company was speaking about on the last earnings call. Maybe you could share what the feedback from customers has been, since you announced it, right? Because I think that was one of the things that you guys -- you put some thought into it, you think $40 million per year is a number that can allow you to have the market share you want. But now that the more time has passed, what's the feedback been like?

Yes. I think that's a big misconception about the impact of -- no pun intended, but fixing the fixed contracts to a specific dollar level. A lot of this is prepaid licenses. So these are on deals that have already been won, they're shipping there with a small group of our customers, mostly Tier 1s, mostly in Japan and Korea. And what happens is the product starts shipping, mostly the procurement organization comes back and says, I want to buy an inventory of licenses, usually about 6 quarters of license. And for that, I want a discount, but I'll pay you upfront. Discounts have historically been around 15%. Take a draconian situation, where we just cut it off to zero, there's nothing they can do except go back to running royalties. They can't pull our technology out. So it becomes a customer relation composition, right, which is why we kind of pegged it at 40%, where we felt -- and I work closely with the CEO and the Head of Sales, and our Head of Sales has been involved with this business as long as I have been longer. And we felt that the $40 million would allow us to continue to build the strong relationships with these Tier 1s. But to constrain this to a level that -- and the real reason we did this, is because in the few months that I've been the CFO and going and talking to many of you, it became obvious to me that we had made it too complex to model the business. And you can't have a business that our investors and our key investors can't model, because you didn't know the level of fixed contracts we're going to do. You didn't know what the consumption roll-off is. And we were confusing our investors and not getting the opportunity to talk about the things we've talked about the last 15 minutes or so around the strength of the business. Now we will have to go through this transitionary period. It's a 100% gross margin license business. So it impacts gross margin and impacts the EBITDA, it has an impact on cash flow. We tried to provide some additional disclosure with our Q3 earnings. For the first time, we kind of gave this pro forma gross license number. So that would be the revenue, if we had never done a fixed contract. We provided the consumption number. And then we put this little goalpost out that said by 2025, $40 million of fixed contracts should above equal consumption. So then it becomes neutral, right? And I felt it was really important Stefan, Rich, as we talked through this, that we drive through this transition. And then when we get to the other end of it, we show really strong growth in '24, '25 and '26, because a number of factors. You've consumed a lot of the previous ones. But additionally, we've had some very strong bookings over the last 18 to 24 months. And as we said, it takes 12 to 18 months to implement those, 3 of the biggest bookings we've ever had. Those will start to get into production in the back half of '23, which is kind of our Q4, Q1 '24. They come with strong ASPs. They come with strong connected services on top of our embedded. But most importantly, it will allow us to really highlight that, let that flow through the financials, and see the strong outcome of that from a gross margin and EBITDA and a cash flow perspective.

I don't know if you can give any more quantification on how much margin pressure there is in the short term from this decision. But longer term, if these discounts go away, I would think there's a long-term margin uplift.

It's not so much the discounts although it will help on a certain level, not giving those additional discounts. It's really the short-term impact, the 5-quarter impact of losing -- of taking license revenue that's 100% margin. That's the big impact. You can kind of see that through the guidance that we provided in Q4 and the impact of gross margins and EBITDA. There'll still be an impact in FY '23, but it will start to abate. And then as we get into '24, it becomes a much smaller impact and then you add on these other growth initiatives that we're driving. And you should see us return in the '24, '25, '26 to more of the historical financial model that Cerence had, right, growth, really strong gross margins, really strong EBITDA and cash flows.

I want to ask on the Services business. It's an area you had very strong growth in recently. I think it was up 36% year-over-year last quarter, if I'm remembering that correctly. How sustainable is that? Is there any sort of onetime nature to that, or is this more of a sustainable level?

Well, our strategic objective is not to become a PS company. So please don't model that into the future. The reason it's so strong this year is, we're starting to implement these strong bookings, and in that implementation period, we're invoicing and we're doing NRE with our customers, and that drives without PS revenue. And we also talk about SOPs on a quarterly basis, and we had 43 of them last quarter. That's the start of production. So that's -- the PS work is done, gets into production and then we start to get the revenue curves. So it's up because we've had a large number of SOPs that we've been working on this year. It will -- it's still a very important element of our business, because it is what helps us to drive that differentiated solution that we're building for our customers. But our objective is not to make it a growth driver. I mean ideally, we want our solutions to be easily implemented. That requires work with -- not just by us, but by the Tier 1s and the OEMs. We'd really love to shorten that cycle time between bookings and revenue. And the more successful we are at that over the medium and long term, that will actually drive PS revenues flat or down a little bit in the medium and longer term, right, which is actually a good thing, because you're getting into production faster and you're getting the real revenue that we want, which is the license revenue and the connected services revenue, right? So it's kind of a catch 22, right? It's nice that the PS revenue is growing, but that's not what you really want from a long-term model standpoint.

Understood. Let me stop here. I do want to give the audience a chance to ask a couple of questions if they have any.

Hi. I just wanted to ask about your percentage of production which has Cerence tech on it, that stat that you produce, which seems to have just stalled in the low 50s. And given all the kind of aspects we're talking about, the electric vehicles, upselling of technology, which is then luxury vehicles coming down the spectrum. Just why has it stalled? Why is it not increasing? What's going on there?

Well first of all, it's a really big number. So it takes a lot to even move it by a percentage point. A lot of the growth in auto, if you look at IHS, it's in some emerging markets where some of the technology is not even deployed. And then, of course, it all depends on within a particular quarter, what the mix of models that each OEM is choosing. And right now, we don't have access to the choices they're making on how they're deploying the limited number of chips they have. So we're subject to overall volume. We're subject to geographic mix. and then we're subject to how they deploy that. But what I think is really important is that we'll see over time, higher consumer -- connected services penetration. That's really not going to impact that 50% because it's already part of our embedded solutions, and we're going to see the higher ASP. I think those are the 2 bigger numbers to try to pay attention to, as you look at us as a company, because that's where we're really driving our competitive advantage. That's where we're really capturing value for the solutions that we're providing. Yes, we haven't broken that out. I'll take a look at it and see if that's something we can maybe give some visibility to.

You guys mentioned that you can coexist with the big tech companies. Is there a content trade-off, though, where if you're one of multiple parts of the overall vehicle voice opportunity that -- is there a lower ASP, if there's Google there, as well as you guys?

Not really because we're -- if we're in there, we're defining the core stack, right? And we're just allowing coexistence with the tech companies, right? I mean, you're not going to get away from not having an Apple Car Play button on the head unit. But that doesn't really impact the technology stack that we're selling to the OEM, right? And they really don't, at this point, who knows what they'll do 5, 10 years from now, but they're really not mixing the stack, right? They're either using us -- or kind of Google GAS is really the only other option. And we've announced win backs from tech companies, too. So I don't think that's a big piece of the equation right now, and we don't see it in the short term, medium term.

One more question to close this out, and maybe I'll ask both of you. What do you think is something that could surprise investors or industry observers over the next year?

Well, hopefully, production starts to come back. It is kind of one of the big drivers of license revenue. I think for us, as a company, we have to get through this Investor Day. We have to lay out the specifics around many of the things that we had put the numbers behind it. A little handcuffed as to exactly what I can say right now. And we just have to go execute around what we're saying, what we're going to do. And hopefully, that won't be a surprise to investors. It will be a reassurance of what we're trying to accomplish.

Yes. I mean I think we know we have work to do with investors to -- and we're focused on executing. We're focused on delivering a strategy that's executable, that is one that we think is going to set the company on a path of good growth into the future and get back to where we were before. And we do have this transition period, but we're pretty confident that once we get through that, we will be in a good spot.

Great. Well, unfortunately, we're out of time. Tom, Rich, appreciate you taking the time to speak with us.

Thank you.

Thank you all for listening in. Appreciate it.