athenahealth, Inc. (NASDAQ:ATHN) Q4 2017 Earnings Conference Call February 2, 2018 8:00 AM ET
Dana Quattrochi - Director, IR
Jonathan Bush - Chairman, President & CEO
Marc Levine - Executive VP, CFO, Treasurer & Principal Accounting Officer
Robert Jones - Goldman Sachs Group Inc.
Ross Muken - Evercore ISI
James Stockton - Wells Fargo Securities
George Hill - RBC Capital Markets
Donald Hooker - KeyBanc Capital Markets
Sean Dodge - Jefferies LLC
Steven Halper - Cantor Fitzgerald & Co.
Mohan Naidu - Oppenheimer & Co.
Nicholas Jansen - Raymond James & Associates
David Grossman - Stifel, Nicolaus & Company
Sean Wieland - Piper Jaffray Companies
Glen Santangelo - Deutsche Bank AG
Jeffrey Garro - William Blair & Company
Charles Rhyee - Cowen and Company
Matthew Gillmor - Robert W. Baird & Co.
Welcome to the athenahealth Fourth Quarter and Full Year 2017 Earnings Conference Call. As a reminder, today's call is being recorded. At this time, I would like to turn the call over to Dana Quattrochi, Executive Director of Investor Relations for athenahealth. Please go ahead, Ms. Quattrochi.
Good morning, and thank you for joining us. With me on the call today is Jonathan Bush, our Chief Executive Officer; and Marc Levine, our Chief Financial Officer. On today's call, Jonathan and Marc will share brief highlights from the prepared remarks we published yesterday, and then we will take questions.
I would like to remind everyone that certain statements made during this conference call are forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 as amended and Section 21E of the Securities Exchange Act of 1934 as amended. All statements other than statements of historical fact made during this conference call are forward-looking statements, including statements regarding management's expectations for future financial and operational performance and operating expenditures, including the implementation and impact of our strategic plan, including our savings; changes in leadership, including the impact of our new Chief Financial Officer; changes in our market; our goals and strategic objectives; our position for the future; our ability to drive profitable growth and enhance shareholder value; and statements regarding the focus of our business and the impact of actions to improve our business.
Forward-looking statements may be identified with words such as will, may, expect, plan, anticipate, upcoming, believe, estimate or similar terminology and the negative of these terms. Forward-looking statements are not promises or guarantees of future performance and are subject to a variety of risks and uncertainties, many of which are beyond our control, which could cause actual results to differ materially from those contemplated in these forward-looking statements.
The risks and uncertainties include those under the heading Risk Factors in our most recent annual report on Form 10-K and other periodic reports filed with the Securities and Exchange Commission, which are available on the Investors section of our website at www.athenahealth.com and on the SEC's website at www.sec.gov. Forward-looking statements speak only as of the date hereof. And except as required by law, we undertake no obligation to update or revise these forward-looking statements.
Finally, please note that on today's call, we will refer to certain non-GAAP financial measures in which we include certain noncash or nonrecurring items, such as stock-based compensation, from our GAAP financial results. We believe that in order to properly understand our short-term and long-term trends, investors may wish to consider the impact of these items as a supplement to financial performance measures determined in accordance with GAAP. Please refer to yesterday's press release announcing our fourth quarter and full year 2017 results available on our website, www.athenahealth.com, for a reconciliation of these non-GAAP performance measures to our GAAP financial results.
With that, I'll now turn the call over to Jonathan Bush.
Thank you, Dana. Good morning, everyone, and thank you for joining us today. Last night, we announced our Q4 results. Our performance reflects the metamorphosis we are going through as an organization and our commitment to drive more profitable growth. Revenue came in at the high end of our guidance and we exceeded the profitability target we set for Q4. We're executing remarkably well on our plans to produce a leaner, more efficient growth company. At the same time, we need to do a better job driving demand for our services, enhancing our product market fit and improving client satisfaction and retention.
Before I discuss the quarter in more detail, I would like to introduce Marc Levine, our new Chief Financial Officer. Marc joined athenahealth just over a month ago, and I am thrilled he is here with me on the call today and part of our team. Marc brings significant experience driving improved operational performance across complex growing organizations. His skills and expertise will enable an enhanced level of financial and operational leadership and rigor that will better position us to go after our highest return opportunity, expand upon the value we bring to our clients and create real shareholder value. I know Marc will be a valuable partner to me and to the rest of the leadership team.
Turning to our financial and operational highlights. We closed out the year with a solid fourth quarter. Total company revenue came in at $329 million, an increase of 14% year-over-year. Non-GAAP operating income grew 83% year-over-year to $78 million during the fourth quarter. Our Net Promoter Score improved sequentially to 33.3 in Q4 and was up 39% year-over-year. We also delivered a strong growth across our network. Compared to Q4 of last year, the number of providers on athenaCollector grew 15%. Covered lives on our population health platform were up 48% and hospital discharge bed days increased approximately 350%.
Despite our solid financial results and the network growth during the quarter, there are areas where we know we need to do better. First, we fell short of our bookings goals for 2017. Our industry is in the midst of a sugar low after years of rapid growth, driven by government stimulus. Demand for our services has been weaker than we expected this past year. We added $293 million of bookings for the full year, including $250 million worth of athena-branded bookings and $43 million worth of Epocrates-branded bookings. Despite these results, we had one of our highest win rate years ever with independent medical groups during the year.
While we're winning in an increased percentage of deals we're in, we're seeing less buying activity among our prospects overall. As we evolve to fit our products to the needs of our current market, we remain confident in our ability to draw more prospects into market and connect them to athenaNet. Medical practices and health systems are facing increasing financial pressure and the need to dramatically cut administrative costs. Our capital-free and cash-producing business model is a compelling solution to help them thrive in today's challenging environment.
Second, client retention decreased 200 basis points year-over-year to 92% in 2017. While we had anticipated this decrease after the rollout of our new athenaClinicals experience and a new low point in our customer support during 2016, we're not satisfied with the results. We know where we need to do better and we're already taking action to improve client retention. In 2017, we introduced a new release cadence. We now deliver new functionality three times a year to enhance our alpha and beta processes and provide smoother release processes for our clients.
We also committed to tightening our customer service levels and drove better performance on metrics that measure tactical support we provide for clients than we ever have. In our customer support center, we improved both average speed to answer and complex case resolution rates by approximately 50% year-over-year. And client feedback on athenaClinicals has also improved. Providers now tell us they find the new Clinicals experience easier to use and like it better than before. In 2018, we plan on building this process to stabilize client retention by adding product depth to the enhanced customer user experience.
Let me update you on our key strategic objectives. We entered 2017 focused on four things, deepening our services, executing in the small hospital market, building out nationwide connectivity for our network and investing in our platform. During 2017, we began to reorient our services towards increasing the value we deliver to our clients and improving the provider experience. Our clients are asking us to do what we do better. This past year, we analyzed and focused on operational work that we could remove from the medical practice and further reduce the administrative burden on behalf of our clients.
We focused on areas such as overpayments, coding-related denials and patient insurance-related denials. We launched our first machine learning model to automate how faxes are processed and analyzed. And we piloted services like authorization management, which we plan to expand within our client base throughout 2018. We expect the groundwork we laid in 2017 to further shift the burden of millions of tasks off our clients' plates and position us to go after the long tail of micro aggressions against the practice of medicine that we also abhor.
We also continued to make progress in the small hospital market. We view 2017 as a foundational year in our effort to serve rural hospitals, disrupt the traditional software vendors and build a scalable, network-centric hospital service. We exited 2017 with 62 hospitals fully live, nearly doubling the number of hospitals live on our network over the past year. The activity on our network also maintained strong momentum in the fourth quarter with discharge bed days up 38% sequentially and up approximately 350% year-over-year. We're balancing our opportunity to rapidly grow in this market with our objective to build a modern, scalable hospital service. In support of this effort, we enjoy great fellowship, alignment and partnership with our hospital clients and our hospital Net Promoter Score continued to trend well above our corporate average during the fourth quarter.
We also advanced our position as the most universally connected health care network in the country. During 2017, we launched and rapidly expanded our patient record sharing capabilities with CommonWell and Carequality. Patient record sharing helps providers exchange patient records with outside care sites and use those records to positively impact patient care. Built directly into athenaClinicals, patient record sharing automatically searches the growing network of care sites on CommonWell and Carequality to locate our clients' patient records and display them directly into athenaNet.
Today, our clients are connected to 1/3 of all Cerner sites and over 3/4 of all Epic communities. Nearly 40% of our clients are benefiting from patient record sharing by exchanging clinical documents with care sites on both CommonWell and Carequality. We expect our connections to continue to expand as additional Cerner and Epic clients join the network. This is the beginning of a true interoperation. With access to this data, we're focusing on servicing the right patient information to the right provider at the right time. This will not only enhance the utility of the connections we're building but also enable athenaNet to truly eliminate the clipboard and create a much better experience for patients and providers.
In terms of our platform strategy, 2017 was a year of platform investment and evolution. As I mentioned earlier, we introduced a new release cadence to deliver a smoother release process for our clients. We also invested time to optimize our database architecture and code so that we deliver enhanced performance and utilization at a better operational cost to our business.
And as part of our evolution to a microservices-based architecture, we developed the first microservices in 2017, including our calendar service and provider directory service. Just as other leading tech companies that have made this transition in industries outside of health care, we believe this is a key step for us on our journey towards becoming health care's first true platform.
With that, I'd like to turn the call over to Marc to review our financial results in more detail.
Thanks, Jonathan. I'll start by saying that I'm thrilled to be part of the athenahealth team. This is a special company driven by our shared commitment to make a difference in the way health care is delivered. It's a company that offers a compelling vision and has the capabilities to combine that vision with innovative technologies to solve problems, deliver insights and open access across the health care network. These opportunities for future growth as well as the company's commitment to execute with greater discipline, clarity and consistency are what attracted me to athena.
Over the past month, I've had the opportunity to meet with people across the company, experience our capabilities and witness the shared passion, commitment and focus of the team to transform health care. I look forward to working with Jonathan and the rest of the leadership team to deliver on our commitments and take athenahealth to the next level. I also look forward to meeting with many of you in the weeks and months ahead. With that as background, I'll begin the financial portion of today's call.
Our fourth quarter revenue was $329 million, an increase of 14% year-over-year. athenaOne revenue grew 16%, driven by a 15% year-over-year increase in the number of athenaCollector providers. A decline in Epocrates partially offset the growth in athenaOne. For the full year 2017, revenue was $1,220,000,000, an increase of 13% versus the prior year and in line with the high end of the guidance from October.
During the quarter, the company executed on its cost reduction plans. The majority of the planned Q4 actions taken by the company were related to workforce reductions. These have been substantially completed. In addition, we made progress on our plans to reduce our real estate footprint exiting our San Francisco office in early 2018.
However, the fourth quarter was also favorably impacted by the organization aggressively acting on short-term discretionary spend items. The business saw reductions in short-term marketing spend, third-party activity as well as from greater rigor in the application of basic company practices. The organization's ability to execute on what we said we would do is one early data point toward better, more disciplined execution.
To realize the planned savings, we recorded a restructuring charge of $19 million during the quarter. Following the progress of the fourth quarter, we're reiterating our commitment to eliminate the previously communicated $100 million to $115 million of gross annualized run rate spend by the end of 2018. The execution of the cost reduction plan and focus on expense management drove strong profitability during the fourth quarter.
On a consolidated basis, GAAP gross margin for the fourth quarter was 55.6% as compared to 53.7% in the same quarter of last year. For full year 2017, GAAP gross margin was 52.6% compared to 50.7% in 2016. Our service automation rate, formerly referred to as non-GAAP adjusted gross margin, was 68% during the fourth quarter as compared to 66.2% in Q4 of last year. And for the full year, the service automation rate improved by 50 basis points to 64.6% from a rate of 64.1% in 2016.
Looking below the gross margin line. GAAP selling and marketing expenses decreased $7 million year-over-year to $60 million in Q4 2017 and decreased $4 million year-over-year to $252 million for the full year. Compared with the same quarter last year, GAAP research and development expenses were flat during Q4. On a full year basis, GAAP research and development expense increased by $39 million versus 2016, driven primarily by our platform strategy work and investments in our core ambulatory and hospital services. In 2017, we also had a higher mix of expensed versus capitalized research and development projects versus the prior year. On a cash basis, research and development spend as a percentage of revenue came in at 15.4% in 2017 compared to 15.7% in 2016.
Lastly, our GAAP general and administrative expenses in Q4 increased $10 million year-over-year to $41 million, and for the full year 2017, increased $14 million year-over-year to $145 million. This increase in both the fourth quarter and the full year was driven primarily by higher advisory fees related to the strategic assessment of efficiency and effectiveness opportunities in the business.
We generated GAAP operating income of $39 million during the fourth quarter and $71 million for the full year 2017, reflecting initial improvement in operational execution. This compares to $12 million and $27 million during the fourth quarter and the full year 2016, respectively. These operating income results include the approximately $19 million of pretax restructuring charges I referenced a moment ago.
On a non-GAAP basis, fourth quarter operating income increased 83% year-over-year to $78 million. For the full year 2017, non-GAAP operating income grew to $175 million, representing a 32% increase over the full year 2016. Driven by the expense savings initiatives and top line growth, non-GAAP operating margin for 2017 improved 210 basis points year-over-year to 14.3%.
We exited the fourth quarter with GAAP net income of $32 million or $0.78 per diluted share. This compares to $10 million or $0.24 per diluted share in Q4 of the prior year. On a non-GAAP basis, we generated net income of $45 million or $1.11 per diluted share during the fourth quarter, up from $25 million or $0.62 per diluted share in the same period last year. On a full year 2017 basis, GAAP net income was $53 million or $1.31 per diluted share versus $21 million or $0.52 per diluted share for 2016. Non-GAAP net income for 2017 was $101 million or $2.48 per diluted share versus $76 million or $1.90 per diluted share in 2016.
Turning to cash flow. For the full year 2017, we generated $241 million of operating cash flow and invested $83 million in capitalized software and $80 million in capital expenditures. As a result, operating cash less total capital spend came in at $78 million versus $24 million in 2016.
In terms of our efforts to improve profitability, we remain committed to achieving at least the 15% non-GAAP operating margin we had previously communicated. This target does not take into account the impact of accounting pronouncements not yet implemented, like Topic 606. We plan to introduce our fiscal year 2018 guidance at our upcoming Investor Summit on February 15.
With that, I'll now turn it back over to our conference call moderator for Q&A.
[Operator Instructions]. Our first question comes from Robert Jones with Goldman Sachs.
Jonathan, your comments on the industry going through what you refer to as a sugar low is certainly consistent with the message that you've been sharing throughout 2017. And obviously, there's some things you've talked specifically about the company needs to do better to drive bookings. But I was wondering if maybe you could just take a couple of minutes to give us a sense of how much of your ability to drive better bookings growth you feel is within the company's control versus maybe just a total lack of demand right now in the end market.
It's a very good question. Certainly, when we were positioned to be the best EMR or practice management information system, there's nothing we can do to grow demand because by that definition, everybody has got one and, by and large, it's a pretty new one because everybody was kind of forced to buy one during the HITECH Act. What we were before the HITECH Act and what we are at our core and always will be and what we are moving to being is a service -- a national network that provides a set of services. Most people don't have that and everybody needs that. Everybody will need to be on an integrated national network and have highly scalable services replace the very manual tasks that go on in hospitals and practices.
The economics of medicine just do not support individual kind of craft brewing of the back office of medicine, and it's getting worse. So we're really down to not whether but when. What we are doing to draw people into market is, as I have said, deepening the services so that you can see real cash more certainly sooner by switching. And so that's the work that has been going on in '17 and '16. And it will really hit the base in force during '18. We actually have things that we've been testing that we're doing on -- we do all the referral and authorizations now, we propose in all the small group prospects we have been. So the last 300 small groups to go live on athenaNet, they're not doing any of their authorization work. We just, yesterday, flipped over the last practice for overpayment handling, where all 6,750-odd practices now, whereas in '16 and through early '17, they had to handle overpayments.
Now we handle the overpayments. During '18, we expect to see some coding trickle into the base. We will have all of the referral and authorization get turned on throughout the base. Referral and authorization alone, according to our outside consultants we brought in to set this product up or this enhancement up, is an average of about half an FTE per doctor on athenaNet. So you think about the full cost of an FTE of a staff worker in a clinic or a hospital, that's about the cost of all of athena per doctor. And we believe we can get that cost out across the base. Now going on athena is not about, Gee, I need an EMR, who should it be? It's about, I need my administrative costs lower and my payment faster, how can I get that? And so that's what made us into the company that got venture capital, that got to go public, that got to have a great run before the election of Obama. And we know how to be that company, and that's what's going to pull people in sooner. So it's not whether but when. And I think we're doing a good job of making when now, given how early people's recent EMR investments are.
Our next question comes from Ross Muken with Evercore.
Can you give us a sense both in ambulatory and inpatient of sort of the competitive landscape in terms of we're seeing or hearing about a number of larger institutions as they merge? Think about sort of consolidating on to one product or transitioning parts. How are you thinking about that as a positive or negative relative to your business over the kind of the next 12 months, given the fact we're seeing kind of this elevated level of activity amongst the providers and physician practices?
I don't know that I would call it elevated. There's obviously one that everyone's very -- watching very closely as are we. But it's been going on throughout the HITECH Act as these additional administrative burdens get added to practices. And as admissions per 1,000 continue to trend gently downward, you have -- you look for scale. And merging is one way of getting scale if it's done properly. We win in those mergers when we make our case that athena produces more physician efficiency and more administrative efficiency than software. And when that's what they value, we win. When they value other things or when we don't make our case well, we lose. But we think our case in this front is made stronger by the fact that we now have the legal right to go to any hospital system where our clients are treating patients and insist that they connect.
So we are now able to create an integrated inpatient, outpatient view of the record. Clunky at first, getting slicker by the day, that kind of trumps that box-out, I love athena, but you've got to go because I need inpatient and outpatient on one computer system. No, you don't, you need inpatient and outpatient data in one chart, and we can now provide that thanks to a lot of really good R&D work over the last two years and a lot of really good legislative affairs work over the 3 years before that.
Our next question comes from Jamie Stockton with Wells Fargo.
Maybe just another cut at that last one. The 92% client retention rate, you guys obviously had the noise around Streamline. I feel like we've probably had maybe a couple of years of elevated enterprise churn. So could you give us a sense for what is a realistic target for the client retention number going forward?
Well, I don't know that we're talking about specifically 2018, which we'll share at the Investor Day. But I'll tell you viscerally what I've told you in these calls for years. I expect 5%. 5% is the normal course. Anything over that, I feel like we could get back, and then it's on us. 5% is normal. Doctors retire, mergers happen, politics -- we don't win, we don't get make our case, we make our case badly, we underdeliver. All kinds of things make it fine by me to have the 5% attrition that we've had for all 20 of our years, except last year. And we'll get into how it all fits together. As you know probably or as you can surmise, attrition is a lagging indicator. It relates to somebody getting mad enough or sad enough or getting bought far enough in the past that they could go out, buy a new system, set it up, train their staff, switch their payments over and ramp up. So if somebody left financially in 2017, it means they left kind of emotionally in 2016. And for the people who left emotionally in 2016, we are very sorry and we deserved it. And for the people who stayed, you will not regret it.
Our next question comes from George Hill with RBC.
And Jonathan, nobody likes to be left emotionally. I guess, as we think about the bookings weakness, can you talk about maybe gross to net? And is the weakness year-over-year? Is it more a reflection of the attrition that we saw and gross bookings were strong? Or is the net number year-over-year was just basically kind of reflective of the environment?
I'm getting these faces that say. We don't talk about that. Is that right?
Yes, we don't. It's just bookings, net bookings is what we discuss.
So when we said bookings, it was after all the netting out of people who charge-back or anything else. So we did all the bad news for you. I suppose I could give you this color because we've talked about it. In new markets, in a new specialty, in a new adjacency like hospital, we have tried it both ways. We've tried going for getting everyone no matter what. And we've tried going for those where we could win and build a scalable service that thousands and thousands more could use as well. And the latter is better -- we're better suited to the latter. So we have been very disciplined in the hospital market to not take live folks that we were worried might be disappointed. And that means if they're sold and they want to go, we say, Well, we can't get you, let you go until next year when this thing that you know how to do, athena does better. They would charge-back. So they would be someone who got sold but then wouldn't show up in bookings because we both agreed to either wait or they agreed to walk away. And that with the [indiscernible] of hospital, it happens a lot.
Our next question comes from Donald Hooker with KeyBanc Capital Markets.
Just want to maybe get a view from maybe Jonathan from the top level around Epocrates. And why are revenues there struggling in your opinion? I would assume there's a lot of pharmaceutical products coming to market right now. And I would think -- I would speculate that there might be a robust environment for advertising potentially. But maybe can you talk about trends at Epocrates and what you think is going on?
We are, for the first time, really excited about the core product on Epocrates. We've always been excited about the promise of Epocrates. But the platform -- literally, we still had, as of last year, a Palm OS, remember the PalmPilot, server running somewhere in the bowels of Epocrates. We bought Epocrates for the network. We believe that increasingly network software will be easily grabbed, untethered mobile apps that replace enterprise systems that are, well, you know what they're like, they're big and cumbersome and expensive and they don't evolve well. And part of that would be to create something that disrupted our own medical records, some sort of Sonos personal app that could attach to whatever EMR you happen to be near. That's always been our vision for Epocrates. But it occurred to us after we bought it, that we would have to rewrite much of the stack underneath Epocrates.
So two years ago, we stopped trying to run it like a stand-alone business, moved ownership to our just greatest product managers. We have a lot of great product managers. We love all our product managers. These guys are pretty extraordinary, and moved to essentially rebuilding the core stack. This last Q4 was the first time since we bought it that we actually upticked in users. Average weekly users went up on the new stuff that's coming out. It's been a lean team, but the team is really tracking well. This year, we will begin to add untethered connections into athenaNet. You will be able to see a patient in Epocrates. You will be able to see a care plan for a patient in Epocrates. The same type of stuff you might see in a care manager dashboard of population health or in the quality section of an electronic record. That has never been true of Epocrates before. Unfortunately, you probably remember when I said that was my vision for the acquisition when we were both 16 years' old, shaving for the first time. And so I apologize for how long it has taken.
But under the leadership of Jonathan Porter and Jason Bornhorst, this thing is really coming around. And we will ultimately, despite the time in the time series here, consider it a very positive NPV acquisition. And we will see those ad dollars start to go up, too, right alongside the users going up.
Our next question comes from Sean Dodge with Jefferies.
So maybe on the cost initiatives and the sales and marketing rationalization, what I want to understand better is the balance there between reducing spend but doing it in a way that doesn't impair future revenue growth. So the actions you took in the fourth quarter, were there just some programs that weren't yielding any benefits, so easy decisions to shut down? Or are you finding more new lucrative sales channels you're getting better returns on? Anything around that would be great.
Yes. So you make probably the most important point of the call, which is that the profit level that we are at would not be a prudent profit level to maintain because of the size of the growth opportunity around us. We sort of thought at $1.2 billion, there's a base camp there, where with all of our fixed costs and everything, and we ought to able to contain a very scalable business at no less, as our previous CFO said on the previous call, than a 15% EBIT. And then a lot of profit ought to come down along the way as we grow. But we should be investing in growth. We did make growth a lot more efficient. As you know during the HITECH Act, there was this weird dynamic where everybody was going to be forced to buy and nobody had heard of athenahealth. And so we had the opportunity of a lifetime and the threat of a lifetime. And so we had to run out and frantically try to get our mug into every single doctor's office in America.
And we did not mind that it would be inefficient. Starting when Obama was elected, we would tell investors, we try to spend $0.50 to get $1 of recurring annualized revenue. By the time Trump was elected, I think we might have gotten to $0.90. But it was because we were trying to make sure that while this weird sort of forced open season was underway that we got more than our fair share, that we punched above our weight class. We waited on rearchitecting. We waited on -- we were capable over being under control because it was the right thing to do, given the regulatory environment. When that environment changed, our tone has changed. We literally -- aside from the cost, we had thousands of doctors put us on the do-not-call list because we were so annoying, forget the inefficiency. But we just needed to make sure that they knew that this stuff could happen on the Internet because they didn't.
And by the way, if you look at our -- and part of it was acquiring Epocrates. If you look at the name ID, the awareness of athenahealth between the beginning of HITECH and the end of it, you're talking about, can I say in order of magnitude, 7% to 70% name ID. So it worked. The environment is different now. We are spending a lot of money on replatforming, on microservices that are more scalable, on better instrumentation of quality measures and on a more intelligent sales journey -- customer journey. And that is going to make a much more -- even when we put a little gas on in this month here and start to make sure we hit our sales numbers, we're still going to be a much lower cost per $1 of annualized revenue booked than we were before our restructuring. Yes, but we're going to grow. We're going to keep growing, double digits.
Our next question comes from Steve Halper with Cantor Fitzgerald.
So relative to the bookings performance in 2017 and the headcount reductions, do you feel like you have the right personnel in place in order to get the bookings growing again in 2018? And then my follow-up question, have you worked on a tax rate assumption for 2018? I know that's a little bit for guidance purposes. But any information on that would be helpful.
Yes, I feel very good about the leadership team. We made a lot of hard decisions about the leadership team so that we would feel good about it. We had a window, a strong support from our shareholders to take a hard look at ourselves. And we, in retrospect, definitely needed to do that. And I feel extremely confident about the team we have down through every layer. So this is a company that's got 1 fewer layer or 0.7 fewer layers. It's got 9% fewer people. The managers that are in place have full swath of management responsibility and aren't multitasking across other duties. So from the very top to the very bottom, including, by the way, I wish I could be talking about our Chairman search, which is almost over and we'll be getting back to you on that really shortly, including our board.
Yes. And I'll comment on the tax side of your question. So you're right, we intend to talk about the tax impact in the guidance that we offer when we get together in a couple of weeks. But I think it's fair to say that obviously the company hasn't really been a taxpayer, given the lack of profit and also the NOLs that the company has carried for a while. And that's going to be factored into -- and we'll share with you, that will be factored into the assumptions going forward.
Our next question comes from Mohan Naidu with Oppenheimer.
Jonathan, I just want to hit on the bookings question one more time. The shortfall in the bookings, is it fair to say it's predominantly on the ambulatory segment? And on the bookings on the small hospital segment, is the pipeline strong enough to offset some of that and hold you until the ambulatory weakness shakes off a little bit?
No, I would almost go the other way around. The core ambulatory business did very well on all fronts. If you look at group practice, or large group practice, they grew much faster than our average across -- in fact, the large group team beat their numbers soundly on bookings and revenue and attrition. What we had over the years between kind of beginning as a whisper in '14 and ending with a roar in '17 is the creep of not fully scalable new products being assigned big numbers, not fully scaled up new market segments being assigned big numbers and then those missing. Or they hit, but what goes into implementation crumbles and attrits and isn't good.
So one of the things that we started with Jack Kane and are persisting, we will obviously persist with Marc Levine, is to not let that stuff count as bookings. Sell it, install it, let it grow, and when it's a real business, start to try to communicate on it and manage it. So most of the stuff that went away were the mega deals, the new products, the emerging stuff, a lot of the bookings miss was related to our -- during the course of the year and during our restructuring, tightening up the entryway into the hospital service so that only hospitals that were a really tight fit for what we could do well today got through. So I think we've allowed a slightly inaccurate sort of market assessment that there's no more growth left in ambulatory, where we have 12% of doctors, and that we need hospital to be able to be a fast grower. We obviously need every vertical. We need labs and pharmacies and ambulances and hospitals and surgical centers and nursing homes and skilled nursing facilities and visiting nurses, and all of them need athenaOne.
They all need a mobile cloud-based connected service that allows them to know what's going on with their patients and get rid of the crap that abrases their life and gets in the way of their clinical life. So we need them all. We need them in order. We need them to come in, in a way that is scalable. And we need to talk about them when we're ready to talk about them. And I think I confessed to a little bit of denial that we would ever not grow 30% organically in a year and just kept whipping the team in a way that what -- that did not allow for the sophisticated, very hard to compete with, very financially forgiving for both customer and seller cloud-based service concept to mature. So what we're talking about now with you all is clean, ready, bookable, unlikely to attrit growth. And we're working on the other stuff. But we're working on it in an ordered way so that when it gets to maturity, we can count on it and talk to you about it in ways that you can count on it. So I would sort of flip your concern.
Our next question comes from Nicholas Jansen with Raymond James.
Just in terms of the macro environment. Do you think 2017 is the trough where in terms of inertia from the ambulatory marketplace or in things like MACRA and MIPS in '18 and '19 can help reaccelerate? Or do expect this 2017 time frame, where we've seen this lack of buying activity to continue into the intermediate term?
I certainly do not think that MACRA and MIPS will drive any market activity. The dollars in play for MACRA and MIPS are incredibly small and they are delayed. If you think about change management, right, you want childbearing, dog raising, sales promotions. You have size and speed. If you do this, we'll give you, no questions asked, on the barrelhead, rebate, whatever. MIPS, you get the check something like 2 years after you make the move and the check is smaller than cable bill. So it's just not going to move the needle. And I don't think the current administration is going to bring all of Harvard down to Washington and invent a new farkakte scheme for us to go all report against. I think what's going to drive people back into market is the drip, drip, drip of financial pressure. I think I saw it at JPMorgan. I have seen it in the calls and earnings releases of our publicly-traded clients and prospects.
The ability to eliminate a whole lot of administrative cost is a really, really valuable arrow to have in your quiver. And then obviously, right after that, right after we get the full value of the administrative cost work that we've been inventing for the last 18 months that's going to be going into the network this year, we're going to start talking really aggressively about market share. Are you connected to Apple? Are you connected to Amazon? Are you connected to the Walgreens app with 23 million Americans having a password at the backdoor of the Walgreens app and a telemedicine button waiting for them on the other side. Are you connected and serving Teladoc? Are you out on the front end of an increasingly shop-aware health care consumer? And we are. We can do things in that realm that enterprise software companies can offer tools to do. And the IT department at the hospital can hire a bunch of people and try to get the very best and brightest of social media to work at the local medical center. But we just have a very, very strong competitive advantage over that approach by being able to marshal the sheer scale of appointments that we have to fill with our client base and patients that we can access to fill them.
So it is the growth uptick, the renaissance of our -- the return of our growth rate to levels that excite me. And you -- well, it definitely won't be in '17. It's related to replatforming, that financial pressure building and our ability to pull that cost out and bring in patients. And you'll see evidence of that, you'll watch for it and you'll see evidence of that each quarter here during this year as we push these things out.
Our next question comes from David Grossman with Stifel Financial.
Jonathan, could you just clarify your bookings comment you just made? Was a big part of the shortfall, in fact, just tightening up the definition of what a booking is, which drove perhaps a larger-than-normal reset in 2017 but should theoretically drive better bookings realization on a go-forward basis?
We absolutely expect -- I'm not sure you would be able to calculate it. But the relationship between a booked dollar and a revenue dollar, we expect -- was definitely part of the restructuring effort, is a huge area of focus. It is absolutely the case that the bookings number that you saw for 2017 was aggressively netted out. In fact, in some ways more, the rules were applied more aggressively and more rules were applied on that front than ever before. And it is very much our intention to use this hiatus, this restructuring, to restructure all of that. There has been, as I hope it's true in lots of companies, a little bit of a drip, drip, drip in the definition and the enthusiasm and in the tightness. I will never forget the Bausch + Lomb case about where they had to hit certain numbers, and there were people moving boxes across yellow lines on warehouse -- on shipping docks and calling it a booking.
I never want to get to that place. We're nowhere near that place. But I want to be more aggressive than anyone has ever been on that stuff so that it's clean. We really want to make sure that nobody is under pressure ever to meet a non-real thing. So yes, what you saw in the end of the year, we could have easily, by definitions used in the past, crossed over into the bottom end of our range. And nobody felt like it was a good idea.
Our next question comes from Sean Wieland with Piper Jaffray.
So you mentioned that athenaNet is now connected to 1/3 of the participating Cerner sites, 3/4 participating Epic sites. So what does that really mean? How many sites does that--
It's actually 100% of the participating. It's just that they're not all participating. Now they have to participate. So over time, what it means is that those are the sites that have either subscribed to CommonWell or Carequality and allow our widget to crawl through their patient directory looking for matches. There's a -- CommonWell sort of provides the key. Once you're a patient on a CommonWell site, you get a CommonWell key. If I get the key, we can match. And now I can pull out a continuity of care document out of the medical record system of that hospital and pull it back over -- in our current form, it shows up as a tab, Hey, I found this person's chart down the road or across town or whatever it is, across the country. And so you can hit that tab next to the regular flow sheet and face sheet in athenaClinicals and see in a moderately well structured way what you would have seen if you logged into the local hospital's emergency room or whatever about that patient.
The next task is to tease apart that data, find pieces that just we should add to the face sheet and reconcile for the doctor and separate it out from stuff that is unlikely to be seen, but it's nice to know you can go look for it. And ultimately, we'll build an increasing set of APIs, where doctors can just issue orders and interact with hospital systems out of their athena screen without ever knowing what system the hospital is on. We've got that running with a small number of Cerner sites as a result of our acquisition of Praxify. There's an engine that they spent 8 years on that we're very excited about that allows the doctor in athenaClinicals mobile to issue orders, not just see the chart and have an integrated view of the patient, but actually act like they're standing on the hospital hall from their living room couch inside of their athena view. So it's a slow build, starts with get access, right? You know our brand, right? Open the network, step one. Step two, multiply its intelligence. And then step three, free people to do what matters. So that's where we are. We expect -- any hospital we really point our energy at kind of has to do something. And so it's just a question of getting through the queue of hospitals.
Our next question comes from Glen Santangelo with Deutsche Bank.
Jonathan, I just wanted to ask you sort of about the bookings conversion rate in 4Q '17. I mean, given everything that's sort of going on with the sales and marketing infrastructure changes you're making, are you seeing that have any impact on your conversion rates?
So our close rates, so meeting to win, initial meeting to win, were higher than ever. There may have been some quarter that I'm missing. But certainly, this feels like an all-time high in terms of close rates. The number of proposals that went out was not at an all-time high. And then the number of really big enterprise deals was at an all-time low. So the conversion rates look good. It's the ability to get somebody to believe that athena represents not just an EMR choice but a financial and clinical performance choice. And we think we can do that, but that's still to be proven.
Our next question comes from Jeff Garro with William Blair & Company.
I want to ask about the hospital space. And I see that hospital customer NPS is very favorable while the hospital performance metric is trending unfavorably. So want to ask, is the hospital performance metric entirely attributable to the mix of new customers? And how should we think about the pacing of implementations or the backlog of hospitals already signed and to be implemented going forward, given your new focus on the success of those customers?
Yes, so it's a good question. Net Promoter Score is high and satisfaction is high. The number of hospital types that we will serve is lower than we expected. So that's why we have 62 hospitals live and not 162. The performance metric you referred to is the percentage of hospitals that get to 104% of their previous year's cash flow. If you look at it at an average, right, we actually average at 106.9%, but the way...
Live a year.
Yes, at live a year. I thought that's what I said. And the problem in the metric is that we look at the percentage nominally that gets there, which is a more aggressive way of looking at it. We feel like the value is there on the collection side. And certainly, that's what our hospital clients feel like. In a market where hospital collections are painful, to average 106.9%, that feels right. What we need to do in hospital is get some of the tasks that hospital staff still has to do out and on to athenaNet, and that's underway, coding, for example. And we have to get the number of hospital types that we are willing to serve up. And we are being judicious about that. But we know we'll do it eventually, but we don't feel like we're in a rush to do with ahead of the strength of the platform.
Our next question comes from Charles Rhyee with Cowen.
Maybe just to follow up actually maybe to Glen's question earlier, obviously you're talking about strength in the win rate when you're getting deals. But can you talk about sort of the implementation time lines of your backlog and how that's trending? So obviously, we had sort of weaker-than-expected bookings this year. But is the mix in the bookings such that we can ramp that up and then get them live in a relatively short period of time? And then maybe following to that, the last question was has your inpatient strategy changed at all, given that we now have greater interoperability and you're now able to give that sort of view of the patient record for the inpatient, outpatient? Does that change sort of the focus you might have had with inpatient versus a few years back, when you bought a bunch of assets and you're trying to build out perhaps a much bigger hospital system? Do things change within your interoperability [indiscernible]?
Well-called, Charles, well-called. That's exactly right. So the inpatient movement was two things to us, right? It was a new line of business, an adjacent market that could reuse athena data synergistically. And it was an answer to an existential threat that we call the box-out. I'm not sure if we use that term with you, but we use it internally all the time, where the customers like us more, the doctors like us more, we're a better financial deal and we're fired anyway because the buyer is the hospital and the hospital believes they need inpatient and outpatient on one computer system. And what we need as an answer is to be able to show the doctor or the inpatient, outpatient, skilled nursing, pharmacy, surgical center, everything on one screen, more so than just inpatient and outpatient as our answer. Initially, we were blocked out. People wouldn't connect to us and they were allowed to not connect to us. Then it was illegal to not connect to us, but the technology was weak.
Now you've got the law, the mandate, you've got real traction. The box-out is still a very big deal, probably our biggest deal but is a much smaller deal quarter-by-quarter than it was and a fraction of the threat to us than it was two years ago, when we entered the hospital market. So it's absolutely the case that the strategic intent of the inpatient business today is now purely leg 1, right, and not leg two. It's purely to do -- to build a wonderful new area of growth and synergy for our platform and our Network Services. And it is not a critical short-term weapon in blocking the box-out, which is why you hear us talking about scalability and discipline on who we serve so much. It does mean that you're going to see a better traction through the pipeline to live and better satisfaction, better financial performance. Quickly, just so you know, only 50% of the hospitals are over 104% of cash flow, which is why that number looks bad, even though the average of all them is that 106.9%.
Our last question is from Matthew Gillmor with Robert W. Baird.
I wanted to ask about utilization. That's been a headwind you've called out in the past. Do you have an estimate for how much of a negative impact that was for 2017 revenues? And then as you think about 2018, I know you're not giving guidance, but just curious if you have any perspective on utilization in general terms. I was going to say, is this a new normal? Or do you think this stabilizes and that headwind goes away?
Yes, we did see a -- I think you could see in the -- I'm not sure if you could see it. But I would say that we -- so we presented at JPMorgan through Q3 and we showed kind of a relative sizing of visits to the doctor over the years, and you saw the only really real red year, other than very early in 2009, kind of after the crash, was this year in terms of same-store productivity of the doctor. It got a little better in Q4. But to your final part of the question, I don't expect this to go away. I think that more and more Americans will have bigger and bigger deductibles. And they will be more and more shrewd about how they use the health care system.
So you'll see more of it back-weighted. One of our executives freely admits to have gone to children's hospital and demanded that they move his surgery. He's a millionaire, but he wanted that thing in 2017, not 2018, so he didn't have to pay the extra $1,000. I was like, God bless you, I'm glad you're working for me. But that's a wealthy guy. Can you imagine what the average American is willing to do? So you will see spike year Q4s and leaner Q1s as more and more Americans get hip to this phenomenon. I also think for some of the more sort of easily utilized things, ED visits, precautionary imaging, you're just going to see less. Look at some of the publicly traded ED-focused companies, these are well-run companies by top-flight teams that have been doing this for years. But when the tide goes out, the ship hits the bottom.
And I think all of that is part of a new normal. And by the way, while I feel for our client base because it's hard to fight in that environment, we can -- our utility goes way up. So the ability to actually be a part of going out on to the Internet and getting market share, coordinating care, staying in touch with the people who call you their patient and call you their doctor and making sure that they get their screenings, making it appealing and easy, if you look at the #1 arbiter of market share to date, it's access. And we've said this all the time, if you show them a really easy appointment that's convenient and on a day when they're not working or before or after work, et cetera, you are multiple times more likely to get that patient to do what they should do clinically than if you say, Yes, I think I can fit you in, in the middle of work in three weeks.
So there's a huge amount that athena can do to change the game here by nature of our business model that even the most wonderful enterprise software company just can't do. They can provide the tools and hope that each customer comes up with nifty marketing departments. But I emphasize that these areas of financial administrative pressure and need for new market share are areas which we expect to turn into tailwinds for athenahealth.
This concludes the question-and-answer session. I would now like to turn the conference back over to Jonathan Bush for closing remarks.
In raising anything that you care about, a family, a child, a company, there's nothing worse than that period of time where you know it's coming and they can't see it yet. I hope today was a day where you could see the things that we here can see coming around the bend. And I hope the time that it has taken shows both our willingness to execute aggressively and our passion for the business as well as our insistence on doing things well so that we build a company for the ages that take on a problem that is fundamental and existential to our society in a way that is lasting and irreplaceable. That's what we're doing. And I hope today, it started to show. Thank you all for your attention and for following this wonderful adventure.
Ladies and gentlemen, this concludes today's conference. Thanks for your participation, and have a wonderful day.