Good day, ladies and gentlemen, and welcome to the Autodesk fourth quarter fiscal 2016 financial results conference call. At this time, all participants are in a listen-only mode. Later, we'll conduct a question-and-answer session, and instructions will be given at that time. If anyone should require assistance while the conference is in progress, please press star then zero on your touchtone telephone to reach an operator. As a reminder, this conference is being recorded. I would now like to hand the meeting over to David Gennarelli, Senior Director, Investor Relations. Please go ahead, sir.
Thanks, operator. Good afternoon. Thank you for joining our conference call to discuss the results of our fourth quarter and full year fiscal 2016. Also on the line is Carl Bass, our CEO, and Scott Herren, our CFO. Today's conference call is being broadcast live via webcast. A replay of this call will be available at autodesk.com/investor. As noted in our press release, we have published our prepared remarks on our website in advance of this call. Those remarks are intended to serve in place of extended formal comments, and we will not repeat them on this call.
During the course of this conference call, we will make forward-looking statements regarding future events and the anticipated future performance of the company, such as our guidance for the first quarter and full year fiscal 2017, our long-term financial model guidance, the factors we use to estimate our guidance, including currency headwinds, expectations regarding our restructuring, our transition to new business models, our market opportunities and strategies, and trends for various products, geographies, and industries. We caution you that such statements reflect the best judgment based on factors currently known to us and that actual events or results could differ materially.
Please refer to the documents we file from time to time with the SEC, specifically our Form 10-K for the fiscal year 2015, our Form 10-Q for the periods ended April 30, July 31, and October 31, 2015, and our current reports on Form 8-K, including the Form 8-K furnished with today's press release and prepared remarks. Those documents contain and identify important risks and other factors that may cause our actual results to differ from those contained in our forward-looking statements. Forward-looking statements made during the call are being made as of today. If this call is replayed or reviewed after today, the information presented during the call may not contain current or accurate information. Autodesk disclaims any obligation to update or revise any forward-looking statements.
We will provide guidance on today's call but will not provide any further guidance or updates on our performance during the quarter unless we do so in a public forum. During the call, we will also discuss non-GAAP financial measures. These non-GAAP measures are not prepared in accordance with generally accepted accounting principles. A reconciliation of our GAAP and non-GAAP results is provided in today's press release, prepared remarks, and on the investor relations section of our website. We will quote a number of numeric or growth changes as we discuss our financial performance, and unless otherwise noted, each such reference represents a year-on-year comparison. Now I'd like to turn the call over to Carl.
Thanks, Dave, and good afternoon, everyone. Let's start by wrapping up Q4. I want to take a few minutes to talk about long-term value creation for our shareholders, we'll end with guidance. As you saw in the press release, we had a terrific fourth quarter, our results demonstrate that our business model transition is working. The end of the fourth quarter marked a key milestone in Autodesk transition to our software-as-a-service business as we hit the end-of-sale date for perpetual licenses for individual products. While the end of sale went well, what makes our Q4 results particularly good is that they were not driven by a dramatic surge in last opportunity buying of perpetual licenses. In fact, our analysis suggests that only a small portion, around 10% of our Q4 unit volume, was related to end-of-sale activity.
More importantly, strong subscription growth, both new model and maintenance subscriptions, and a record number of large transactions drove these results. Our total unit volume for the fourth quarter was flat compared to the fourth quarter last year, which was a major accomplishment. Recall that in Q4 last year, we had a significant uptick in demand related to the end of our upgrade program. Said another way, we were able to cover last year's Q4 uptick in unit volume with our new model subscription offering. When we look at total unit volume for the year, we were pleased to see volume of approximately 650,000 licenses. The total of subscription additions for the quarter was better than expected at 109,000. As we've seen all year, the additions were led by new model subscriptions with strong growth in all 3 types, desktop, enterprise, and cloud subscriptions.
Desktop subscription remains our fastest-growing new model subscription offering and grew nearly 300% year-over-year. More specifically, desktop subscription is thriving with our channel partners. In just the past four quarters, the channel volume of desktop subscription has increased by more than 400%. In Q4, approximately 65% of our desktop subscription volume came through our channel partners, which is up from less than 50% in Q4 last year. What's also notable is that we achieved our highest quarter for new model subs for the year and highest ever in terms of organic new model sub-add in the same quarter that we hit the end of sale of perpetual licenses for individual products, a remarkable achievement. For the year, we added 50% more new model subs than maintenance subs, even as maintenance attached and renewal rates moved to record highs.
The growth in new model subscriptions fueled 74% constant currency growth in new model ARR. Total recurring revenue grew to 53% of revenue, compared to 47% in the fourth quarter last year. We're also pleased to see the unit volume through our eStore post strong growth, and we believe that our online business will continue to grow to increase meaningfully. When you add in the business we do directly with enterprise customers, our overall direct business grew to 21% of our revenue volume, compared to just 17% and 16% for the past two years. We set a record for the number of large deals in the quarter. Overall, a 19% increase in total deal value. The majority of these large deals were flexible enterprise licenses or EBAs, which are helping drive subscription growth.
Specifically, existing enterprise customers that moved from one of our older license agreements to the newer Token Flex EBA have resulted in subscriber growth three times higher than before. That's pretty meaningful. Just as a reminder, for those of you keeping score at home, the new Token Flex EBAs that we signed up this quarter, over 20,000, won't start to show up in the count until next quarter. Large enterprise deals signal that we are becoming a more strategic and essential partner to our largest customers. Our long-term vision and position as the thought leader in the industry is an essential part of their commitment to us. To summarize, Q4 provides a nice proof point that the business model transition is working for our customers, our partners, and us. I couldn't be happier with the results. Now that we covered the past quarter, let me turn to long-term value creation.
Given the short-term impact that the model transition is having on our traditional financial metrics, the recent turmoil in the stock market, growing fears about a global macroeconomic slowdown, and the noise created by investors focused on the short term, I think it's worth stepping back and taking a few minutes to outline how we are creating long-term value for our shareholders. First, we are dramatically increasing the lifetime value of our customers with our new business model. While this creates short-term downward pressure on our traditional financial metrics, we have repeatedly called out how it creates significant financial returns over the next three to five years. Second, we are simplifying our entire go-to-market strategy to align with the concept of being an all-subscription company. This involves a significant increase in our direct customer sales and marketing efforts, both at the enterprise level and through e-commerce.
Not only will this change the cost structure, but it will greatly increase how effectively we serve our customers. Third, and most importantly, we are exploiting the cloud to deliver more compelling products to our current customers and dramatically expand the size of our market opportunities. Our early investment in cloud-based design and engineering tools have given us a once-in-a-generation opportunity to lead in both the construction and manufacturing software market. Mainstream mobile and cloud technologies have opened up a market with more than 100 million potential subscribers. We have made substantial investments to achieve our leadership position, and let me be as clear as possible, the technology platform for the future for all design and engineering software. No, really all software is the cloud, and the company that wins will have substantial and sustainable long-term advantage.
To summarize, our framework for building long-term shareholder value is to increase the lifetime value of every customer, change our cost structure, the means by which we reach customers, and finally, build the best cloud-based products and services in the industry. While generalities are important, let me specifically highlight recent activity in terms of these long-term objectives. Most importantly, our business model transition is underway and exceeding expectations. By the second half of this year, we will fully be in a subscription-only mode. Customer and partner reaction to our new business model has been very positive, which gives me great confidence about our approach to the transition. Next, we continue to diligently control our costs. In Q4, we kept spend flat year-over-year.
We recently did a sizable restructuring that balances the need for financial discipline with an ability to invest in critical initiatives that drive the long-term health of the company. In September, we discussed annual operating expense growth of 5%-6%. The restructuring, combined with other cost-cutting measures, has allowed us to put a significant cap on spending for the next two years. We are committed to keeping spend growth roughly flat to slightly down through FY 2017 and FY 2018. Finally, our cloud-based products are the undisputed leaders in their respective categories. While many people are focused on the business model shift, winning a leadership position in the cloud leads to a long-term, sustainable competitive advantage. Building a great company over the long term requires balancing the interests of shareholders, employees, and customers. Done well, these are not mutually exclusive.
While it's easy to talk about, it's harder to do. I'm very proud of the many awards we received that recognize our success in doing just that. A few in particular worth mentioning are we were selected for Fortune's World's 25 Best Multinational Workplaces, Fortune's 100 Best Companies to Work For, and we're in the top five of Fortune's Most Admired Software Companies. We also were given countless awards for innovation and product leadership. Bringing it all together, we are updating our long-term cash flow model. As we look to FY 2020, assuming that the number of outstanding shares remains constant, we will be able to produce about $6 of free cash flow per share.
In the following years, free cash flow continues to dramatically increase. We see a path to free cash flow that exceeds $2.5 billion, or more than $11 per share of free cash flow by FY 2023. As you are aware, our cash flow generation this year and next will be impacted by the transition from perpetual to term-based licenses. The silver lining is that we now have an opportunity to switch to a new operating structure to support our business needs. This change will result in an increase in our U.S.-based cash while lowering our long-term tax rate. This will greatly help with the long-term structural imbalance of ongoing revenues and cash flows as we move through the business model transition. We will give you more details during this year as our plans materialize.
As we turn to fiscal 2017, we couldn't be more excited about entering this next phase of our business. This year will be the most unique in the company's history as we complete the transition from perpetual licenses to cloud-based subscriptions at the end of Q2. It'll be an interesting and unique quarterly pattern. None of the traditional seasonality patterns for sales metrics will be applicable, nor will year-over-year growth rates of traditional financial metrics be helpful in understanding how we are performing through the transition. Q1 will be our first quarter with only subscription offerings for individual products. We have a number of sales promotions in place to accelerate our customers' move to desktop subscription. As you think about Q1, the best result for us would be to be higher on subscriptions and lower on revenue.
Now, this may seem counterintuitive, but in fact would result from an accelerated transition to desktop subs. The end of Q2 will be the next major milestone of our business model transition as we stop selling perpetual licenses for Suites. We expect the dynamics will most closely resemble what we saw in Q4 with the end of sale for individual products. Q3 will be our first quarter of subscription-only sales across the board and will likely experience sequential slowdown. Q4 should begin to show more normal sales trends in our new subscription-only model. In our view, the better metrics for tracking our progress over the next few quarters will be subscription additions and ARR growth. The entire company has been realigned and is hyper-focused on subscription additions and growth in recurring revenue.
We have put in place both the processes and incentives, such as new sales comp and bonus plans, to align with our goal of growing our subscription base by a 20% CAGR over the next 4 years, which will drive a 24% CAGR in ARR. Many of you have asked about the impact of the macroeconomic environment on our business. We've said for the past several quarters that the global conditions have been uneven. Most of the emerging markets have been a mess, but most of the mature markets have been decent. As we evaluated our strong Q4 results, we didn't see any further degradation in the demand environment. However, there is no denying there is clear divergence in sentiment with regards to global macroeconomic conditions. Coupled with the uniqueness of this year, our forecasting process has never been more challenging.
As such, we believe it's prudent to take an appropriately conservative approach to our outlooks for the year while remaining confident in our ability to achieve our long-term targets. For the year, we're projecting to add approximately 500,000 subscriptions, with new model subscriptions contributing the vast majority of the growth. As expected, this will be the year where the transition causes the biggest revenue impact as we move more fully into the subscription model. On the earnings call last quarter, I talked about 1%-2% spend growth for FY 2017. Given that our typical annual spend increase starts at about 4% just from merit and healthcare costs, et cetera, some cost actions were already factored into that guidance. I've spoken repeatedly about reducing our spend growth while staying the course with the investments required to make both the business model and platform transitions.
Examples would be investments in building out a low-cost, high-touch inside sales team or in building out our cloud platform products. We are conscious of the impact that cutting spending too much or not reducing it enough could have on the rate and ultimate success of the transition. We believe our current course of managing spend growth flat to modestly down for FY 2017 strikes the right balance. As I previously mentioned, we will maintain flat growth through FY 2018 as well. We aren't updating the entire framework we laid out at the Investor Day last September, there are a couple of data points that should help you with longer-term modeling. As I said earlier, with the updates to the model overall, we see FCF per share of roughly $6 in FY 2020, given our current view of top-line growth and our spend trajectory.
Our revenue projection for FY 2017 is slightly lower than the framework we provided in September for a few reasons. Continued FX headwinds are pushing revenue down. We are assuming a higher mix of new model subs based on our recent performance. As I mentioned earlier, we're taking a more conservative stance given all the moving parts this year. Similar to our view on Q1, hitting the low end of our revenue range while exceeding our subscription guidance is a desirable outcome for the year. If we experience greater than expected demand for the last perpetual licenses for Suites, it will likely drive revenue to the high end of guidance or beyond. To wrap things up, we're really excited to be one step further along in the transition. Our long-term strategy will provide our customers with greater flexibility, more compelling products, and a better user experience.
In doing so, we're driving higher lifetime customer value, simplifying our go-to-market, and significantly increasing the size of the available market. Simultaneously, we are reducing our cost structure to accelerate our transition and increase our profitability, as well as evolving our operating and tax structure to better meet our future business needs. In short, we are creating a more predictable, recurring, and profitable business for Autodesk in the years to come. Finally, I'd like to thank all of our employees and our partners who worked so hard to make last year a success. Operator, we'd now like to open the call up for questions.
Thank you. Ladies and gentlemen, if you have a question at this time, please press star followed by the number one key on your touchtone telephone. If your question has been answered or if you'd like to remove your line from the queue, you may press the pound key. In the interest of time, we ask that you limit yourself to one question and a single follow-up. Our first question comes from the line of Philip Winslow from Credit Suisse.
Hi. Thanks, guys, and congrats on a great quarter. Carl, really appreciate that update on the outlook. It certainly was impressive on the cash flow metrics, clearly above your old guidance and above even where we were. When you look at those cash flow long-term targets you gave up there, I know you're not updating the whole framework, but it certainly implies it feels like margin's solidly ahead of what you talked about at the Analyst Day. Obviously, you talked about fiscal 2017 and fiscal 2018 on the cost side. Wondering if you could give us just how you're thinking about the framework of margins in that long-term guidance, and then from a top-line perspective, how you're building up there. Obviously not giving the full detail that you guys talked about. Then just one quick follow-up to that.
Let me just quickly touch on it, and Scott can jump in. Most of the uptake to the guidance was not a change in the overall demand. Let's just put that out there, that kind of assuming the same level of demand. We were pleased in Q4 to see greater demand, but no difference in demand from when we talked about the model. A lot of the upside is driven by greater cost control. A lot has to do with some of the stuff we outlined for FY 2017 and 2018. It's just as you roll that through the subsequent years, just starting from a lower cost base has a tremendous advantage. In a lot of ways, that's what we're responding to. Many of our investors had said, "I understand the transition.
I love to where you're getting on the other side, but can't you control costs a little bit better through this transition?" I think in response to that's where we got to, and that's where you see a lot of it. The other slight shift, I'd say, is we're more optimistic about the move to the new model. We've seen more positive signs of acceptance by both customers and partners. Remember, a little bit of our reluctance to move more quickly was making sure we had acceptance of both groups and the ability for them to successfully navigate the transition as well as us.
Phil, this is Scott. The only other thing I would add is the framework we laid out is not intended to be something we update every quarter. As you roll through the changes that we've talked about in spend growth out through time, then of course, it doesn't step function back up after fiscal 2018. It takes a while for the spend engine to accelerate again. We do see margins expanding. I think what we showed on the 29th was in that mid to high 20% range in fiscal 2020, and it's probably a good 4 to 5 points higher than that, given what we just talked about in spend.
Great. Well, I was at $8-$9 in free cash flow, so I'm a big fan of your 11. One last just follow-up on the quarter itself. You guys talked about the quarter's strength was not driven by the end of sale of licenses for some of the products. Carl and Scott, I wonder if you'd talk through just the verticals in geos, what areas were standing out as particular areas of strength, and how are you thinking about that in terms of your guidance?
Yeah. A couple things about geo. Let's start with the geos because it's really black and white. Every emerging economy, developing country, whatever you want to call it, practically, stinks. The exception for us was China. Despite the news that you may have read, China continues to perform well, but there are places where our business in emerging markets has just totally tanked. When you look at a number of these places, the Middle East, Russia, which are largely driven by oil. Brazil is bad. You look there. When you compare geographically, our EMEA results look not as strong as the others. Two factors there. One is EMEA had a fabulous fourth quarter last year, and the second part is really the MEA part.
The E part of EMEA did great, but the MEA part just struggled, and it really is due to oil and commodities and other stuff. The geographic part is pretty simple. Don't see a big change. I think relative to what we hear others talking about or read in the newspapers, if anyone still reads newspapers, China is better than you would expect from taking your readings. The only weak part of the developed world, but no delta there, is Japan. Japan has been weak for a while. Currency is an issue, but generally, we saw flat demand. Across the markets, both construction and manufacturing and our media and entertainment were all good, and there wasn't a lot of difference across it that I would point to or call out any trends that we're seeing that we're projecting into the future.
Yeah, I agree with that. By the way, the commentary that we just gave on emerging markets is really no different than what we've seen all year. We've seen China actually be a bit of a bright spot for us in the emerging markets and everything else, and in particular, Russia and Brazil have really struggled year-on-year. The only other thing that stands out when you look through, it was kind of strength across the board in Q4, and in particular, the number of large transactions that we did, the deals greater than $1 million, is heavily driven by our EBAs, our big Enterprise Business Agreements. We had a very strong quarter there, and in fact, did more than a handful of eight-figure deals there.
These are all ratable, you don't see them all dropping into the revenue stream in Q4, but a handful of eight-figure deals, and the largest manufacturing deal we've ever done, we just got done in Q4. I'd say strength on the high end as well.
Thank you. Our next question comes from the line of Saket Kalia from Barclays.
Hi, guys. Thanks for taking my questions here. Carl, you talked last quarter about selling anywhere from 500,000-800,000 units in a given year, I think you said that it was about 650,000 this year. Correct me if I'm wrong. You've given us the subscription component of that with the subs guidance for fiscal 2017. Just normalizing for all the noise in 2017, how do you think about unit volume for fiscal 2017?
I think unit volume will be up. Not a huge amount, but up. One of the tricks that we're kind of referencing in the forecasting is that, in particular, I think we have a couple of quarters where it's hard to know which way people will go, Q1 and Q2 in particular, even extending into Q3. We're doing our best job on triangulating from what we're hearing from our partners, what we're hearing from our salespeople, top-down models, extrapolation of trends. To be honest, it's a little bit hard, and that's why we have a slightly broader range in terms of guidance and why we gave the slightly paradoxical guidance that I wouldn't mind seeing revenue down.
I think I've now done something like 40 of these earnings calls at the end of the quarter. I don't think I've ever rooted for revenue to be down, but this is certainly a first. It's just different than it's ever been, and I can't highlight enough that the next four quarters are just going to be a little bit odd. That's why we pointed to, and I think you're on the right point, which is look at volume. We think volume will be up. Look at ARR. There will be a number of metrics that underlie, really the foundation for the business doing well and for underlying demand. Then a little bit, we're taking is the math will follow as it will.
Just as the math is ugly in the next year and a half, the math gets beautiful in the subsequent years. We'll take that as we can. One other addition, let me just add what our bias is, and I think strongly supported by many of our investors, is to do anything we can to transition our customers to the new model as fast as possible. There are a number of things like this that we're doing, which again we think could adversely affect some of our guidance on revenue, but we think will also help on exceeding on subscriptions. Many of you have noticed things like promotions in Q1. I think you'll see promotions in Q1 around our new model businesses. I think you'll see the same thing in Q3.
It would not be out of the question for us to do stuff like that. A number of things, but our bias is we had already moved up the end of sale date by six months, and now we're using all the knobs and levers we have to continue to put pressure on the system to move to the new model as quickly as possible.
That makes sense. Then for my follow-up, I'm not sure for Carl or Scott, but one of the difficult things that I find modeling externally is that ARPS number, particularly around new model ARR. I'm not sure if this is something that you've looked at, but on a steady state basis in Q3, Q4, and beyond, based on the historical mix that you've seen of historical volume and subscription pricing, what's a fair range to think about new model ARPS in sort of a steady state? If that makes sense.
Yeah. I understand that the model drives you to want to focus on that metric, Saket, I think the key is what Carl said earlier. We've geared the entire machine, from sales comp through the variable comp bonus plans that every employee has. You look at our promos, you look at the way we will point our channel partners next week at our sales kickoff event. We've geared everything toward driving subs, because the more we drive subs, and in particular, the more we drive the new model subs, ultimately, that's what builds the strength of the model and gets us to that vision of fiscal 2020 that we've laid out. ARPS, to us, is more of a derived number. We're focused on driving subs and driving ARR. When you divide those two, you get ARPS.
I think if you wanted to think about it a little bit more mechanically, I would give you two pointers. One is the fastest-growing piece of our desktop business, which is what's driving the new model subs as much as anything, is the low end. It's the AutoCAD LT desktop subs. That's naturally going to put downward pressure on ARPS. The second is the way we calculate ARR on a quarterly basis is we total up our recurring revenue for the quarter, multiply that by four to get it annualized, and that's our ARR. The linearity in a given quarter will drive the ARR calculation.
In other words, the more of those sales that come in late in the quarter, there's fewer revenue that actually accretes into the reported revenue for that quarter, you don't get to see the full effect of it until the following quarter. There's a little bit of seasonality and linearity effect on ARR at any given quarter that I think can also throw off your ARPS calculation.
I mean, just in general, the one thing I would say about ARPS is I could see it going slightly higher this year as more of the Suites and our full-fledged offerings hit. As Scott pointed out, our lower-cost products like LT will put pressure on it downward, as well as many of the cloud subscriptions. As we've talked about, many of the cloud subscriptions are for larger groups working together as opposed to individual design and engineering products. Things like BIM 360 and PLM 360 can have hundreds of thousands of users. The average cost per subscriber will be substantially lower than a design product. It really is all caught up in that first word of average.
Got it. Very helpful. Thanks very much, guys.
Just driven by mix. Yeah, thanks.
Thank you. Our next question comes from the line. It's Sterling Auty from J.P. Morgan.
Yeah, thanks. Scott, the comment that you made that in the desktop, it's the lower end that's really kind of driving the subscriptions at the moment, do you think that's just a matter of who is interested in adopting first and we'll see that mix change over time? Is there something about the subscription format that's going to drive users more towards a lower-end product versus some of the higher-end products?
It's just that we introduced it first.
Yeah. From a volume standpoint, Sterling, that's absolutely right. There's nothing inherently advantageous to buying a subscription on LT versus buying a subscription on AutoCAD or PRDS or any of our other products. It is our highest volume product, and given that it's our highest volume product in perpetual sales, it'll be our highest volume product on recurring sales as well.
We really focused on it first because one of our concerns going into this transition was that the customers who were more paying for the flagship products would be very comfortable with the new models, but that some of the LT users were more casual and might not want to enter into longer-term commitments. We focused very heavily, given that it was our highest volume product, on making it go there. I think what we're seeing a little bit is we're measuring the effect of our influence on this. A better time to take the measurement will be as we hit Q3, and when we get to the end of sale of the perpetual licenses for Suites, as well as introduce our new offerings that are really the desktop replacements for Suites.
Got it. As a follow-up, you mentioned a couple of times the good results in terms of the large deals. Was there anything in particular that you saw stood out to drive some of those deals in terms of the timing, whether it just be year-end budget flush or particular types of problems that those customers were trying to solve with doing these larger deals?
I think there are a couple of things that have happened. Some short-term, some long-term. One is there has always been a more of a seasonality to our large enterprise deals. Some of it is driven by their end of year. Remember, even though one of the quirks for this company is our calendar and fiscal numbering, it does give us a second close. We not only have a customer end of year, but our salespeople have an end of year too. There's always been a stronger influence on Q4 because of that. There's a second thing is, if you go back a handful of years, that was a non-existent part of our business, and I think we've done a number of things over the midterm to change that. One is we built out a sales force that did it.
We changed the nature of some of our products to better accommodate it. We have licensing models that more align with what they want to do. I think our drive really to take a leadership position in the industry, whether that's what we're doing in media and entertainment or with BIM in AEC or what we're doing on the cloud in manufacturing, has led companies to view Autodesk as a more strategic partner. I don't know how many years I'd have to go back before I would find that a $1 million deal in a quarter with a big deal, and now it's fairly routine. We've invested heavily. As we've detailed at Analyst Day this year and last year, we think there's a lot more there. We think the offerings coming enable this to grow much more quickly.
What we love about it is it does have some trickle-down effect in that all of these companies greatly affect their supply chain. We're really excited about the enterprise business, and we're going to continue to keep investing in it. As a matter of fact, the biggest investment came during the 2008, 2009 downturn, when we recognized our statistical run rate business was falling off. One of the places we could really invest was in the enterprise. It was really out of that soul-searching in the darkness that we really landed on this. It's just been a continual success.
Got it. Thank you.
Sure.
Thank you. Our next question comes from the line of Heather Bellini from Goldman Sachs.
Great. Thanks, Carl, for taking the question. I just wanted to ask you a little bit about your maintenance ARR, which declined in the quarter. I guess after you cancel suite licenses at the end of the second quarter, should we expect the declines in the maintenance ARR to accelerate? I'm just thinking about it. It's a high-margin maintenance stream. What's baked into your assumptions on how this should decline for fiscal 2017, given your earnings guidance? Then I just have a follow-up.
Yeah. Let me just say one more, then Scott. Yes, you absolutely should count on maintenance ARR going down after the end of Q2.
Yeah, Heather, I wouldn't say, I don't know that accelerate is the right word, but this is again, one of the trends we talked about back at Investor Day, that as we sell our last perpetual license, that will be the apex on maintenance subscriptions, which will come down at a rate consistent with our churn rate. Of course, maintenance ARR will trail down as well. There's a lot of maintenance sitting in deferred revenue, it's not going to come off quickly out of the reported revenues, but it will trickle down. It will become, in the second half of the year, both maintenance ARR and maintenance subscriptions growth will be a headwind to our total ARR and total subscriptions growth.
Okay. The follow-up question, I guess there's actually two quick ones. Your new model ARR increased by about $35 million in the quarter. How do we think about that ramping throughout fiscal 2017? Is there a target we should be thinking of, or do you guys have a goal in mind that you could share so we could benchmark how you do versus expectations? The other question we're getting asked a lot about, and you gave us some of the pieces, how are you thinking about your cash flow for fiscal 2017?
Yeah. On the first, Heather, we're not providing ARR guidance for the year, certainly in that level of granularity. I think, again, that we've set up the machine, and when I say that, I mean every part of the company to focus on driving subs. The easy place to track right now through this year will be to track how we grow both total subscriptions, more importantly, the new model subscriptions, because that's what's going to grow the company longer term. On cash flow, I'd say a couple of things. Cash flow for Q4 was artificially low. We had very interesting linearity in the quarter, where the third month of the quarter, so January for us, saw more than half of our total sales get done.
Since we bill net 30, in a normal linearity, a normal 30, 40, or the normal that we have experienced in the past few years, we would have seen about $95 million more cash flow from ops fall into Q4. Given the linearity of the quarter, given those sales got done late, that all is sitting in receivables. By the way, our receivables is 100% current. There's no risk on that. It's sitting in receivables and will now be collected in Q1. What it means is there's a headwind on the cash flow from ops we just reported for Q4 that'll just ripple over into Q1.
If you remember what we said for cash flow or free cash flow back at Investor Day, we said we thought it would be break even to slightly negative for fiscal 2017, that now goes up by about $100 million year-on-year. Think of it being certainly positive, somewhere around that $50 million-$100 million range.
Thank you. Our next question comes to the line of Jay Vleeschhouwer from Griffin Securities.
Thanks. Good evening. Carl, in the second half of this year, after you terminate the perpetuals for Suites, you'll be introducing your so-called Collections. You've stated before that this is one of the opportunities you have to reduce the complexity in your business. Could you talk about how you've anticipated perhaps the cost savings in the selling process from introducing these new Collections and improving or aiding the overall salability process of introducing that?
Let me just break this part into two halves, Jay. I don't think any of our financial guidance is just predicated on what we're going to do with Collections, but we do think about it a lot. We would really like to simplify the offering. For a whole number of operational reasons, it's incredibly important that we do that. The second, and probably more important thing that we're trying to accomplish with Collections is to give an incentive for the existing customers who are on maintenance for Suites a more attractive thing to move over to. As we said, we want to try to use the carrot as much as possible to convince our best customers who are on maintenance subscription to move over, and that's what we really are looking to do. At the same time, it is a great way to simplify our product portfolio.
We're going to use it for both those things. I think it'll be important, and we'll get a better read on it as we go through the year. A lot of financial modeling that we've been talking about so far this afternoon is really predicated on overall volume, and we're not assuming any kind of outsized performance from Collections.
At the analyst meeting, you mentioned that about a tenth of your named accounts had already adopted EBAs. I was wondering if you now have an updated adoption rate in light of the strength you highlighted for large deals in Q4. Looking out, how would you talk about the pipeline or assumptions of close in the EBA pipeline for fiscal 2017? Our understanding is that you've got a number of interesting names in terms of industrial and automotive accounts, for example, and then perhaps you could talk about that.
Yeah. We do. We have a lot of really large manufacturers and large construction companies. There are certainly other companies in there, a number of medium entertainment companies. It's there, but it is focused on large manufacturing and large construction companies. I don't think we've hit 20% yet. We're somewhere in that 10%-20%. I don't have a more precise number at this point, but there's still a lot of room to grow. What's particularly good about this business, the growth in the business is just not the number of accounts, it's the growth within the accounts that's been so substantial. It's not like we sign an EBA, and it's the end. Remember, there's a consumption-based aspect to this as well, and we detailed why we think it's a win-win.
The existing EBAs continue to grow, and we have more than 80% of our large customers are still not on these flexible licenses. The last part, just to tie both your questions together, Jay, is essentially Collections is a form of flexible licensing for smaller customers that can be sold through the channel. What we really want to do is use the pattern, just see the model that's worked so well for the enterprise and bring it to small and medium businesses and allow our partners to sell more flexible, consumption-based licenses as well.
Thank you. Our next question comes from the line of Keith Weiss from Morgan Stanley.
Hey, guys. Good afternoon. It's actually Stan Zlotsky sitting in for Keith. I wanted to come back to the restructuring that you guys announced, 10% headcount being taken out. Where are the costs that are being taken out? Where are they being reinvested as we head through fiscal 2017?
There's only a small amount of reinvestment. If you want to think about it, there are probably three areas where we're doing reinvestment. One area is moving to more inside sales, which I think is really important. The second is we continue to build out our back office. Remember, these flexible license offerings we talk about and are so proud about, we actually need to build the infrastructure to support that. Consumption-based model is needed. We're still investing in doing that. The third thing that we're doing is, I know it's gotten a lot less focus. I feel like I could stand on my head and talk about this, and it still doesn't matter.
The really big thing we're doing is investing in the cloud and mobile, and we're trying to change the dynamics of the market, and we are continuing to invest in that. Some of it is tied to cloud infrastructure and things like that. As you step down a level, you would see things like continued investment in our enterprise sales force, things that are working. Some of that stuff will fall in the category of the regular rejiggering you do within a business. Clearly, we have less people selling in Russia than we used to, and more people on major accounts. Hopefully, Stan, those gives you the areas.
Yes, that's very helpful. Maybe one more for Scott. What layers of conservatism, or maybe how did you approach your thinking around the midpoint of 500,000 guidance for subs for fiscal 2017? That's it for me. Thank you.
Stan, we really built this bottoms up. We look at a handful of things. One that we have already talked about, which is we will hit the end of sale of professional license for Suites mid-year. We do expect some buying behavior, much like we saw at the very end of Q4 when we hit the end of sale on individual products. I think that will put some upward pressure on maintenance subs in the first half of the year. In the second half of the year, there is no new maintenance subs. While we have a high renewal rate on maintenance, it is not 100%. Maintenance turns around and becomes for the full year, becomes a headwind for us on some subscriber adds.
The remainder is looking at what we have seen this year, what we think will happen next year in terms of unit volumes, knowing that for the second half of the year, those unit volumes will all be in some form or fashion a subscription. It is putting those two together and really building it bottoms up, with the assumption on renewal rates that leads us to that midpoint.
If you just look at the numbers, those are the ones that we can most have the greatest confidence in. Place where, the question about conservatism, we probably are conservative on the side of some of the cloud subscriptions. I think there is a lot more to be done there, I think this is going to look much more like the adoption of any new technology. There will be a tipping point. I cannot tell you exactly where it is. We are happy with the growth rates we see today, there will be a tipping point, just as we have seen with the adoption of almost every other technology we have ever introduced.
Thank you. Our next question comes from the line of Ken Wong from Citigroup.
Hey, guys. Maybe a question first for Scott. In terms of your spend target for next year being kind of flat to down 1%, any conservatism baked into that particular number? I get from the top line and from subscriber base perspective, it's particularly tough to gauge what might happen, but from spending, you guys generally control most of that.
Yeah, I wouldn't say there's aggressiveness or conservatism built into that number, Ken. That's one that's fully in our control. Really, the only variable there is if variable comp goes way up because we have a blowout year. Beyond that, and by the way, we've gauged that on subscriber growth. I think that would actually be a good thing if it went up. The spend number is one we've got a pretty firm handle on.
Gotcha. Then the follow-up, Carl, in terms of your kind of new subs, what are you guys seeing on renewal rates? Maybe not specific renewal rates, but relative to internal targets, is that coming in kind of better than expected? How should we think about that going forward?
Yeah. So far so good. All of that looks really good.
I'd say when you look at both the on the maintenance side, we look at attach rate and renewal.
Yeah.
Then, of course, from the new model subs, we look at renewal. We've seen upward pressure across the board on both attach rate and renewal rates.
Yeah.
I'd say that's one of the, as Carl talked about, we look at next year or fiscal 2017 with a lot of excitement, a lot of that's based on what we've seen now in terms of the acceptance of the new models and what we're also seeing on kind of the general bias upward.
I think it will continue. If you want to think about this, with the end of sale of the perpetual licenses for individual products last quarter and the ones for Suites next quarter, customers are more incented than ever to stay on the program that they're on. In the past, somebody could get off and then get back on relatively painlessly. That is no longer a possibility by the second half of this year. I think it will continue to lend pressure on an upward bias.
Thank you. Our next question comes the line of Steven Ashley from Robert W. Baird.
Thanks so much. I was wondering if you might be willing to disclose what kind of unit volume you had in the year just ended, FY 2016.
Yeah. Unit volume is about 650,000 licenses.
Perfect. Then, at the Analyst Day, you gave an outline chart, showing what you thought that revenue might be going out to FY 2022, what expenses might be, what margins might be. You've already commented that you're tweaking it a little bit with screwing down the expense growth a little more than that, and in the near term, maybe the revenue isn't what you thought. If we look in those out years of FY 2021 and 2022, is your revenue expectation for that still similar to what it was back at the Analyst Day?
Yes. Let me start at the end, and then I'll work backwards. The answer is yes. Our revenue outlook in the out years is still exactly what we outlined.
Okay.
To go back to the beginning of your question, I would say relative to September, our expenses are much lower, considerably lower.
Yeah.
From 5%-6% that we said was kind of the ceiling, now we're talking about flat for the next 2 years and then moderating from there. It's substantial. When you run the model, it has a dramatic effect, particularly having gone flat in the first 2 years of this transition.
Yeah.
The revenue, we're tweaking slightly, the revenue is not about unit demand, as we said. It's really whether people buy the new model or the old model.
Right.
All through this year, we will still have some variability and some inability to predict exactly what happens. The good news is we move to the end of the year. There's only one thing that people can buy new from us. While in many ways this is painful to go through the transition, we do hit a point probably in the middle of next year where 80% or 90% of the revenue in the quarter comes off the balance sheet. If I go back to when I first started, where you started every quarter in zero, you're now going to be at a place where you start every quarter with 80% or 90% of the tank full. We use the word business model transition, and maybe sometimes it gets lost, but that's really what it means.
Our ability to predict that gets very good at that point, and your ability to model it gets very good at that point, as you watch, say, the ARR.
Yeah, Steve, Carl touched on this earlier. It's a somewhat paradoxical year. To the extent that customers continue to move more quickly to the new models, that's going to put downward pressure on the current year's reported revenues. However, when you look out in time, especially when you get out into fiscal 2019 and fiscal 2020, we talked about this, the overwhelming majority of our revenues are going to be driven by new model subs. While a faster move to new model will result in lower end on reported revenues for this year, it actually is good news for building up that base and getting to the numbers we laid out in the out years.
Thank you. Our next question comes from the line of Gregg Moskowitz from Cowen and Company.
Okay. Thank you very much, and good afternoon, guys. Carl, was wondering if you could give us a bit more insight into recent customer behavior in conjunction with the termination of perpetual standalone. For customers that opted not to buy one last license prior to the end of January, did a lot of them buy a subscription instead before the end of the month? Did they wait one, two, three weeks after the quarter and then move forward? Have they yet to make a decision? Just wondering any more color you could provide on that typical customer, just from what you've seen and heard so far.
What it looks like to us is most of the customers are moving to the desktop subscription. That's the general one. There's still enough choices out there and still enough people that we can't totally account for, that we won't really know. We'll get a much better feel in Q3 as this becomes the only way to buy. From all the anecdotal info, talking to our partners, what people are tending to do is buy desktop subscription. Like we said, we thought about 10% of the volume was people who were taking advantage of the last moment opportunity to buy. I think we'll see that again in Q2, maybe slightly stronger with the Suites, I'm not convinced of that yet. Many people are finding desktop subscription a really attractive alternative.
Maybe when we started talking about this a couple of years ago, we were the tall pole in the tent or something about this. We are no longer the tall pole in the tent. The entire software industry is moving to these kinds of licensing models, and customers are just getting much more comfortable, and they understand what the benefit to them is of buying this way. I think there will be some noise for a while, by the second half of the year, we're well into the single model, by next year, it's kind of all done.
Gregg, I think one of the best indicators is if you look at Q4 results, not only do we see, as Carl said, a small buy ahead on the perpetual side, we had the biggest quarter we've ever had for organic new subscription adds. We actually saw both at the same time. It wasn't, I'll buy one or I'll buy the other. We had the biggest quarter ever for new model subscription adds, and we had the end of sale, which drove some buying activity on perpetual licenses. I think it's tempting to say, I'm going to do this, or I'm going to do that. What we're seeing is both is happening. They're two different customer sets.
Okay, great. Thank you.
Thank you. Our next question comes from the line of Brendan Barnicle from Pacific Crest Securities.
Thanks so much. Just as a follow-up to that question on new model subs. At the Analyst Day, you guys shared that I think 40% of the desktop subs were new to Autodesk. Do you have any updates on what those look like as you look back over the year in terms of net new customers to you on the subs?
No, we don't have anything that we can update you on yet.
No, I wouldn't suspect there's any change there, but bear in mind that there's a couple of statistics there. One is within desktop, how many of the desktop customers are net new to Autodesk? The other is, this is really where the future lies, you remember some of the statistics we have on the cloud products where they are overwhelmingly net new customers that are coming to Autodesk. It's still a small number today, but as you look further out in the future, we talk about TAM expansion. We talk about the upside of making not just the business model change, meaning the pricing change, but the platform shift to the cloud. That platform shift really drives. We see a lot of net new customers to Autodesk and expands our TAM.
Yeah. If you're looking at this carefully, just to put it in size and why we defer to the existing base, our cloud products that are doing well are in the tens of thousands, as opposed to our other things, which are in the millions. The growth rates can be high, the percentage of any particular characteristic can be high, but we're still comparing 10,000. On the other hand, 10,000 is growing at 50% a year, start turning into real numbers.
Then just following up, you mentioned you'll be meeting with the channel next week. You mentioned the Collections product as one sort of new offering for them. Any other changes that you expect to see in the channel this year, and anything related to the reorg and spending cuts that are going to change anything on that side?
No, I think what you'll see in the channel, yes, they will see Collections, and they will hear about that. They'll hear about new products. Some of the improvements in the back office, which should improve the efficiency of their businesses as well. There's always minor tweaks to the programs we have in place. I don't see anything hugely different coming out of it. Just one more year. Our partners follow us closely, as you know, and we've been working pretty closely. I would be disappointed if anything we said was a huge surprise to our partners.
Thank you. Our next question comes from the line of Matt Hedberg from RBC Capital Markets.
Hi, it's Dan Bergstrom for Matt Hedberg. Thanks for taking the question. To build on Heather's question and even Steve's, with fiscal year 2017 free cash flow moving higher with the updated targets, should we still think of free cash flow bottoming in fiscal year 2017? Then thinking back to the mentioned free cash flow glide path chart from Analyst Day, is there a quarter or year you'd point to when free cash flow would start to diverge from that chart and move more towards the updated targets?
Dan, I would say fiscal 2017 still looks like the bottom for free cash flow. Fiscal 2018 still looks like the bottom for net cash versus net debt. I don't think that's changed. I think the hard part about pointing to the quarter that is the nadir, that is the actual bottom, is linearity within a quarter has such a huge impact on free cash flow when you take these 90-day snapshots. We just talked about the linearity of Q4, which was a strong quarter from any dimension. Because of the linearity, we saw a lot of the sales that came in in Q4 come in in that third month, and depressed our reported free cash flows for Q4 by about $95 million, which now have shifted from Q4 and into Q1.
I think with linearity being such a large variable there, it's really hard to say this quarter definitively is going to be the bottom. Think of it as happening in fiscal 2017 from a free cash flow standpoint, and then from an accumulated cash standpoint, think of that as fiscal 2018.
Thanks.
Thank you. Our next question comes from the line of Brent Thill from UBS.
Carl, I think maybe your answer is going to be, you're going to have to wait to my question. When you think about the suite bundles that you're going to have available, you have a very complex lineup of products, and I think everyone's just trying to understand how you think about directionally, the number of bundles, what they look like. Is this just something we're going to have to wait, or is there any hints in terms of how we should think about that?
No, you don't have to wait that long. We're going to be rolling them out to our partners. Basically, it looks like think of it as a couple per industry. Kind of in rough terms. I could give you a lot more detail, but just think of it as a couple per industry. Then there'll be a handful of standalone products that generally are in different price points and different levels of sophistication and complexity. For example, some of our complex simulation tools like Moldflow, we're not going to wrap into a bundle because it just doesn't hit a broad enough section of the base. The things that we're putting into the Collections are the things that we think will be broadly used, and we're going to put together a couple offerings in each of our three large industries that we serve.
Brent, the way to think about it is the goal is, first of all, to simplify. You'll see us bring down what is a fairly large menu of offerings into a much simpler set of offerings, point one. The second is to drive consumption, to build those bundles in a way that drives broader consumption, much like we see with the Token Flex on the very high end.
Yeah. Okay. Just as a follow-up. Carl, many of the analysts that are on the call have covered software for close to two decades. It's been rare you see a seven-year-out goal, on the free cash flow number, just going looking at history, you've never achieved three-quarters of a billion dollars of cash flow, yet you're given a $2.5 billion goal. I know this is an aspirational target, there's a lot of things that could change, but what gives you the confidence at this point to do that when historically, these long-term targets have been tough to hit?
Yeah. Good question, Brent. I think it's partially what I just answered, is that for example, we start building a model in which almost all of the revenue comes off the balance sheet. Just starting with that idea that quarters used to start at zero and will now start at 80 or 90 or 95 is different. You will also be able to look at things like the churn and see the renewal rates and extrapolate from there. We generally haven't gone out that far, but people really didn't understand the dynamic of this, we were just trying to put it in perspective. To the extent that you want to discount FY 2023, feel free to. It's way in the future. I think the FY 2020 is there.
The problem we had, really the reason why we gave details around FY 2023 is because FY 2020 or FY 2021 is still so much in the steep part of the curve. When we were doing that and only talking about FY 2020, people were saying, "Is that the terminal? Is it the terminal cash flow? Is it the terminal EPS? Is it the terminal operating margin?" It wasn't. Just to the extent now that traditional financial metrics are going to go down predictably as a result of how we account for this, it's just as certain that they come up on the other side. I'm totally fine with people wanting to discount FY 2023. We just wanted to make sure that people understood that that's what our model looks like. If you back off it. Look, we've been pressed.
There are some people who think we should be able to get to a lot more. There's some people who get to less. You can then look at it in a pretty reasonable way about thinking about the number of subscribers. Remember, if anything, we have been attacked for our ARPS. There's not a lot of assumption about the ARPS going up. It really is all about subscribers, ARR really translates directly to revenue. Put whatever discounts you want on it, Brent.
Thank you. Our next question comes from the line of Steve Koenig from Wedbush Securities.
Hi, gentlemen. Thanks a lot for squeezing me in. I wanted to ask you a little bit about the long-term trajectory when it comes to thinking about new users. I know that there's definitional issues there. If we suppose that out of your 650,000 seat sales run rate, let's just say for the sake of argument, half of those are kind of new users versus half are people that are active but weren't regularly paying and maybe buying a new license or something. Unless I mangled that, if I'm thinking about that the right way, if we can assume that, what happens to that new user addition number as you move through the transition? Can that 300,000 go up due to adding pirated users? It probably should go up. You should be adding those old licensed users as subscriptions, those become renewals later in time.
Yep.
How can we think about the unit sales and the seat sales as we move along the transition?
Yeah. We would hope to see those continually going up. We've been pretty modest in our assumptions. You're right, you have the old subscribers. Remember, the definitional issue just really gets to this thing of a user could be at a different company, but they have been a user before. Let's just ignore that for a second, but the old subscribers would be there. Piracy, we think this will put a significant dent in casual piracy. As we said, the new offerings are the places where we're getting new customers. Coming at this a different way. We talk about it sometimes as TAM expansion. We talk about it as new users. I referenced the 100 million people in manufacturing construction who did have a need for many of the tools we previously sold, who do now, whether that's BIM 360 or PLM 360.
Those are new users. Those new users may be inside of existing customers, or they may be in new accounts. We've even given you an indication of in many of these products are relatively high. In almost all cases, the majority of the customers that are doing it are not traditional customers. The other thing that I think is really important, and I know you guys do not want to bake too much into this, into your model, but we're running a business, not a spreadsheet. In our world, the investments we've made in going to cloud-based engineering tools has led us to have a leadership position in that. We deeply believe that the future of all software is in the cloud and in mobile, and it opens up big markets.
I've always believed, for people who invest in Autodesk, this is one of the fundamental tenets you should believe in. There are less difficult ways to get there, we think the business model transition starts the curve we just talked about with Brent. If that is going to continue into the out years, that's based on the fact that we expand into new markets, get new users, and win the platform of the future.
Great. I'll leave it there. Thanks a lot, Carl.
Okay, good.
Thank you. That concludes our question and answer session. I would like to turn the conference back over to Carl Bass, CEO, for any additional comments.
Yeah. I just wanted to say one thing. I kind of hesitate to do it because it's all been going so swimmingly. I'm really just trying to avoid work for Dave and Scott and I over the next couple of days. In having the same conversation several dozen different times, because I've had it in private, even as I go down to the coffee store, I get asked by all the investors in our building. That's just the question of what's going on with our activist investors. I just want to take that head on and happy to talk about it privately with others, but seems like a mistake not to address it to some degree. Let me just give you my viewpoint on this.
The first thing is that I think a number of the 13D filers, as well as some of the activist investors, have raised legitimate concerns. The one part that kind of troubles me about it is that there are other long-term investors, really constructive investors, who have raised the same issues in a much better way. While we are absolutely talking to the activist investors, I think too much credit is being given to them and not enough credit is being given to our long-term investors who have asked for change in a constructive way. They've understood kind of the subtleties and complexities of running a business. They truly have taken the time to understand what the challenges and the opportunities we face.
I feel like they're getting the short end of the stick when they've really pulled the laboring oar in supporting the company over the long period. Many of my problems with sort of some of the more activist investors is that their focus is extremely short-term and somewhat simplistic. It just ignores some of the complexities of running a business. Let's just get on a lighter note for a second, and then I'll return to being more serious. Let me just tell you what my concern is. It's that it's easy to sit on the outside and have lots of good ideas how to do this or that thing better, or why things should move more quickly.
As a matter of fact, if you turn on sports radio or political talk radio, on sports radio, you can hear thousands of people who know what the coach or general manager or the owner should have done differently. I'm worried about that thing. I watched the game last night. I have a million ideas of what the coach should do, but I didn't actually have to do the job. In a nutshell, my concern is people who essentially want to turn Autodesk into a Men's Wearhouse. If you look, I don't think any of the people who got involved in Men's Wearhouse intentionally meant to do damage to the company. I don't think they meant to screw it up, but I think they had some simplistic views. What troubles me is seeing those same views expressed about our business.
Actually, I got one little funny story here. I looked at Men's Wearhouse the other day. I was trying. They changed the name of the company and the symbol. What I thought was interesting is changing the name makes it actually hard to track a price and how poorly it's performed. The symbol is a little bit of a telltale sign, because if you just look at it's squint, it's spelled T-U-R-D. It's actually TLRD, but it's what it looks like, and it's probably more like it. Let me just try to end this on a more serious note. We are talking to the activists. We would like to shift the conversation to the active investors to a more constructive dialogue. There is certainly a possibility that together we can do that.
Most importantly, I just want to say thank you to our long-term shareholders who have understood what the challenges are, but really the fantastic opportunities the company has. Thanks.
All right. That concludes this afternoon's call. As many of you know, we'll be at the Morgan Stanley conference on March 3rd. You can just contact me directly. You can reach me at 415-507-6033. Thanks.
Thank you. Ladies and gentlemen, thank you for your participation in today's conference. This does conclude the program, and you may now disconnect. Everyone, have a good day.