American Well Corporation (NYSE:AMWL) Q4 2021 Results Conference Call February 24, 2022 5:00 PM ET
Dr. Ido Schoenberg - Chairman and Chief Executive Officer
Robert Shepardson - Chief Financial Officer
Conference Call Participants
Ricky Goldwasser - Morgan Stanley
Charles Rhyee - Cowen
Cindy Motz - Goldman Sachs
Glen Santangelo - Jefferies
Eric Percher - Nephron Research
Stan Bernstein - Wells Fargo
Allen Lutz - Bank of America
David Larsen - BTIG
Jailendra Singh - Credit Suisse
Jessica Tassan - Piper Sandler
Good afternoon, and welcome to Amwell's Fourth Quarter and Full Year 2021 Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded.
Leading today's call are Dr.Ido Schoenberg, Chairman and Co-Chief Executive Officer; and Robert Shepardson, Chief Financial Officer. Ido and Robert will offer their prepared remarks, and then they will take your questions.
The Amwell press release and webcast link are available on the Investor Relations section of Amwell's website.
Please note that we will be discussing certain non-GAAP financial measures that we believe are important in evaluating Amwell's performance. Details on the relationship between these non-GAAP measures to the most comparable GAAP measures and reconciliations thereof can be found in the press release that is posted on our website.
Also, please note that certain statements made during this call will be forward-looking statements as defined by the Private Securities Litigation Reform Act of 1995. Such forward-looking statements are subject to risks, uncertainties and other factors that could cause the results for Amwell to differ materially from those expressed or implied in this call.
And now I'll turn the call over to Ido Schoenberg, CEO of Amwell. Ido?
Dr. Ido Schoenberg
Thanks, operator, and welcome, everyone. On today's call, I'll start by reviewing some highlights of the quarter and the full year. Then I'll take a moment to discuss the market for our solution and anticipated drivers of our growth. I'll conclude my remarks with a review of our strategic priorities for 2022. Bob will then review some key metrics, our financial results and our guidance. Then we'll open the discussion for your questions.
At Amwell, our mission is clear: To define the fundamental infrastructure that connects providers, payers, patients and innovators and enable them to deliver on the promise of greater access to affordable high-quality care. We are transforming care delivery from in-office to integrated in-person, virtual and automated care modalities.
Much like Amazon assembled a distribution framework for sellers and buyers to find one another, Amwell's Converge solution equips providers, patients and the industry that supports them with the tools to realize a better health care experience.
Q4 was a great quarter that advanced our progress in many ways as we turn the page to Converge. Our teams are executing well and the market for our solution is strong.
I'd like to take a moment to recap a few highlights of the quarter. First, we are making good progress on our development and deployment of Converge. With a substantial part of our platform delivered and in live deployments, our R&D efforts are now focused on bringing Converge over the finish line with several remaining models that are in high demand from our customers and prospects.
Second, underscoring this progress during Q4, Converge received major new industry recognition. Gartner ranked Converge at the top of their recent review. Converge also won the recognition of Chilmark Research, the MIT Leadership Council, and was also named Best New Application in Telehealth by UCSF Health Awards. We are honored by this amount of favorable responses to Converge, and for Amwell, as a leader and innovator in our space.
And with Converge, we registered large multiyear bookings in Q4 that are indicative of new clients' receptivity to our new platform. We expect this strong momentum should continue to build as Converge functionalities completed and as we work to continue to bring in large new health care payers and systems on Converge and add modules for existing customers.
Turning to some of our newer initiatives. We are well underway with the integration of newly acquired Conversa and SilverCloud, which advance our approach to long-term longitudinal care and higher-margin revenue mix over time. With Conversa, we ended 2021 with subscriptions to 179 automated virtual care programs designed to remotely manage patients on behalf of leading health care organizations across the country with some notable Q4 wins, including expansions at existing Amwell legacy and Converge health system customers.
These programs are enhancing patient access and experience while improving outcomes. Automated programs help providers to identify patients who need more support and reduce the burnout among frontline workers by offloading some of their workloads. And we are also building momentum with SilverCloud, our digital mental health delivery module.
During Q4, we won several large health systems bookings for SilverCloud. SilverCloud is also a success in international markets, where we recently won a strategic contract in the U.K. providing nationwide behavioral health services to over 3 million wealth citizens.
Now I would like to shift to some highlights from our services team. Our services team is currently busy converting existing health system customers from our legacy platforms to Converge. As we look to describe our progress, we believe visits on Converge are the most meaningful metric of progress converting our business to the new platform.
We ended Q4 with about 10% of our visits on Converge, and it is our goal to significantly increase that metric during 2022. In addition to these successful migrations, we are seeing growing engagement across all our products. Our EHR integrated solutions are a big driver of our number of active providers, which grew 26% over a year ago. And the number of visits per patient grew a healthy 14% over a year ago, something we view as indicative of sustained adoption among patients.
To wrap up Q4 highlights, I want to add that I'm proud of how our teams are executing, putting in place the crucial elements of our solution, landing new names, driving engagement and executing towards our mission to define the fundamental infrastructure, enabling the future of health care delivery.
Next, I'd like to spend some time briefly reviewing what we are seeing in the market for our solution. It is now well-accepted that the pandemic dramatically accelerated telehealth adoption. It is also clear that with this acceleration, urgent care visits have been commoditized. Providers, payers and innovators are now seeking to realize the much broader potential of hybrid care and are facing significant challenges. These include complicated and confusing rules and regulations, difficulty to simply and practically implement and integrate sophisticated workflows, disparity of point solutions portals and fragmented nonintegrated documentation and payment systems.
Our conversations with our customers are indicating that the market is primed for more complete unified solution for digital health delivery, one that requires a true partner to define and implement their digital health strategy. At Amwell, we are defining a distribution enablement platform for health care that addresses many challenges and solve for the industry requires.
It's not just about the virtual visit. It is extending care beyond telehealth to a digital care continuum. More than connecting providers and patients for a video visit, we are creating and managing the logistics to support the emerging hybrid care delivery model. We are delivering on that vision with Converge.
For example, a provider network using Converge at the heart of their IT infrastructure, will be able to significantly increase the reach and impact of their providers by optimizing the use of physical, virtual and automated interventions. Payers will be able to share insights on gaps in care and empower providers to address these important gaps. Patients will be continuously connected to their health care system and enjoy the convenience and efficiency of digital-first health care experience. All of this means a powerful impact on clinical and financial outcomes benefiting all segments.
In addition, ongoing rapid innovation requires a modular infrastructure that unifies a fragmented landscape and future proofs providers and payers, supporting their ability to adopt new functionality. Innovation is on the rise and Converge will be the infrastructure to help health systems prioritize and embed all those solutions into harmonized ecosystem, an ecosystem which streamlines the exchange of data, services and capabilities underpinning a truly hybrid multi-model solution, including natural language processing, as well as AI and analytics that are moving from the sidelines to center stage.
With Converge, we are expanding well beyond traditional telehealth. It's a transformational vision defining a new digital care delivery system that empowers providers and offer meaningful new ways for patients to receive health care. We are addressing a growing TAM with a more complete solution set and the market is taking notice.
Now I'd like to take a moment to highlight a couple of our key priorities for the coming year, which are aligned behind our overall mission and our goal of achieving sustainable, high-margin revenue growth.
We will complete the build-out of Converge. We will make the final investments that will complete this distribution system and enable a digital transformation that extends beyond the traditional care model. This final effort adds essential Converge features in high demand by our large providers and payers, including on-demand and white label enablement capabilities. With Converge nearly complete, the major focus for us in 2022 will be to migrate existing customers to the platform and drive sales with new customers.
We will also strive to demonstrate the modularity and flexibility of our solution by helping our customers expand the use of our solution and adopt new innovative elements of hybrid digital care. With some of our largest new Converge customers, we are building out infrastructure elements that will apply broadly to the market and set the standard for best practice in digital health care delivery, accelerating our ability to deploy our solution in the future.
Finally, our customer footprint and mind share give us a unique point of view at the forefront of the digital health landscape. SilverCloud and Conversa are a great example of carefully chosen investments that allow us to leverage Converge to deliver crucial elements of longitudinal care to our customers and their patients.
While future M&A is not needed for us to achieve our goals, our strong balance sheet and promising financial model enable us to invest internally and externally for growth and to further differentiate our solution. With these top priorities in focus, we believe that we are well-positioned to add new customers, expand within our installed base, grow the mix of our higher-margin revenue, and over time, demonstrate the potential of our long-term operating model.
Now I would like to turn the call over to Bob, who will cover our Q4 financials and some key metrics on the quarter, plus our guidance for 2022 and path to profitability. Bob?
Thank you, Ido, and thanks, everyone, for joining the call this evening. Today, I'm going to touch on some key operating metrics from 2021, run through our fourth quarter results, provide our initial outlook for 2022 and conclude my comments by sharing our view on Amwell's Path to Profitability framework.
First, I would like to review a few key metrics and try to put those in context. Active providers on our platform, specifically non-AMG active providers, is one measure of platform adoption among our customers. We ended the quarter with over 91,000 active providers, 88,000 of which were non-AMG. That's relative to 76,000 in third quarter of '21 and 68,000 in 4Q '20 for growth of 16% in the quarter and 30% for the year.
Turning to visits. As a further measure of the continued acceptance of virtual care, we saw healthy total visits of 5.8 million in 2021, which rivaled 2020, a time when many care facilities were closed and virtual care was a pandemic must have. In 2021, virtual visits became more ingrained as a standard of care. Scheduled visits increased 30% over 2020 and grew to 70% of our total visit volume. Specialty visits, inclusive of behavioral, which we view as an important part of our long-term strategy towards more longitudinal care, grew almost 50% versus 2020. AMG visits declined about 10% in 2021 and were 1.4 million.
Average annual contract value, or ACV, is a good indicator of the value we are delivering to our customers and the success of our land and expand strategy. And a real bright spot for the quarter, health plan ACV increased 18% to $723,000 in 2021 compared to 2020. ACV for Health Systems saw a more modest increase from $334,000 to $356,000 as we expected during this time of Converge transition.
Also, as expected, average customer counts across both health plans and health systems were relatively flat at year-end. We expect both ACV and customer counts, especially for health systems, to reaccelerate in the second half of the year once the full features and functionality of Converge are broadly available in the market and recent customer bookings are implemented and go live.
We exited 2021 with 49 customers on the Converge platform, primarily on Amwell Now, our entry-level experience for patients and providers. We also have several customers deployed on Converge EHR, which integrates scheduled visits, a key feature of our platform.
A metric we would like to highlight today that we think demonstrates our progress in customer migration is that approximately 10% of our visits were taking place on Converge at year-end. We expect that percentage to rise significantly over the course of 2022 with health systems leading the way.
And now on to Q4 results. Total revenue was $72.7 million for the quarter, reflecting growth of 20% over Q4 of last year. Normalizing for M&A-driven churn in our customer base, which we have discussed in prior calls, growth versus 4Q '20 was 24%. The components of revenue are as follows: Subscription revenue grew 14% to $30.1 million in Q4 compared to Q4 of 2020. Again, normalizing for M&A-driven churn, growth versus 4Q '20 was 22%.
Subscription revenue in Q4 increased versus 3Q '21, primarily due to several go-lives in late third quarter and fourth quarter, as well as a small amount of revenue from Conversa and SilverCloud.
Visit revenue continues to be influenced by the pace of the pandemic. For example, visit revenue of $31.2 million grew 19%, and AMG visits in the quarter at $392,000 were up 10% versus last year, reflecting a rise in urgent care visits in Q4 due to the Omicron variant average revenue per visit was $80 for the quarter, up from $74 a year ago in 4Q '20, but lower than in prior quarters of '21. Our AMG business is an important differentiator in the market and critical to many of our clients, and we view the offering as a supporting element of our Converge strategy.
As we have said previously, our primary focus going forward is to drive high-margin recurring revenue associated with sales of the Converge platform, plus a growing number of modules and services like AMG.
Our services in Carepoints revenue grew 45% to $11.4 million, with the strength being driven by the timing of marketing engagement campaigns on behalf of our customers. As a reminder, these programs are dependent on our clients' marketing campaign strategies and are variable. These two aspects of our business have remained stable over time and provide us great visibility in a loyal customer footprint.
Largely due to the revenue contribution of our services and Carepoints business, gross margin was 350 basis points lower than last quarter at 40%. In general, we believe as we ramp up Converge deployments, the shift toward higher margin revenue will lift gross margins.
Turning to operating expenses. And in support of our Converge strategy, we ramped R&D spending in Q4 to $33.8 million from an average of approximately $25 million per quarter over the last year. Also contributing to the increase is that Q4 was our first full quarter consolidating the spending of SilverCloud and Conversa.
Sales and marketing spending also increased in Q4 to $21.3 million, mainly due to increases attributable to the acquisitions. We expect R&D to continue to increase in the first half of '22 as Converge spending peaks, then begin to decline during the back half.
Finally, I have a few comments on our EBITDA profitability. As we prioritize the investments in Converge and absorbed costs associated with SilverCloud and Conversa, our adjusted EBITDA was impacted, and in Q4, represented a loss of $41.1 million. This is up about $5.7 million compared to Q4 of '20. We view this rate of loss as temporary and important strategically.
Transitioning to the balance sheet. We ended the quarter with $746 million in cash and short-term investments. Our balance sheet is healthy and provides us the resources to fund this temporary period of investing and allows us the flexibility to consider and pursue strategic opportunities that are aligned with our goals.
Turning now to our 2022 outlook. For the full year 2022, we expect revenue to be in the range of $275 million to $285 million. We expect a modest bump in subscription revenue, roughly flat subscription revenue from core Amwell products as we continue to focus on our migration to Converge, plus upside subscription revenue contributions from our automated care and behavioral health programs through SilverCloud and Conversa, consistent with our commentary at the time of these acquisitions. We also are taking a conservative view on visits with paid volume expected to be largely flat over 2021 at around 1.4 million to 1.5 million visits. We expect Carepoints and services to remain stable at 10% of revenue, and we anticipate revenue seasonality similar to prior years.
Our revenue outlook reflects some previously described impacts on our 2021 bookings, including delays by some health systems who have chosen Amwell, but are waiting for the completion of Converge in order to go live on that platform, and some who were prioritizing COVID-related challenges over adding new features to their existing infrastructure or transitioning to a new software platform. These are timing issues, and we believe the healthy Q4 bookings Ido talked about earlier will begin to convert to revenue in the second half of 2022. Combined with Converge bookings accelerating in the second half, we should exit the year set up for a strong 2023.
Now on to our guidance on profitability. For the full year 2022, we expect our EBITDA loss will be in the range of $190 million to $200 million. Underlying our profitability outlook are three strategic investments we are making, which are temporary in nature and will set us up for significant success going forward.
First are the elevated Converge-related R&D investments as well as the cost of maintaining the legacy Amwell platform during the transition. Second is a somewhat higher level of services spend we plan on to ensure customer migrations to Converge go seamlessly and pave the way for future success. Finally, there is the impact of consolidating the losses associated with our acquisitions. Importantly, we believe we will contain all of these impacts to 2022, and as we put this period of investment behind us, we will be positioned nicely for future profitable growth. The rest of our GAAP and non-GAAP guidance can be found in our press release.
Now that I have completed my first quarter as Amwell CFO, I'm more pleased than ever to be here as we turn the page from telehealth to health care delivery infrastructure, from our legacy platform to Converge. I would like to take this time of transition as we begin the year to make some additional commentary about our longer-range view of our income statement.
This quarter, we are providing a new graphic at the end of the earnings release intended to illustrate our Path to Profitability Framework. I'd like briefly to summarize, and we will welcome your questions after you've had time to take in the details.
Our model assumes the three temporary impacts to EBITDA that we just described. These items lowered our adjusted EBITDA margin by 10 percentage points in 2021, and our guidance today assumes an additional impact of 25 to 30 percentage points during 2022. As we look to 2023, our acquisitions will be integrated, synergies will begin to be realized and we expect our R&D spending to decline on an absolute basis by 40% to 50% compared to 2022.
With the temporary costs meaningfully abating and our revenue mix shifting toward higher-margin revenue, we believe an increase in our gross margins will combine with operating leverage across the income statement resulting in an improvement in our 2023 adjusted EBITDA margin of well over 50%. From there, we see a clear path to adjusted EBITDA breakeven on approximately $500 million in revenue.
I hope that's some helpful context. This model is firmly rooted in our belief that our market can support 20% revenue growth ongoing and that our high-margin visible subscription revenue will grow more rapidly than visit revenue, as well as our Carepoints and services revenue. And while it is further out, with breakeven behind us, we see a path to 20-plus percent EBITDA margins that is a matter of time and execution.
Finally, to provide you with some additional detail into our plans for steady-state R&D spending, we intend to normalize R&D at a level of approximately 30% of our subscription revenue. This is in line with SaaS companies today, and we believe this rate of investing provides us the resources we need to continue to invest, deliver on our vision, lead our market and achieve sustainable profitable growth.
Before handing the call back to Ido for some final remarks, I would like to summarize. Q4 completed a strategic year for us, one in which we made important investments and good progress towards our goals. I look forward to working with all of you as well on this journey.
With that, I will turn the call back to Ido for some final remarks. Ido?
Dr. Ido Schoenberg
Thank you, Bob. We are proud of our team, our progress in Q4 and the full year of 2021. The opportunity in front of us is clearly large and expanding.
Before I end the call and take your questions, I want to summarize by pointing out that at the heart of many of our customers' growth strategies is a focus on digital-first hybrid care programs. We are optimally suited to enable these programs given our unique vision in our differentiated solution. We have a substantial installed base and our vision is grounded in customer mind share from the largest payers in the country and health systems representing over 2,000 hospitals and over 90,000 providers.
Converge is the connective tissue that will integrate the disparate assets and operating environments required to deliver on the potential of hybrid digital health care. The market is moving towards us. We believe ours is the right vision, and we are well on our way making the right investments and executing well to deliver on the promise of this meaningful opportunity.
Thank you for your time. Operator, please open the call for Q&A.
[Operator Instructions] Your first question comes from the line of Ricky Goldwasser with Morgan Stanley.
So Bob, thank you for giving sort of the bridge of the road map to revenue targets through 2025 into profitability. So just to start, how should we think about the cadence of the 20% revenue growth from '23 to '25? Should we assume that in '23, you're going to be above 20% as we expect to see kind of like the pent-up demand from Converge migration? Just helpful to think about the cadence.
And then what are just the additional underlying assumptions that you're making in terms of subscription revenue versus business mix and sort of kind of gross margin targets by 2025?
Thanks, Ricky. Yes. So look, we thought it was really important to give people a view of -- especially with the spend this year, how the income statement would work going forward. The estimate of 20% revenue growth. I wouldn't take as -- is it going to be '25 and '23, and '23 and '24. I would think of it as over this period of time, the composite revenue growth for the enterprise should be around 20%. It's a, I think, a conservative assumption that bakes in a higher growth rate for the software business, call it, mid-20s and a mid-teens growth rate for visits that would get us to a composite growth rate of around 20%.
And that in and of itself, when you think about the impact on EBITDA, is really meaningful because you -- we're talking about software with the benefits of Converge probably taking gross profit margins on the software business into the mid- to high 70s. And the visit business getting incrementally more efficient but at a much lower gross profit margin there.
So the pickup in gross margin is substantial from around 40 -- low 40s now well into the 50s by 2025, just from the disparate growth rates and the pickup in efficiency that we're going to see from running one platform on Converge as opposed to multiple platforms on our legacy.
That's very helpful. And just one follow-up question. Ido, in your prepared remarks, you talked about the move towards more scheduled visits and specialty visits. Maybe you can give us some context and where you're seeing sort of this new adoption health from the urgent care that we saw sort of 1 year, 1.5 years ago to a more sort of steady state and fundamental demand.
Dr. Ido Schoenberg
Ricky, it is really astounding to see how far we came in such a short period of time. So for many years, telehealth was an adjacent service alternative to the rest of health care. And it's focused on urgent care as the convenience in case it was too hard or too long to wait for a doctor.
Very much because of COVID, but also based on other factors, what happened is that the mainstream health care players, starting with providers, have discovered the power and efficiency of digital connectivity with their own patients. That is not a simple substitute, but rather a whole transformation that really expand the use of digital technology not only to urgent care, but really across the entire care continuum, and opens opportunities to collect data continuously to analyze it, to rethink care plans and really bring in other important participants starting with payers an ability to integrate depths and care and work in an efficient way that is much more appropriate and efficient to really improve the outcomes.
It is also important to state that while in the early pandemic days, the utilization of telehealth was very simplistic in the sense that it was a video conferencing that replaces the in-room visit. But shortly thereafter, our clients have discovered that a technology can bring so many other elements, some of which I just described, that eventually are really transforming the way that people experience health care and are really contributing to a dramatic rise in access to care, in efficiency, in many ways democratizing health care.
What we saw even with our own health systems, it's very dramatic in the sense that when we started, it was almost like a sideshow for allowing people to talk to some of the doctors some of the time. And it is now very quickly becoming a necessary mainstream modality of communication and connection between them and their trusting patients.
Your next question comes from the line of Charles Rhyee with Cowen.
Yes. Ido, maybe -- you highlighted earlier, obviously, this year, there is some impact from customers who have maybe delayed purchasing models or other clients what selected them but are waiting to deploy Converge first. Maybe just overall, if you look at the number of health systems clients that you have, how many have -- at least raised our hands to say we will be migrating to Converge? And for those that are choosing not to, what are some of the reasons why someone might not want to transition over to Converge?
Dr. Ido Schoenberg
Charles, I'm hard pressed to think about the single client that refuses to upgrade to Converge. And it's very easy for us to think about so many that are strongly raising their hand waiting for the upgrade. Well, the receptivity to the new platform is much better than we ever hoped. Our clients are very excited to do that. And of course, the timing issue that Bob mentioned is really the key factor that is driving it together with the availability of the various modules depending on the complexity and the needs of each and every client. So I'd like to point out that all this will be behind us by the end of the year when the entire spectrum of functionality will be available across the ecosystem to all our clients.
Okay. That's helpful. And maybe as a follow-up, Bob, when I look at the slide here, you kind of bucket the three things, right, the M&A impact, the Converge impact. And I'm assuming that the services spend to support clients that's in that bucket. But obviously, the core EBITDA loss improved dramatically, and I know you highlighted a number of factors. Is this all just really just mix shift as we get more subscription revenue? Or what else could be contributing to this big improvement? And I guess the question really is, what's the visibility you have to getting that part of it realized?
Thanks, Charles. Look, a big chunk of it is what I mentioned on the R&D line item, which we have a lot of visibility on, which is we expect that to decline by 40% to 50% in '23 relative to the spend in '22. So that's a dramatic increase or decrease in cost.
And then on the -- it's a more modest impact from mix shift from faster growth on a year-over-year basis on subscription versus visits. But the two, if you just look at the low end of my estimates in terms of what they would deliver, it's somewhere in the $100 million to $110 million of a benefit from those two items from gross profit improvements and from R&D.
And so if you think about the walk from the guidance of, call it, midpoint of $195 million negative and take $105 million off of that, that in and of itself is a huge delta in terms of the negative EBITDA margin that we expect that we think we can achieve in 2023 going from around 70% negative to around 25%.
So those are the two biggest components, and we would expect to see, as I said, some operating leverage across the other items as well.
Your next question comes from the line of Cindy Motz with Goldman Sachs.
I just want to follow up a little bit more on the Converge migration. I know you said to think about it in terms of visits, and people seem really excited about it and happy with it. But in terms of -- are you being asked to -- like with the ramp in R&D, are you finding that it is -- people are asking for more things or unexpected complexity? Just trying to get an idea of that.
And because I had thought maybe there would be some delays because of the Omicron variants and maybe distraction. But just curious about that R&D number. And I know you said it would go down. But maybe just a little more color on what they may be asking for.
Dr. Ido Schoenberg
Cindy, you asked many things. I'll try to answer fully. As I mentioned earlier, we really saw two things that are influencing our view to the year. One that you mentioned is the Omicron dynamics. The Omicron dynamics really does two things. One, it's unpredictable. And we saw that in the last two quarters of 2021. For example, the unexpected separation between demand for urgent care and be able to help was a surprise. And we adjusted conservatively our view into the year based on that.
The second element was that during Omicron, despite the fact that the population overall was much more relaxed, hospitals were full and that impacted the ability of some of our clients to implement a new project or extend the existing ones.
Another thing that happened is the appetite for very large enterprise platform grew much faster than we realized. I talked about the bookings of Q4. We are very proud of what's happening in the market, some of the largest, most sophisticated organizations are choosing Converge. As you can imagine, when you deal with such scale, sales cycle is longer, deployment cycle integration is longer. But it's very much worth it because the upside is also much more significant and much more sticky. Such deployment usually signifies the potential relationship for five years or even a decade when you think about the span of digital care delivery today.
When we began to deploy such clients or negotiate it with such clients, we didn't only focus on what we developed so far in our short-term plan, but really shared some of the multiyear roadmap. The enthusiasm really continue there, and they really hard pressed us to begin to expand and accelerate our growth in a way that it's still very lucrative growth for them and for us. We leverage our healthy balance sheet in order to make that investment in acceleration and are going to end the year with a platform that is far broader, far more differentiated and far more capable than we originally planned, well beyond the original plan. And also in that way securing some very sought after clients in the market that we believe we can go together for many years. I hope that answers your question.
No, sure. That makes sense. And just as a follow-up, in terms of -- I know you broke it out with the Converge adjusted EBITDA impact. How should we think about 2022? Like, is it -- I would assume is it more back-ended that we start to see improvement or -- I know you don't give quarterly guidance, but if you could give us any help there just on the cadence, that would be great for 2020.
Sure. Cindy, I would say, top line that if you think about our seasonality, overall, the quarterly breakup in '22 should be similar we think to what we saw in 2021. So that's kind of the top line.
Qualitatively, I would say we would expect to see on the subscription side, as Ido talked about, some of the fourth quarter bookings that we saw to be implemented and start to generate revenues in the back half once we go live on those.
From a cost perspective, you're going to see R&D continue to ramp from that fourth quarter number through the first half of the year, and then start to tail off in the back half of the year. And so those are, I think, really the two most important line items there.
From an M&A loss perspective, I don't think there will be really much in the way of quarterly differential there. The big driver is really on the cost side going to be the ramp-up and then the subsequent ramp down in R&D.
Your next question comes from the line of Glen Santangelo with Jefferies.
Ido, I just want to go back to some of your comments you made in your prepared remarks with respect to the overall competitive landscape. It kind of sounds like the 91,000 physicians that you mentioned that are currently using the platform, could you maybe comment on the exclusive nature of those relationships? I think what investors are really trying to get their arms around is how the competitive landscape may be evolving in telehealth more broadly in terms of what you're seeing because that sort of leads into my second question for Bob.
You're forecasting subscription revenue growth in the mid-20s through 2025. And I think we understand 2023 will probably be a little bit of a bigger year as maybe you move out of the transition year and you probably play a little bit of catch-up. But what gives you the confidence in those growth rates in '24 and '25? And could you maybe talk through that algorithm of how the revenue model works once you convert one of these clients to Converge? Can you charge them more like is there an upsell attached to it? Just any thoughts around those general questions. Sorry, this is kind of a long question, but any thoughts would be appreciated.
Dr. Ido Schoenberg
Sure, Glenn. I'll take the first one and let Bob take the second. As it relates to the competitive landscape, we're really seeing two trends that are important. There is the, I would say, the simple end, it's not necessarily the low end, where you really see simple connectivity. And that's a very crowded market with Zoom and Epic and Teams and long list of many other known players are playing. And we definitely see a lot of competition there. The products are similar. Price is a big issue. And for that type of use case, it's very difficult to differentiate.
In addition to that, as I mentioned earlier, what we are seeing is that a growing number of our customers are looking for things well beyond simple connectivity, workflow integration and many things that I described earlier. The growth that you saw in our active providers is mostly non-AMG. These are not adopted to providing urgent care and our services as part of our network. These are really our client's providers. It's an indicator of traction within our 2,000 hospital clients. And they use our platform for various things, almost all of them require well beyond simple connectivity and are becoming more and more sophisticated with a very large number of the modules.
The most sophisticated you are, the bigger you are, the more confident we are that we provide more value and differentiate. Although Amwell now is actually also very well received for a simple connectivity task.
As far as what you call exclusivity, it is not uncommon to see with some of our clients using multiple video platform for simple connectivity needs. These don't essentially compete, although, in theory, they could grow and become Amwell someday. But right now, they don't come near to what we offer and provide. And when you think about digital care delivery platform, we feel quite secure that we don't -- I'm not recalling any client of ours that has two of those. They typically have one infrastructure because of the power of having one connected platform. It doesn't actually make much sense that we have a few unless there is an usual situation, like an M&A or things of the nature, that eventually resolved by having one platform take over the other.
And I'll let Bob answer the second part.
Glen, so I guess a couple of ways to address that, the second part of your question. I'm going to let Ido address some of the vision aspects here. But as we look at it, we've got -- and I know there's a lot of TAM fatigue out there on the part of investors. But we see, at least a $75 billion TAM for our market, and looking at what some of our brethren have out there at 150 to 250, we may be a little bit conservative in that estimate.
But as we think about growth, we certainly have that underpinning it. But we also have this land-and-expand strategy where once we're in, there are ample opportunities to upsell. And the way we structure our contracts, especially with some of the larger, more sophisticated multi-silo buyers are -- have a lot of volume sensitivity across the usage there.
So to give you a sense, I mean, we estimate our bookings in the last year or two and kind of where we are right now. bookings are probably split, 40-60, something like that between new customers and upsell. Over time, over the next few years, I would expect that, that gets closer to 75% of our software bookings are from upsells.
And so that's -- when you think about the TAM and you think about the types of customers we have and how their service offerings are going to evolve over this time, we still think we're in the very early innings of the evolution of the care delivery modes that we're going to see and they will become much more hybrid in nature to think that we could grow our top line software subscriptions at a rate in the mid-20s doesn't seem like -- it doesn't seem too heroic.
Your next question comes from the line of Eric Percher with Nephron Research.
Question on behavioral trend. I know as we end '21, there's been a discussion of decoupling from COVID. Could you speak to your expectations for '22? I think a reminder on just how impactful behavior is to the overall organization and some of the expectations for SilverCloud would be helpful.
Dr. Ido Schoenberg
Absolutely, Eric. So I want to really point out that when I spoke about behavioral, I was really focusing on our ability to have paid visits done either on urgent care or the everis driving the AMG revenue part of our business. When you think about the ever health in a much broader scale, the opportunity is to really serve enormous unmet need in the market by connecting and amplifying the available therapies and other types of providers in the market so they can go ahead and do a much better job. And we mentioned the example of what SilverCloud is doing in the U.K. with NHS for many years with proven results, extracting much more services for each therapy [Technical Difficulty] is a much broader topic than therapy alone in the sense there are variable elements in every type of longitudinal chronic care program that really requires the collaboration and many other elements to really achieve the outcomes that are promised as part of that program.
With Converge, we are now able to really inject the Ever Health elements in a very efficient way, combining physical, virtual and automated to everything our clients are doing and not by selling those services ourselves, but rather by connecting available services in a very efficient way, to an ecosystem that is using a unified platform. I hope that's helpful.
And I guess, the question there would be, is there a metric relative to the percentage of visits that are behavioral? And I also -- my second question here was going to be in that R&D ramp down, is there an expectation of sunsetting the legacy system as early as '23, '24, '25?
Dr. Ido Schoenberg
I'll start with the end. The answer is absolutely yes. We have -- it's important to maybe note that it's not a legacy system, but they're [indiscernible] systems. We have quite a few systems running in parallel, and we are definitely planning to sunset all of them.
I'd also like to point out that behavioral health visit is one facet of behavior health care, a lot of behavioral health, CBT, for example, is totally automated and does not require even a single visit in order to be effective. And maybe I'll let Bob complete my answer.
Eric, I don't think we have made decisions yet on exactly how we're going to report SilverCloud, for instance, going forward in our KPIs, whether you'll see just the impact blended across ACV or in another way. So I guess, the way that right now you can see its impact is on visit revenue, average revenue per visit. And on the SilverCloud side, again, I don't think we've really put a pin in that. So I'll keep you posted, but that's probably not something we're going to address until maybe this time next year.
Your next question comes from the line of Stan Berenshteyn with Wells Fargo.
I guess, I'd like to ask a question on the Oracle acquisition of Cerner. Maybe a two-parter here. So first, maybe high level. What are your expectations for your strategic partnership with Cerner after this deal goes through? And then maybe related to that, has this merger in any way change the buying behavior of health systems that are currently using Cerner? Any comment on that would be great.
Dr. Ido Schoenberg
Sure. So I don't recall any pushback or otherwise from joint clients, Amwell-Cerner clients related to the acquisition. We also didn't notice any change in our relationship with Cerner, at least not so far. As you know, we have many common clients. We are well integrated, and those clients are happy. And that's great for both Cerner and Amwell. We complement each other well.
Our initial very -- the initial dialogue with Oracle is positive. So we have no reason to believe why this relationship will not continue in the growth.
Got it. Got it. That's helpful. And then also, I want to go back to actually a comment you made last quarter. It was in reference to a survey by HIMSS where you called out 56% of health systems are planning to invest in virtual care. I'm curious as your sales reps are speaking to health systems on the ground or through Zoom or whatever it is, are health systems communicating this type of meaningful adoption beyond what they already have in terms of growing their capabilities longer term. Is that something that you're hearing? Or is that just your expectations, without any commentary from clients?
Dr. Ido Schoenberg
Yes. So of course, we are spending a lot of time with the clients. And I would suggest that there are two things that we see. One is the COVID related, the pandemic-related visits are definitely on the decline. So the siege from patients that are trying to reach out digitally because they can't do it otherwise, it's really driving smaller traction.
But on the flip side, the discovery of the enormous value of digital connectivity is definitely there, and the dialogue has changed from quick wins to connect because of necessity to really a see change in thinking about how they're going to really connect with patients and participate with a digital care distribution platform that is really connecting entities that didn't connect before.
Saying it differently, there was a time where a hospital did not have an EHR. And now nobody is even thinking about the hospital without an EHR. I'm pretty sure that based on our dialogue with our customers, growingly, the leaders in those organizations are seeing don't imagine a future where the digital envelope of the relationship with their patients and the rest of the ecosystem will not be enabled by an enterprise-grade solution.
Your next question comes from the line of Allen Lutz with Bank of America.
I guess, how much revenue from Conversa and SilverCloud was there in the quarter? And then do you still expect revenue for those businesses to double in 2022?
Thanks, Allen. We -- on SilverCloud and Conversa, the guidance we gave on the M&A call is still very much what we expect to see. So kind of a minimum of around $30 million top line. The -- we're not breaking out revenues for the two of them going forward or historically here.
I will tell you, it was a very small single-digit kind of number in the quarter. And what we're seeing in aggregate out of the two acquisitions is just terrific opportunity from a cross-sell perspective, both ways to their base and them into our base. And we're really looking forward to moving forward on the integration and realizing a lot of these synergies.
And I think another one for you. The increase in ACV across health systems and health plans was helpful. I guess can you talk about how much of that was due to new module growth? If any of it was due to price? And then if you think about, I guess, just how do you think about price as a lever to growth within Converge over the next few years.
I wouldn't call it price. I think it's really upsell, right? New modules and volume-based triggers the way we're structuring some of these contracts is the way I would be thinking about it. And we really only saw meaningful movement on the health plan side, and that was really just more adoption and more programs into those plans. So that's working as expected on the plan side. I talked about and I referred to as well, some of the headwinds that we've seen on the health systems side. And that's, I think, why those numbers were a little flatter.
Your next question comes from the line of David Larsen with BTIG.
I think you mentioned that about 10% of visits -- or what percentage of visits are on Converge now? Was it like 10% in 4Q? And then how should that trend over the course of 2022? Will you be at 100% by 4Q '22 and that's what's going to enable you to have all the R&D savings in 2023?
Your lips to God's ears. But I don't think so. It was 10% at the end of the year, so call it in December. We expect -- look, we don't control all of that. Some of that is dependent on our customers dedicating the resources and having the time to allow us to make those conversions and migrations. We expect that to rise meaningfully over the course of the year. And I think as the year goes on, we'll certainly be reporting what that number is. I hesitate right now to give you a target for year end.
We are starting with our health systems first. And then we'll move to work on migrating our health plans after that. So we certainly won't have the 100% because we're not really working on 100% immediately. But we do expect to have the lion's share of the visits for our health systems taking place on Converge by the end of the year.
Ido, maybe you want to comment?
Dr. Ido Schoenberg
Yes, absolutely. I mean, in essence, everything we need to win our entire market share in a way of R&D and capability will be there this year. And even towards the end of the year. But in the first half and then completing it during the second half. As I mentioned earlier, the demand for Converge is very, very high. People really want to upgrade to Converge. As a reminder, it's included in their license. So there is no financial barrier to do that.
However, there is an operational time-related to the barrier in the sense that it needs to be prioritized and deployed. We feel that we are going to, as Bob mentioned, have the lion's share of our customers on the platform. But it's not an exact science, and we fully expect some to linger towards '23, although it's really difficult to predict it exactly how it's going to fall out. We're then going to sense on our existing platform, of course, benefit not only from the efficiencies of Converge, but much more importantly, from the great expansion opportunities for so many models and programs that the new platform brings.
Okay. And then just one more quick one. With the 20% annual revenue growth, what portion of that will be organic versus acquired in your view? And it seems like there's plenty of assets that could add a lot of value to your customers. Any certain areas that you're focusing on?
Yes. That would all be upside to the degree that we are successful in finding assets as terrific as SilverCloud and Conversa, that would be incremental revenue in this model.
Your next question comes from the line of Jailendra Singh with Credit Suisse.
I actually wanted to get your thoughts on virtual primary care opportunities we have been seeing a lot of activity from health plans and other virtual care vendors in that area. Just trying to understand what traction have you guys been seeing? How big of a component is that product in your '22 guidance as well as your long-term revenue projections?
Dr. Ido Schoenberg
Jailendra, that's a great question. Virtual Primary care is really part of digital first. It's the assumption that patients are going to interact with the health care system by going online is almost always and certainly first. That's part of the heart of Converge. So it's not a product on Converge. That's the way Converge is built. We introduced virtual primary care last year, as you remember. We're selling it very successfully now in the market as part of Converge. I don't think that we expect almost any client not to use that capability. We think that it's an essential part of any type of digital care delivery in the future. And we see it as core. We are not going to report any as a separate product.
Okay. And then my follow-up around the telehealth reimbursement environment for providers. Clearly, despite the bipartisan support on telehealth providers across the country are still waiting for clarity on the reimbursement in post-pandemic and public health emergency once that is over, just trying to understand what is reflected in your long-term outlook of $500 million with respect to telehealth reimbursement environment and implications it might have on the next leg of growth in telehealth adoption or exploration among providers.
Dr. Ido Schoenberg
Yes. So it's very difficult to estimate the size of the change and the speed of the change. As such, we use fairly conservative assumptions to drive our model. It's also difficult to really understand the change between free-for-value and free for -- fee-for-service to fee-for-value. When you're value-based organizations, all the many modules and functions of Amwell that Converge that drive efficiency are becoming very, very relevant for any type of organization, in many cases, in a way that is unrelated to reimbursement.
If you're able to process people in your ER in a better way to avoid the readmission, to bring specialists or be very healthy therapies to different points of care, that's good even in different reimbursement environment. So I would say that the main purpose of Converge is really improve the efficiency of care, improve financial and clinical outcome. Of course, if there's going to be more direct reimbursement, not only for virtual visits, but also for digital therapeutics which we begin to see, that's going to put a lot of fuel in the fire.
But as you rightfully mentioned, we certainly are not smart enough to guess exactly how this is going to play out. Although I would only suggest that because there are so many proof points that financing virtual interventions is so much -- selling so much that we believe that it's likely that the different payers, and certainly, the government will end up reimbursing almost all.
Your next question comes from the line of Jessica Tassan with Piper Sandler.
So I wanted to follow up a little bit on SilverCloud and Conversa. Can you guys just maybe remind us how SilverCloud and Conversa are build, whether they're 100% subscription revenue streams and are they kind of enterprise SaaS contracts or build PMPM or per purchases per month?
Conversa from a revenue and customer size, average revenue per our ACV, I should say, and type of customer, very similar. All -- and it is virtually all subscription revenue. SilverCloud, is a little bit different from a customer size perspective. They still have a subscription revenue that we would characterize as subscription revenue. And it would be primarily in the health system category. So that's how I'd be thinking about categorizing them.
That's really helpful. And then just as a quick follow-up, when we think about Conversa specifically, can you help us understand how this virtual triage solution is kind of competitively differentiated relative to some of the other offerings on the market? What are kind of the key value propositions? And is it integrated within Converge?
Dr. Ido Schoenberg
It doesn't really count like one question, but I'll treat it as such. The Conversa is being implemented as part of Converge. It's going to really become the core of our offerings. As I mentioned in my prepared remarks, we have over 170 different programs that were tried and tested to automate certain workflows that, of course, can be described as part of the virtual primary care experience or in other scenarios as well.
Some of the differentiation of Conversa to other programs is its open nature. It is not only used as a tool that provides the actual programs, but it's a very powerful offering. So we can take programs that exist and amend them to the needs of your organization. And in some cases, create programs, new programs altogether. So there is a library that is ever expanding that is really customized to the need of every customer. And that's very important for many of our clients, especially the bigger ones, that has a very strong opinion about how workflow should be managed, how triage should be managed, litigation, steerage and things of that nature.
There are no further questions at this time. Thank you for joining. This concludes today's conference call. You may now disconnect.