CCC Intelligent Solutions Holdings Inc. (CCC) Earnings Call Transcript & Summary

Good morning. Thank you for joining us at the CCC Intelligent Solutions session at the Goldman Sachs Communacopia & Technology Conference. I'm Gabriela Borges. I cover emerging pathway here at Goldman. My colleague on the far right, Callie Valenti and we're delighted to have the CCC team on stage with us, Githesh Ramamurthy, CEO; and Brian Herb, CFO.

So Githesh, one of the consistent learnings over the past year has been in order to have an AI strategy, you need to have a data strategy. And a more proprietary data set, the more interesting things you can do with your AI road map. So let's start with the data set. What do you view as the most differentiated about the data set that CCC has?

Sure. Interestingly enough, the data set has been an essential part of our strategy for decades, because there's a massive amount of data that's involved in the process of settling an auto claim. If you think about even a single auto estimate or repair, a collision repair, when you go to an auto estimate, you're literally talking about 700 different data elements for that particular claim. So over time, what we've done is that because of a variety of other analytics that we've been doing over the years. We've actually built very large data lakes and data warehouses over time with very granular level of detail. And that granularity is the key difference between how you can use it for AI. So that level of granularity getting down to parts level detail getting to the specifics. So we have about $1 trillion of historical data. We collect 500-plus million photos a year. And that has been enormously valuable and we started early on before AI, we started building predictive models with it. And the AI itself, we've been at it for about a decade now.

It leads to the question on what comes next. And so I think AI has been, you've mentioned, very consistent part of your strategy. What are the 1 or 2 AI priorities on your road map for the next 10 years?

Sure. So what's been awesome has been that we started working on actually one of the hardest problems in this industry for AI, which is I look at a photo of a car, that's been an accident. Each crash is unique. Cars are the same, but each crash is unique. So now how do I take a 2-dimensional picture, convert it into a 3-dimensional image of the vehicle and the parts that are needed, the labor that's needed, the overlap that's needed and say, I can give you a very accurate prepared prediction. That gave us enormous credibility. Now that took us 7, 8 years to build. And then by '21, when we went to market with it before all the latest noise around ChatGPT and everything else, that give us enormous credibility on solving a really hard problem. And so we have a very large adoption across our customer set. And what we are now seeing is that AI is now an embedded part of almost every solution that we built, not just for Estimate-STP, but straight through processing for the industry as a whole.

Let's stay on the topic of straight-through processing. So at CCC Next, there was a really interesting conversation around intelligence straight-through processing, the platform for insurers. So walk us through. We can see a little bit of the Estimate-STP value proposition today. There is a whole ecosystem around this and a longer term road map to build it into a broader straight-through processing strategy. How do we get from [indiscernible]?

So when you look at right from the point at which an accident takes place, you could have telematics data coming out of a vehicle. We're fully capable of processing telematics data and deriving intelligence from the telematics data. We're also able to impute and compute the physics of the accident and understand the implications on bodily injury from that accident. We're also able to help the consumer decide, what should I do with this? Where should I repair this car? We're able to help the carrier make decisions. We're able to help the repair facility decide what parts to buy. So at every step of the way you can see AI being applied intelligently, so people are only managing the exceptions. One of the things we've seen is that there's about $1 billion -- 1 billion days -- excuse me, not dollars, 1 billion days of elapsed time in auto claims in this country. And that has now gone to 2 billion days. So our customers whether insurance companies, repair facilities, OEMs, parts providers, almost every customer we talk to is coming to us and saying, is there a more intelligent way that I can manage exceptions, because I don't have the labor. I need to become more efficient. And that's why intelligent straight-through processing or ISTP is extraordinarily important for this industry.

One of the dichotomies that I often find listening to CCC talk about the technology platform. Is there's so much sophistication to it? And when you think about the level of technology adoption in the auto insurance industry and it feels kind of painful from the outside in. And so how do you reconcile those 2 things? You have incredibly sophisticated technology and you have a customer base that historically hasn't been the most willing to lean into those types of technology catalysts.

Yes. I would say that fundamentally, the technology has to be seamless and easy to use. In fact, one of our conferences, we had this team, which is an Arthur C. Clark quote, which is great technology is indistinguishable from magic, right? That means it's got to be seamless. It's got to be easy to use. And that is because our customers are not in the technology business, nor should they be in the technology business. Cars are hard enough to repair with sensors, cameras, calibration, all of this. So there's an enormous amount of complexity and our job is to make -- do all the plumbing, do all the hard work and deliver something that's easy and turnkey. And one of the true notes for us as a company are, are we actually doing that is a measurement of Net Promoter Score. So we measure Net Promoter Score. And as you know, enterprise software Net Promoter Score is in the low 20s. Our Net Promoter Score is 82. That means it passes this lens of is it easy to use, have you integrated all of these things, is it simple, is it easy. But we're also seeing across the board, our customers now saying, we need to invest in artificial intelligence. We need to invest in the platform and CCC has been a trusted platform for decades, and that's really helping.

I want to ask a follow-up here, which is, is CCC and yourself and your sales force at or in a position where you can have that AI strategic conversation at the highest level of the organization?

Yes. In fact, I can tell you the dinners I've had with the senior most levels of our customers. across the board, they see the transformative power of even Estimate-STP, which is a small component of what we eventually would like to get done. So this is an example where a customer said, we took a claim. The customer took the pictures, it took 7 minutes for the customer to take the pictures and send us the picture. By the time it went through your AI by the time we got to photos, by the time we made the payment in 10 minutes, which sent the payment directly into the customer's account of which 7 minutes of the 10 minutes was a customer taking the pictures, right? . So here's a great example of a truly transformative example and customers are truly intrigued by what this could mean for a lot of other things. So the short answer, I don't know the long answer to your question is yes.

Are you at the point where you can come away from that and call the account executive on that account and say, because of this conversation we now have a plan to be able to solve products, ABC over the next 12 months, which will increase the account value of CCC at this account by 1.5 or 1.2?

I typically don't do that. The reason we don't -- I don't have to do that is we are in constant engagement across our teams, our product teams, our account teams about the road map for our customers. What is it 3 months out? What is it 6 months out? In some instances, we're working 2 years out. In some instances, we are working 3 years out. So it's actually a part of the biorhythm of the company where that is how we operate. That's how we work. So there's full engagement, a team that might be building a component of AI for subrogation. They're already working with 20 different customers on implementing, finding the nuances, putting these things in place. But I don't have to come back and make those calls.

Very fair. Maybe I frame it as does the road map conversation over the next 3 years lead you and Brian to say there is a real potential for acceleration here to our revenue?

I will turn it over to Brian for that one.

Absolutely. I mean one of the things that we do and get a lot of value is we have ongoing innovation discussions with our clients. So we break out and we have 6 or 7 separate sessions around talking about APD and straight-through processing or with the repair facilities or with casualty or subrogation, we bring our largest clients together and talk to them about what are their operational challenges. We talk to them about what's on our road map. And we really get a clear understanding of where we're going and the needs they have. And so when you look at our road map, and we do this through our strategic review, each of the initiatives have clients that are already engaged in those discussions. So we're not out there or in a vacuum building a solution and saying, hopefully, this will take, and we'll take it to market and see how it goes. The time we bring a solution out, we have clarity of who are the early adopters, an Estimate-STP is a great example. I mean, Githesh talked about the early traction we had. When we brought that solution to the market, we had USAA as a flagship client to be the first on it. There's an article in the Wall Street Journal around them launching that within their organization. And that was already known as we brought that product to market. So that makes it a lot easier as a finance person to say, okay, let's put investment around this because we already know the demand is out there for a handful of clients. But then we also have viewed that once we help them there's plenty more to come because those solutions are working against their true operational challenges that they're dealing with.

So switching really quickly to demand. It feels like the insurance vertical has held up a lot better than many other of these niche verticals that companies sell into. Can you compare and contrast kind of what you're seeing this year as you plan for 2024 versus what you saw last year as you were planning for 2023?

I wouldn't say we're seeing anything dramatically different, because if you look at the underlying structure of the industry itself, it's a very durable industry. Auto insurance is mandatory. Complexity is increasing enormously. You look at the number of new car models that we delivered this year. The number of sensors, cameras. We met with one manufacturer, who wants to make sure that cars can be repaired in collisions. When this car comes out, they're actually going to have an LCD screen on the front bumper. So there's a very steady amount of complexity that's increasing. And so we don't see substantial changes, but we are seeing a material acceleration of our customers' desire to digitize far more than anything I've seen in the last 20 years.

Yes. That makes sense. And then I kind of want to switch back to the products. I think some of the interesting part of this. You talked a little bit about this before, but even new computer vision, AI technology in the casualty space, can you talk about the development process for that? And then just any initial customer feedback?

Sure. So one of the things that we've done over the years has been that roughly out of one accident out of every 5 accidents has -- is a casualty claim. What has been very difficult sometimes is when you get a casualty claim, it might be months after the actual auto accident has taken place. What we wanted to do was to link these 2 things together. So we actually had built a lot of the models that can take the physics of the accident, and I won't take us to high school physics, but the physics of the accident to really compute the impact to the human body and what it could do. And with the work we've already done in photo AI and generative being able to understand accident, we were now able to project for every auto accident, whether it has a medical claim or not, what is the likelihood of injury, what should be done. So we started to put that out with customers, with some early customers, and they were like, "I need to use this on every claim." And so the customers are actually saying this is fantastic. This is very good. So the early feedback has been very positive.

The other interesting thing with that is we can sell that to clients that aren't using us for our casualty platform. So it becomes a bridge from APD to casualty, which is really important as well.

Yes. And then for Estimate-STP, which you talked about a little earlier, how are you working on expanding kind of the use cases of that product?

We've done 2 specific things to expand the use cases. So initially, what happened is, as a lot of customers, who are using our mobile channel, mobile capabilities for consumer self-service where the consumer goes around the car, and our AI guides you to take the pictures, you send the pictures back. That happens roughly 30% of the time. 45% of the time, the collision repair actually sees the car for the first time and writes the estimate. And another 35% of -- the rest of the time, the actual -- the staff adjuster would come up. So we've now provided the same AI capabilities to the staff adjuster so that the AI can actually jump start and help write a lot of the grinding details. So the adjuster still has the expertise and makes the final decisions. But for the collision repair, we've seen -- just in the last few weeks, have seen some great examples where the photo AI allows a consumer to send the photos to the repair facility and the AI actually pre-populates and says, this could be between $1,800 and $1,900, and by the way, here's the line level detail. That is being received very well. So we've expanded the use cases for that as well.

Makes sense. And then payments is kind of an exciting longer-term opportunity for you guys. You've talked about having kind of a base product out there with early customers. Any update on that? And then, Brian, how do we think about the potential impact on the model from a revenue and as well as gross margin standpoint?

Do you want to take the customer. I can do the model.

Sure. We are working with early customers on payments. It continues to be -- there's still a lot of manual steps in the business, both in terms of -- across the board, and we're seeing many different use cases. and we're continuing to work with some early customers to build out the solution, and we feel very good about the need to actually have a much more sophisticated and seamless payment solution as part of digitizing and as part of what you were referring to earlier with straight-through processing, that's going to be an element that's going to have to be there.

Yes. And to think about the revenue model. So across our platform, about $100 billion is flowing through each year. Think about carriers paying medical providers, carriers paying repair facilities, repair facilities paying part suppliers. So all that commerce is being driven off the platform. So we look at that as the opportunity. So there's a market share within that $100 billion. And then we think about a take rate of about 25 basis points. It will be different depending on the use case that we're facilitating and it will come, there'll be a bit of subscription. There'll be some transaction. There'll also be some virtual card that will play into it. And that's why we look at it as an effective 25 basis points for modeling purposes.

Right. Yes, that's helpful. And then another [indiscernible] just going to product or products, but on diagnostics. You talked about kind of 50% of estimates today having pre and post scans for diagnostics and how kind of that number should be 100% longer term. How are you able to kind of drive that number up to the 100%? And how is that important for kind of the entire ecosystem?

Yes. Again, what we're seeing is that to actually repair these cars properly, there's a natural adoption curve, right? When you go back a few years, 3% of cars are being scanned. What our solution is doing is very specifically providing transparency of a trend that is already happening. As you have more sensors, cameras and parts, the diagnostics and calibration capabilities are extraordinarily important. And we have worked with insurers, car companies and the collision repairers to provide very seamless transparency of what is actually happening with diagnostics. Did I do the scan? Was the scan actually completed? What are the implications of the scan? So we are working with all 3 parties and that has actually been -- people have been extraordinarily receptive, because of the transparency that we provide. So our job isn't to really drive up scan or drive down -- that's not really our -- we are helping to make sure we provide the transparency of this trend that is inevitable.

And then so for Subrogation, another big product you guys talk about another growth driver. You talked about replacing a lot of very manual processes today, and that's kind of a pretty advanced technology. There's AI involved in that. How does it fit into the overall STP ecosystem? And how do customers think about the ROI versus the manual processes they're doing today?

What we're seeing is that with Subrogation the individual demands on a subro adjuster are massive because 1 person may have 150 cases of subrogation they're dealing with and each of these could have literally 150, 250 pages of documents. What we've been able to do with AI is to ferret through all of that information and extract the most critical information necessary and also many of the customers that are doing subrogation, most of the customers are customers we've been working with for a very long time. . So the ability to connect from A -- point A to point B across customers is really important. The technology has also been extraordinarily important since we've actually working with well over a dozen companies at this stage and the receptivity to really replacing -- adding this technology to really empower the end users has been very well received, because the ROI is very, very, very big.

Makes sense. And then last one for me, and I'll turn it back to Gabriela. What functionality are you selling into parts providers today and also auto manufacturers? Kind of how does that complement the rest of your ecosystem?

Yes. So if you look at, again, what we are providing is really a seamless ability to order all the necessary parts for repair. Wait times on parts have been increasing. So the ability to pick and choose the right parts at the right time for the right car in the right geography becomes extremely important. So because we're connected to thousands of parts providers in real time, our repair facility customers have the ability and they make the decisions to choose, which parts providers and how this works. For the parts providers, if you saw the recent announcement we did with Toyota, for example, Lexus, for a manufacturer, this provides a seamless ability for them to integrate help their dealers, deliver and sell parts to an ecosystem of 29,000 CCC customers, who are in the business of ordering billions of dollars of parts every year. And then -- so those are really the 2 -- so for the OEM, there's a lot of benefits. For the repair facility, a lot benefits. And then for the independent dealers, who then work with the OEMs and the repair facilities, there's a lot of benefits. And for the insurance carrier, there is transparency in the process of the right parts being -- the right part policies being adopted for the repairs.

It's interesting hearing you talk through the product set, because there are so many modules and solutions that you offer that are broad to customers. And it leads to the question, both from the investor perspective looking in and for your sales training and enablement, you mentioned earlier how the amount of alignment that you have between your CCC folks and the customer is incredibly close. How do you think about the 1 or 2 products that can move the needle? Or perhaps that's the wrong question to ask. And how do you think holistically about executing on a cross-sell?

This is not a new muscle for us, right? If you look at the history of the company, I can look at a time where almost everything we deliver today was at 0 in revenue, right? So every one of these things has grown, because we've designed the right products, we've designed the right solutions and working with customers. So what we are seeing, one thing that's different today than at any other time in the history is that the breadth of what we have to offer is the widest it's ever been. We have solutions for insurers for every facet right from first notice of loss to subro all the way through the pieces. We have solutions for collision repairers from the smallest collision repair to the largest collision repairs, we're bringing banks online to help bring to connect insurers and the banks for title processing. So I would say there's a -- so it's a very broad suite and we've connected all the different parties in the ecosystem. So it's a muscle that every year, we're learning, for example, when we did subrogation, we realized we need to bring some expertise very specific to subrogation. When we did payments, we had to bring very specific expertise on payments. So we try to bring world-class talent into every single area, but we also have to tie this together holistically with the customer.

Absolutely. Going to pause and go to the audience. Questions from the audience. Brian, one of my favorite things to ask you about is the potential for margin model to continue to go up and to the right. And we've talked about and we see this in information services companies where comeback full circle to the proprietary data set that you have. Where in many ways, you have the best of a business model of an info services company, coupled with the best of a business model as a software company. And so how do you think about, I guess, pricing to begin with and extracting and bundling of solutions in a way that is a win for the customer and a win for CCC? And two, where do you think your medium-term EBITDA target of mid-40s -- talked about how there's actually not a feeling on that number? So how does that evolve over the long term?

Yes, absolutely. So starting on the pricing point, and then we'll go to the margin point. So I think one of the things that we do well is really take a ROI approach to pricing. So the way we take our products to the clients, look at the engagement we talk about with the carriers. We price our products on not on a contingency basis, but we set the pricing on an ROI basis. So some of our solutions would be 3:1, some of them are 5:1, some are 7:1. On average, we look at driving a 5:1 return. So we're showing tangible value to the clients and therefore, we're getting our share within that ratio. So that pricing proposition, I wouldn't say, it makes it easy for the sales organization, but you're having a very constructive discussion around the value and the efficiency we're driving to the carriers and then how our fee structure works relative to that. As far as the margin question goes, we've seen strong margin improvements over the last 4 years, we went from 30% margins all the way up to 39% margin. So at the end of last year, our margins were 39%. As we go forward, we're suggesting that we're going to move into the mid-40s over a 5- to 7-year time horizon. Certainly, we've shown the strength of the operating leverage that's in the business, showing those 900 basis points over the 4 years. We remain very focused as a management team on balancing, getting innovation into the business to fuel the future growth and driving that innovation to the market at the same time seeing the margin progress. And so that balance is really a key for us. We can push margin up higher if we want it, but we're really around the long term, making sure we're putting the right investment in the business and delivering the right innovation into the market.

What determines whether a product falls into the 3:1 category versus the 7:1 category?

I don't know if you want to cover that? Or...

Yes. I think what it's not any hard metric. It just -- it's a range we've seen over time. Some very easy, very clear value propositions in terms of efficiency, cycle time, et cetera. And so it's a range. Don't take it as a hard. It's not a hard and fast rule. It just turns out to empirically work out in that range.

When it comes to the growth algorithm, I think you've been pretty consistent in talking about a large percentage of growth coming from cross-sell, because of the customer wins that you already have and your flagship positions at 38 of the top 40. For the first time in the last 3 or 4 quarters, you've been breaking out the emerging product contribution. Should we be reading into that the percentage of the portfolio is shifting more towards products that are emerging that can potentially drive that contribution from emerging products bigger over time?

Yes, that's right. I mean the way we set up the long-term target is we're seeing 7% to 10% growth over time. That's an organic figure. We then break it down and say, out of that 7% to 10%, 20% will come from new logos. And then the 80% will come from cross-sell and upsell. And then out of that cross-sell and upsell category, we're saying half of it will come from established products. So those are products we've had in the market for a while, things like casualty parts upgrades to the repair shop packages. And then the other half will come from these new emerging solutions. So that is Estimate-STP, that's Diagnostics, it's Subrogation and its payments. And so to your point, that 50% of the cross-sell, upsell is coming from these new cohort of products that we more recently have released into the market. What we just talked about last quarter was today, that's 1 point of growth. But that 1 point of growth will move to 3 or 4 points of growth as we go forward. And that -- we see the early success of those products coming into market, we're seeing the adoption curves that giving us confidence on that. And we also believe that just where we sit with these solutions and the trends in the market that they're going to be material contributors to the business over time.

Is the 1 to 2 points going into 3 to 4 points additive to the growth algorithm or substitutive?

Yes. We look at it as -- it's still working towards a 7% to 10%. So some of the more established products will start to taper off because we've gotten further penetration in those and then the emerging solutions come in and drive a larger part of the growth going forward.

What is the limiting factor to taking you a data set? And making it useful to international customers and leaning into new [indiscernible]?

We clearly look at adjacencies, and we've been invited, as you know, there's only one other geography we're in. And we've been approached by many customers in other geographies. We have not to date prioritized that in a very big way, but much of what we do in artificial intelligence, much of many of the technologies we have, the tech stack we have are not only applicable in auto claims, but in other lines of property and other claims as well as international. And we see so many opportunities in front of us right now across the board with all of the solutions we're building. That is something that will naturally take the course of time, but it's not immediate for us right now. But it is a very large -- the auto insurance market worldwide alone is a very large market. And when you go from auto insurance to P&C, it is even way larger than that. So.

Absolutely. Good stuff. Let's leave it there. Thank you both for your time.

Thank you very much for having us.

All right. Thanks, everybody. Gabriela, thank you so much.