Good afternoon, and welcome back to day two of the 43rd Annual J.P. Morgan Healthcare Conference. My name is Ta-Von Wilson. I'm an associate within a healthcare group based in New York. Today, we're here to welcome Andrew Toy, who is the CEO of Clover Health. Andrew is going to provide a presentation on the company, and then followed by, we'll do some Q&A. So with that, let me welcome Andrew to the stage.
All right, thank you so much. I'm delighted to be back here at the J.P. Morgan Conference. As was stated, I'm Andrew Toy. I'm the Chief Executive Officer of Clover Health. And we'll be making some forward-looking statements today. The usual disclaimers apply here. So we were here last year, and we were having a discussion about Clover, what we were intending to focus on, what we were intending to achieve. And so this was a slide that we presented last year, I think maybe in the same room even. The goals for 2024: profitability, focus on star rating, a return to growth, profitable growth, and an eventual expansion of our offerings. Meaning, what does it mean when we have such a strong technology platform that can complement not just our plan, but perhaps others' plans as well?
So looking back on how we did on that, Adjusted EBITDA profitability, we have guided to material Adjusted EBITDA profitability for 2024, and we feel good about that. That's a significant achievement. That has been a focus of the company to get to that profitability. We committed that that's what we were going to focus on. We committed that we were going to do that without raising capital, and we didn't raise any capital. And we achieved that profitability goal. And I'm very, very pleased about that. We doubled up over here. We were focused on Star Ratings. And not only did we return to a three-and-a-half-star rating for this plan year, but we actually achieved four stars for 2026, payment year 2026.
Very pleased that in an environment with challenging Star Ratings, that we are one of the few plans who have received double upgrades on our Star Ratings. And there have been a number of drivers around there, but we executed really well, and I'm pleased with that as well. Growth. We said that once we achieve profitability, we would then focus once again on sustainable, smart growth. And we announced at this conference that our January-to-January growth came in at 27%, which we're once again very pleased by, attaining right around 100,000 members. So that is a strong season of growth. The AEP was good for us. We had strong benefits. We were going into higher Star Ratings. Our loss ratios in 2024 were very strong. So that enabled us to maintain and invest in plan benefits, offer great options to eligibles in our markets, and have that growth.
We launched our Counterpart entity last year as well. We didn't have that at the beginning of last year. Counterpart is a subsidiary of the Clover parent and houses our R&D and technology function. It makes our assistant technology, branded as Clover Assistant within our plans, branded as Counterpart Assistant in an outward-facing third-party environment. It makes that same exact platform technology available to other Medicare Advantage risk-bearing entities, whether they be plans or providers. We make our technology available to help others with total cost of care, help others with quality. Across all of these, we've achieved, I think, tremendous results. We got to profitability. That's been achieved. We are going to be growing and maintaining profitability. That's a fantastic place to be. We are investing in our technology and making it available to third parties via Counterpart.
And that double star rating increase, paying paid on three-and-a-half stars for year 2025, going into a paid-on-year, paid-on-four stars for 2026. Very, very happy with all of those results. So I've talked about this. Our next phase, this is the way I think about it. We are a technology-centric company. We build our own technology for clinicians. We're very clinically oriented. We're very PCP-oriented. We are technology-oriented. We build a technology product, the assistant, that is used by lots and lots of different physicians, not just our own employed physicians, lots of wide-network physicians as well. And I'll talk more about that. And that innovation was what we were focused on in an early period, 2018 to maybe 2022, right? Building that technology, proving it out, building the data platform, applying machine learning, AI, bringing it together, making sure that it worked the way we liked.
That was our first phase. Then we added to that. We maintained that investment in R&D, in our product-centric view of the world. And we built that physician-facing product and used it to help us to get to profitability. That was the next phase. From 2022, 2023, 2024, we were focused on profitability. We talked about that all the time, that that was what we wanted to do, get to Adjusted EBITDA profitability. And we've achieved material profitability, as I said. So that phase has now that focus. Of course, we want to maintain profitability. But what's the next thing? Doing a third thing, which is growing as well.
Adding to our membership within our plans, adding to the total number of patients being managed by the assistant platform through the counterpart entity as well, growing that total number, which should result in revenue growth, eventual Adjusted EBITDA growth, certainly total lives under management growth. All those three things together, that innovation on our technology platform, the profitability to allow us to be sustainable, and now growth as well. We will be delivering all three things, and that is a very, very exciting place to be, so what is our vision? Our vision is to empower every physician with technology to identify, manage, and treat chronic diseases earlier, so if you want to take away something about Clover, I talked about this last year, and we maintained this as our vision, management of chronic disease by doing things earlier. Earlier is key.
By doing things earlier, you might actually end up doing less in the long run. And that's good. Less treatment in the long run is a good thing for patients. It's a good thing for total cost of care because you do things earlier. You flatten the curve. We identify earlier. We manage and treat earlier. And it's a focus on chronic disease. We are Medicare specialists. We do not have a commercial book. We do not have an ACA book. And in Medicare, which services seniors and the disabled, chronic disease management is critical. So it pairs perfectly well with our vision. So when you think about what our technology platform is doing, it's this. We want to empower every physician with technology. And when I say every physician, I don't mean our own physicians using an EHR or EHR add-ons.
Our own physicians, which we do have a home care practice, which I'll talk about, certainly use our own technology. But the vast majority of physicians using our technology today are not employed by us. They are wide-network PCPs servicing our members, but using our technology in conjunction with their EHRs. We're not an EHR company. And why? Earlier diagnosis and treatment, earlier disease management, leading to higher quality and clinical care, and affordable and accessible care. So that's our focus. Earlier care, how are we achieving that with technology? Who's using the technology? Wide-network physicians. Between that, you can see what makes us different. There are a lot of groups out there, verticalized managed care groups, who empower their own physicians with technology. We build a platform that can be used by every physician because that is part of our vision. So that's what makes us different, right?
Because we want to get to accessible and affordable healthcare for all. We're focused on Medicare right now. And on the left-hand side here, you can see that we are focused on investing on technology-centric approaches to care management. And when I say technology-centric, we really mean technology-centric. We, in our own plans, do not do value-based contracting. The vast majority of our, in fact, almost 100% of our Medicare flow through fee-for-service. We don't do value-based contracts downstream. What we are doing is offering our technology to physicians. And when we arm those physicians with AI, with data, we see improvements in quality. We see improvements in total cost of care without having to move them to value-based contracts. That is a very unusual, I would assert, approach. Most people would say, well, you need the value-based contract because why would they do it?
What's happening is we are paying them fairly for their time. We are paying them fairly for the time they are using our technology platform, and we are then making sure that our product does what it's meant to do, making it easy for them to deliver better care, not giving them incentives that might create moral hazards, not giving them incentives that are confusing, nice, straightforward compensation for time, and then arming them with data and tools that let them deliver the earlier care that is critical to the right-hand side, chronic disease management. We also combine this with a focus on managing the sickest population through our own employed physicians, and I'll talk about that in a moment.
But the takeaway from both of these sides here is everything we do, everything we do is about that earlier chronic disease management through our technology and on a fee-for-service chassis, right? Like we've all talked about value-based care at this conference that was discussed, and value-based care is great. Value-based outcomes are great. Combining that with how it actually lands in the real world is challenging, especially when you combine it with technology and I think we found a really good solution for this, so we are really focused on using all that data and AI, as I said, to bring all of these capabilities to primary care physicians and when I say primary care physicians, I mean any physician who sees themselves in that holistic quarterbacking overall care management function. It could normally be an internist. It could be family medicine.
It could be a cardiologist who's taken over full care for a patient. So that's what I mean when I say PCP here, by the way. Our tool is very good in all those scenarios. So Clover Assistant synthesizes many, many data sources, 100-plus data sources. The thing that I like to emphasize is, while it does integrate with EHRs, it can be used outside of an EHR. The reason to integrate it to an EHR is actually for workflow purposes, not for data purposes. So what we can do is, if a physician wants to start using Clover Assistant today, we can turn it on today without having to do a single bit of EHR integration. That's pretty interesting, right? We are not reliant on what's in the EHR. I find that people find it unusual.
But I don't think it's unusual to me because a key part of what makes our technology interesting is we are trying to show physicians things that they don't already know, right? Because if you want to change how care is delivered, you want to show physicians insights they don't already know. Well, logically, that makes sense then. We are less dependent on their own EHR data because that generally are the things that they know, right? So that isn't the richest source of insight. You can get some insight with AI, et cetera. But what we're able to do is the data platform that we have built is stretching across many, many, many data sources generally that physicians do not have access to within their EHR. The classic example I give to set all of your intuition is pharmacy data. Physicians know that they prescribed a medication.
They know where they prescribed it to. That's generally built within the EHR. They have no idea whether that was filled. They have no idea whether that was picked up. We do know that. We can share that back with the physician. Trivial, simple example, but something where you're like, how come they don't know everything? That's an example of something that a physician doesn't know. Very easy for us to bring that back using our data to give them insights based upon that. What we do is a much more deep synthesis than that. But the insight layer can be as simple as that. And that's that middle area, the 100-plus proprietary models that make up our IP portfolio that our R&D is pointing to. We have teams working on getting more data sources. We have teams working on building more models around it.
And then overlaying all of that is using AI to figure out how do we make this clinically intuitive to physicians. I'm often asked, how do you change the behavior of physicians? How do you make them do what you want them to do? And then I immediately say, I think that framing is wrong. That is wrong because it has the words, how do you make them do what you want them to do? You should not be making clinicians do anything, right? That their job as physicians is to practice medicine, and we trust them to practice medicine. What we're doing with the assistant is helping them do a better job by arming them with information that they can consume quickly because their time is under a lot of pressure, as we know, to improve and identify and manage chronic disease earlier.
We are not making them do anything. We are not even tying their compensation to the value-based outcomes. That is how far it goes. We are using our engine. And just as important as how we pull together these data sources is how do we present it to the physician so that it is clinically intuitive to them on what they should be doing. And we're very, very good at that. They should be using their clinical judgment. Bringing this together into our overall care model. I mentioned before that we are very good at fee-for-service physicians. 95% of our membership sits within a PPO product. So that means we are dealing with a wide network open access system. This is how we solve that. We aim for the majority of our membership to go see a Clover Assistant-powered network PCP. What that means is there's no steerage.
You should go see the PCP that you know and trust when you join Clover, and we want that PCP to use the assistant so that they can be armed with our insights and do a great job managing your health, right, so the majority of folks on that green bar, lower to medium to maybe even medium-high acuity is being managed by those PCPs. Also using Clover Assistant, we have one set of employed physicians within our Clover Care Services Clover Home Care Group, and what they do is these are employed physicians, and we are bringing PCP primary care into the home, and we take over the primary care in the home in this scenario, and this is for the sickest population, the most comorbid population.
We are looking at them and saying, we aim to identify folks who need this care, where care is most appropriately now delivered in the home-based setting, and our own MDs will take over with the partnership with their existing PCP that home-based primary care. Almost no PCP in our wide network does home visits anymore. And so we say to them, now at this point, it really seems that this person would benefit from home-based primary care. Why don't we take over? We'll share the data back to you via Clover Assistant. We will make sure you're involved, but we're going to take the primary view here and lead from that home-based setting. And that is a huge important part of our care model as well.
So that home care function has a set of clinicians who are focused on identifying even those who could benefit from this, triaging them, and then referring into the home-based primary care practice, all using Clover Assistant, of course. Between those two things, our wide network using CA, as well as our in-home care function using Clover Assistant, we've had a number of great outcomes. Obviously, being profitable, we're happy with where total cost of care and our loss ratios are sitting. And we talk about that during our earnings. But we're also proud that our PPO, of plans of reasonable size, about 2,000 and above, we are number one in the country in terms of HEDIS scores, HEDIS being the clinical quality scores that sit within the Star measures. We are number one of plans of reasonable size, right? And that is a PPO is number one.
Our HMO is number two, and then after that is quite a lot of other HMOs. We are incredibly proud that we can deliver that on a wide network open access chassis, and that is driven by the fact that we can give our technology to almost any physician, right? I said before, empower every physician, any physician. That is really useful within the PPO setting, and hopefully, you can all see why, and that is resulting in incredibly high quality performance from a HEDIS perspective, so we're very, very excited by this, and this is a big driver of how we move towards that four-star rating for payment year 26. On the right-hand side, you can see here that of our public peers, and we just used the publicly traded peers here as a selection set, the lives-weighted average is significantly lower than what we achieved.
And once again, I would assert that that is because it is difficult to manage this number of lives on a wider set of physicians when you lack a care platform like something that we built with Clover Assistant. We are a leader in readmission prevention measures. So this is, again, a breakout of some of the data that led to the Star measures. And these are the readmission measures. And you can see here that we were sitting at five stars for these readmission measures. You can see here that we're significantly better. We put on the bottom here to make it easier about how much better we were than the national median on those measures. And this is where it starts to make sense where we bring together that Clover home care service together with the wide network.
Because readmission is something where the primary care physician should be involved, but where our in-home care team also leans in heavily. Because an acute event has occurred, there needs to be a greater intensity of care. So no matter where you are on the spectrum after you get admitted, our Clover home care team will get involved post-discharge to help manage that readmission risk in partnership with the PCP. But the PCPs are less equipped to be able to do that. They need to know what's happening. They know the patient really well. But it takes a whole team to do this. And this is where Clover Assistant's powering a lot of the care across the board. But that focus on delivering the care to the right location, the home, really, really helps with readmissions. And I think we're seeing those results flow through here into our Star measures.
So this is a great example that I like to point to where data is important because you need to know when they're admitted, when they're discharged, why, what the meds are. But you also need that care delivery arm to bring it into the right setting because without that care delivery arm, all the data in the world is not going to be able to drive the action and result that you want. And this is where that full care model comes together. And we get a fantastic result for our members as well as for the business. So I really think that coming out of 2024, our technology-driven approach is working.
When we have strong Clover Assistant coverage, and we have strong Clover Assistant coverage now, and I want it to be even stronger, but it's strong, we're able to then drive those better AI-driven clinical recommendations into multiple physicians, right? Our own employed, like I just said, but most importantly, into that wide network. Once we drive in those better AI-driven clinical recommendations off our data platform, we're improving the health outcomes and quality of life. We've shared papers about how we're identifying and managing diabetes earlier, identifying and managing CKD earlier. And that's not just our own clinicians being a broken record. That is the wide network delivering value-based outcomes better than many value-based outcomes using technology. So once we do that, we reduce the total cost of care, right? We flatten that curve. We find things earlier. We manage them earlier. Total cost of care is improved.
That improves as a business our loss ratios, which we're very proud of and we believe are industry-leading, and we can then reinvest that superior margin back into member benefits. That's why we were able to sustain our benefits in the last season. MAs have been under pressure, but even in this environment, we've maintained and improved our benefits. So that's us reinvesting that margin back in, which then results in the growth and even more Clover Assistant usage, so that is our cycle. And I think you can see this is not aspirational. This is what has happened in the last year. We delivered on profitability. We delivered on improvements within Clover Assistant. We reinvested that back into the plan when others were retreating. We maintained and improved Star Ratings when others were going down. We then are able to maintain our product richness to make healthcare affordable.
And then we get more membership. And I want to emphasize one point here. I'm often asked, like, hey, what supplemental benefits are you providing? It's like gym memberships. It's dental. It's vision. And I think that's like an old way of looking at MA, right? I mean, those things are things that people like. But I don't think that's core to the way we think about it. We want to make accessing healthcare affordable to as many people as possible. And so rather than coming up with specific supplemental benefits, which we do, don't get me wrong, that's fun too, what we're looking to do is making healthcare affordable by lowering the out-of-pocket cost to our membership because that way they can afford to get the care that they need. And that is something we definitely see them prioritize: drug costs, like copays, coinsurance, etc.
We see them prioritize that. That's what they're shopping for. That's what drives member acquisition and growth for us. So I want to make sure that I talk about that. I really don't necessarily think it's just like supplemental benefits like it was five, eight, ten years ago. Okay, so our opportunity in front of us. We were proud and we're happy with the growth that we had in the last season: 27% growth, the majority of which came within our AEP area. We anticipate a strong OEP as well because we had good retention. We had good new applications in the AEP. We think that will continue through OEP. Our membership went up, as we said here, from around 79K to 100K, January to January. That's great. And we also feel very well placed to manage that 25 membership. Just quickly walking through why that's the case.
The vast majority of our membership were switchers from other MA plans. The vast majority were in our four-star plan, enrolled in our four-star plan for going into next year. The growth mainly happened within our core markets. So we're very familiar with these markets. We're comfortable. It's not like we launched a new market and got a lot of membership in a place we don't know. We are very familiar with these markets. Even Georgia, which was one of our smaller markets, we've been there for a number of years now. We're comfortable there. But we're also expanding our care capability, that home care capability in Georgia. And we had very strong retention. And we're very pleased with our cohort dynamics of our returning membership.
Between all of these things, while growth always brings somewhat a bit more variability into plan performance, we feel like that we've controlled for the major variables. We're familiar with mix. We're familiar with geo. We should have predictability around where we're going to land in terms of the economic performance of the plan. We had good growth. I think we were smart about it. We grew in the right areas. I feel comfortable about managing 25 membership. With all of those things, there is plenty of room for us to continue to grow within our core markets. I just said we grew in our core markets. I'm often asked about new states and things like that. I'm not against new states. I'm not saying that we won't go to new states with our plans.
However, I do want to emphasize that if we want the next dollar of revenue, the next member, the next dollar of EBITDA, those are almost certainly coming from our core states, New Jersey and Georgia. There's plenty of room to run in those states alone in terms of number of eligibles, in terms of market share. We are not saturated. I feel like we're going to do well in those markets for some years to come in the multi-year opportunity in front of us. People are still preferring PPOs. We're outstanding there. There are a limited number of four-star plans to buy into. We're one of them. We are able to continue to invest our profits into maintaining plan benefits. That's a great advantage. And we have that Clover Assistant network, which is a durable advantage.
Every year as our R&D flows through new features, new capabilities, as AI gets better and better, that will be something we can just quickly launch into production. We will be able to continue our advantage within these core markets, so I'm excited about what we can do there. Once again, I'm not saying we won't go to new places with the plan, but there's plenty of room to run in the markets we're already in. Which brings us to Counterpart, which I've already mentioned. For the areas where we don't have a plan, that doesn't mean that we're not going to do anything. I think that would be giving up an opportunity. Counterpart gives us a capital-light efficient way to bring our model of care into other geographies where we don't have a plan or we might not have a plan in the near future. Same exact technology platform.
Remember, we built this as a technology product. This is something that has been designed to be able to be replicated and deployed into other markets. So it's the exact same technology product, but in this case, we are not the MA plan. We're somewhere downstream of that. We're either helping the MA plan or we're helping a risk-bearing provider. The way I think about this is the customer of Counterpart is an entity that is taking on Medicare Advantage risk. That could be a plan. That could be a delegated entity. That's the customer. And we go to them and say, hey, here's our results, the results I've showed all of you today. And we're going to bring, we can bring, using Counterpart, that same capability to your primary care physicians that you are responsible for performance for.
And we will be able to improve clinical quality, and we will be able to improve total cost of care, we believe. So that's fundamentally what we're doing with Counterpart. I would encourage everyone to think about that in terms of we're doing both of these things: our maturity, our engine, growth. I feel like we're incredibly excited about what we're doing with our MA plans. And I'm also excited about Counterpart bringing that same model of care to additional markets where we don't have a plan. We've announced a couple of deals here already. We recently announced our partnership with Duke. We announced the Iowa Clinic earlier this year. But I would say we'll announce additional deals.
My goal here would be that we would relatively quickly be able to see a world where the number of lives under management for Counterpart Assistant actually has lives outside of our own plans that exceed the number of lives within our own plans. I think that's easily a world we can be in because our plans grow with a B2C motion. That takes the AEP, the OEP. There's a life cycle to that. Whereas Counterpart grows with a B2B motion, you can pull in many, many thousands of lives with one partner. So I think the growth dynamics are different. The economics are also different. So I'm not talking about that. I'm talking about the vision of helping as many physicians as possible, helping as many patients as possible, and having a lot of room to basically increase the number of lives under the assistant management.
Looking ahead, and then we'll have some time for questions. Our focus for 2025: membership and revenue growth. I've talked about that. We're going to take advantage of the MA market opportunity in front of us. Competitors pulling back. Our improvement in Star rating, our durability through our Clover Assistant network. We think we can grow. We will grow revenue. We will be able to grow our membership and our plans. We will balance profitability and reinvestments. There's a tremendous market opportunity in front of us, as I just said. We want to balance Adjusted EBITDA, maintain that profitability with thoughtful reinvestment into growth, especially given the fact that we're going into that four-star year, four-star benchmark year in 2026. Keep on improving our clinical quality and our outcomes. That means maintaining the four stars, beginning a progression towards maybe even higher Star ratings.
But certainly, four stars is our baseline at this point. And then growing that Counterpart business. We hadn't even announced Counterpart at this time last year. So now we've announced it. We will have much more to come on the Counterpart side. I'm very excited about what we're doing there. So hopefully, in a year's time, I'll be able to come back here, pull this slide up, and put some checkmarks next to this as well for all of you. All right. So happy to take some questions at this point from the room or from online. Do we have a mic for?
Oh, perfect. We actually have a question.
Do we have a question up here in the front? We're on super loud. Perfect. We have a mic coming in, I think, for the online folks. I appreciate the loudness, but you might not be able to yell into the internet quite as easily.
Thank you, Andrew. This is fabulous. I mean, it's fabulous. The challenge when we're in tangential businesses, how are you going to engage the providers to routinely get into another new platform? Since they're already asked to be on platforms, and then the pharmaceutical manufacturers want them to be in portals, and they're inundated and they're time-crunched. What is that next little bit that's going to engage them to spend a lot of time and truly engage in your platform?
Yeah. So how are you going to have them engage in a new platform? Definitely one of the things that we spend a lot of time thinking about. It is something that comes up very frequently, like, how do you make sure that there's an overall good experience, I think, for physicians is the way that we think about it. The approach that we're basically taking over here is that we are already able to, and we want to be able to be the platform that delivers what I think of as magic moments for physicians, and most platforms aren't even trying to do that, honestly.
They're like, okay, go here, do this or this, but who's actually thinking like, well, how do I impress or deliver a magic moment? I don't think people think about it that way, and so the way we do is that we have a huge advantage that we're trying to show the physician something that they find useful that they didn't know before, right?
What we're finding is that if we're able to do that, and we are already doing that, like these insights where they're like, oh, that's really useful. I didn't know that, that immediately creates a trigger in their mind where they say, I should check this and come back here more often. So while I can't necessarily take away the other platforms if they have to use those as well, maybe one day I will. But to start with, we want to make sure that they feel like our platform is where they want to be, not where they're being forced to be. That's the way we think about it.
Oh, another question.
Yes.
Hi, it's Dave and Zadim here at Diagnostics. Just a quick question. On those magic moments, can you share some examples what those magic moments could be? Just kind of your top three, maybe. It'd be interesting to know. Thank you.
Yeah, I'll give a couple of examples, but I think that they're so simple, but the fact that they're simple is why they're interestingly magic. One example of that would be we share information to the PCP, the primary care physician, and we'll say something like, hey, did you know that they went to see their cardiologist three weeks ago, and that cardiologist ordered blood work, and we would like to share the result of that blood work ordered by another physician with you, and that alone is a magic moment, and what we've heard, and this is one of my favorite quotes, is that we've heard physicians, PCPs come back and say, "I didn't even know my patient had a cardiologist.
It's not just that I didn't know they went to see them, that I didn't see the lab result, but I didn't even know they were seeing a cardiologist. Okay, that's really helpful to me. And of course, they can then also look at the results that we're sharing from over there. I think another one, and I'll give a second example over here, is that when we look at care and we look at, for example, how you do medication, right? And this is an example that I frequently give, but I think it continues to be something that resonates a lot with PCPs, is that when I said before around pharmacy, a lot of what a physician is doing is prescribing. I think we would agree.
But the fact that they don't have visibility into what's actually happening, so when we go to them and we say, hey, did you know that the patient's sitting in front of you right now? We have no record of paying for their medication in the last three months. The physician's like, oh, they did not mention it to me. That's not something patients just go to their physician and say, hey, I just stopped picking up my meds. They don't say that. So that's another magic moment because in the doctor's mind, they know that the patient would not bring that up on their own, but it's incredibly important to the delivery of care. Good question. Thank you. Was there anything else?
No. I think we're good. No.
Oh.
Thanks a lot for the presentation. Of course. One more question about the assistant. I'm curious to know exactly what kinds of information you're really sharing with providers, and I see on one of the slides you mentioned quality improvements, and you mentioned MLR improvements. How many of the suggestions or surfaced insights are relative to quality or documentation, diagnosis documentation?
Yeah. So I think there's a couple of different things that I can answer that way. There's no ratio that we're going in there. One of the parts that we, one of the things that is actually part of the technology platform is the engine is actually creating thousands, I'm going to say thousands, but it's a lot of different insights. It's not like just rule-based. It's a lot of, it's just generating from the data, and what could we show this physician? Okay, imagine that that's what's happening.
Then another layer of the platform is actually saying, how should I prioritize what I show to a physician, given that I have a limited amount of time to show? We don't mean like, here's 400 things you should look at, right? That's a terrible way to give a good experience. So we then stack rank those things. And then from there, there's another layer, which is like, what's the best way to show the information? So now that we've decided what we want to show, what's the best way to actually show that information to the clinician as well? So it's all of the above of what you just said, but it's a dynamic engine, not a fixed sort of like ratio that we're actually going in.
The one exception to that is there is a recapture period at the beginning of the year where all diagnoses are flushed by CMS. I think that's, I know why we do it, but I think that forces a very kind of bad experience with the physicians where you sort of have to redocument a lot of things at the start of a year. That's just part of the nature of how Medicare is run, and I think that should be cleaned up, but that might be a little different in the beginning of the year, but once we get to the second or third visit, it's going to be that dynamic engine prioritizing things. All right, we have 42 seconds because I promised that I wouldn't cause JPM to go long.
I guess I'll ask just a 30-second question. So I think most of our questions here have been around the platform and the technology. What do you see what's coming next in 2025 for Clover Assistant?
Yeah, so what's coming next is there is a huge opportunity to bring the capabilities around LLMs and boost Clover Assistant around that. And we've been looking at this for a while, researching for a while, of course. But when I talk about AI and ML, we are talking about not LLMs. Most people here probably think of the LLM version of AI. We're talking about more like that clinical stack ranking. We're talking about all those kinds of things, which is slightly different. It can be done by LLMs, but it's not the first thing you would use an LLM for.
So we have a number of features around LLMs that I'm very excited about to bring to market that combines that power of LLMs with what we're already doing with AI, which we're seeing all the results of within CA. And I think that's going to have a compounding multiplicative effect.
Perfect. Thank you, guys, for your time. And, Andrew, thank you for the presentation.
Yeah, and I'll be down to the floor. So thanks, everybody. Thanks for the questions.