Good day, ladies and gentlemen, and welcome to the Autodesk Second Quarter Fiscal 2017 Earnings Conference Call. [Operator Instructions] As a reminder, this conference is being recorded. I would now like to hand the conference over to David Gennarelli, Senior Director of Investor Relations. Please go ahead.

Thanks, operator. Good afternoon. Thank you for joining our conference call to discuss the results of our second quarter of fiscal 2017. Also on the line is Carl Bass, our CEO; and Scott Herren, our CFO. Today's call is being broadcast live via webcast. In addition, a replay of the 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 third 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 our 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 2016, our Form 10-Q for the period ending April 30, 2016, 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. Those non-GAAP financial 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. And now, I'd like to turn the call over to Carl.

Thanks, Dave. We had a terrific quarter, and we reached a very important milestone with the end of selling perpetual licenses. Our Q2 results mark another solid quarter of execution and progress on our two big initiatives; increasing lifetime customer value; and secondly, driving increased adoption of our cloud-based solutions. I'll share more details on our Q2 results and then update you on what we've been doing around creating long-term shareholder value before getting into the current quarter and our outlook for the rest of the year. Starting with subscription numbers, we added 109,000 net new subscriptions in the quarter. We're really happy with the growth of new model subscription additions, which more than doubled year-over-year to 125,000. It was a strong quarter for product subscriptions, which drove the vast majority of the new model sub additions and it’s a clear sign of the progress on the transition with our customers and partners. Keep in mind that we didn't have any big promotions going on like we did in Q1. Product subscriptions had strong sequential growth and dramatic year-over-year growth of over three times. Product subscriptions come to us from a number of sources. We're really pleased to see that one of the primary sources continues to be net new customers, which accounted for approximately one-third of product subscription additions in Q2. It's impossible to get perfect clarity on these new customers, but it's likely that a portion of these people were previously pirating software and now have a much more affordable option with our subscription. Still others have been using alternative design tool and can now get access to better software. Looking at the other new model subscription types, our token-based EBAs experienced good year-over-year growth, but added fewer than we did in our seasonally high Q1. And once…

[Operator Instructions] Our first question comes from the line of Jay Vleeschhouwer from Griffin Securities.

Carl, let me ask you first about the pricing that you plan to take next month. Could you give some specifics as to how you're thinking about what the changes might be? And is this part of what you referred to in the past as SRP realization as a concept for you to help improve pricing and margin? Secondly, with respect to the decay of classic maintenance over time, how are you thinking about the rate of that decline? When we look at one of your competitors BBC, which is moving mostly but not entirely to subscription, the state of rate of decline for their maintenance is low to mid-single digit. So would you anticipate going fully to subscription that your decay of maintenance might be so much faster than that?

Yes. So let me start with the pricing and then Scott and I can kind of tag team the one about the maintenance decline. So starting with the pricing, what we wanted to do was, as I said, kind of rationalize the 3 different price points of AutoCAD LT, AutoCAD and the AutoCAD verticals. Whenever possible, we would like customers to get the right product. More often than not, that is on AutoCAD vertical, particularly tailored to what they -- the discipline they're in. And on the other side, for the more casual users, LT is clearly the more appropriate product. And what we were trying to do is move the price points apart, get the appropriate separation. And what we saw, I kind of referred to in the remarks, in our testing is that we can, with the right pricing, move more LT customers to AutoCAD and AutoCAD customers to the vertical. And so this is a part of the overall pricing. I wouldn't say it's particularly what I would refer to the SRP realization. It's just increasing total pricing, but it's part of a pricing optimization strategy. And let me take the first pass. Couple of things will contribute to the maintenance decline, as you said, classic maintenance. One is going to be the fall off from people who just choose not to renew for all of the usual reasons. The second one will be the conversion of the maintenance customers on to the subscription side. And on the first one, we talked a little bit about it and that we said that we expected the subscription base to be larger by the end of next year, and we gave you some hints or at least we gave you an algebra problem you can work on tonight, Jay. And then on the second one, we also talked a little bit about programmatic ways of doing that and over time how we will continue to encourage customers to move from classic maintenance to the new subscriptions. Scott, you want to add something?

No, I think you said it right. It's the decay rate. We [indiscernible] vary within that pool of maintenance subscribers by the way between low-end LT customers that have a slightly lower renewal rate versus suites maintenance customers, which have a much higher renewal rate. So as that pool moves over, the maintenance pool will go continually down from this point, basically 1 minus our turn rate. And at that point, what you'll see is the mix will shift inside that pool. So if you think of maintenance revenue as a P-time SKU, think of the price, the average price inside there beginning to shift more toward suites customers out through time, but the overall decline being driven by the low-end LT customers. I think the other way to peg this, Jay, as you sit down and do your modeling, is what we've said -- and I think we showed you these curves back at Investor Day, that our expectation is by the end of next year, the number of maintenance subscribers and the maintenance ARR will be below the new model subscribers and new model ARR by the end of next year.

[Operator Instructions] Our next question comes from the line of Saket Kalia from Barclays.

Maybe first to start off, I know ARR, maybe this is for you, Scott, but I know ARR is the better metric to look at. But for those of us that still look at billings, can you just give us a sense for how billings or maybe trending for those new model subscribers or new model billings versus maintenance billings because you obviously have a headwind from maintenance in that deferred revenue accounts? So any color on how new model billings are doing would be helpful?

Yes, Saket. One of the reasons that I don't think billings are the best indicator of our progress of course is they are term specific. And if I sign a new customer, but they only sign up for 1 year, that's 1/3 of the billings if I sign that same customer for 3 years. So I think it's a little bit less of an indicator than things like ARR, which are annualized than, of course, the overall subscription growth. But if you do quick math, deferred revenue changed by $3 million during the quarter, so our billings tracked pretty much to our reported revenues. And when you peel back billings at a high level, they track very much in line with what you see on subscription versus license.

Got it. That's helpful. And then for my follow-up, Carl, I think you mentioned that you saw good crossgrade activity for maintenance to product subscriptions, which of course is a net neutral to the subs count. Can you just give us a sense for how many of those crossgrade you saw this quarter? And then more broadly, are you still seeing customers from product versions older than five years make that crossgrade?

Yes, so let me go backwards. Well, so we didn't see customers from very far back because we didn't run any of the promotions. So the typical kind of legacy promotion where people had to be on subscription already. What I did mention is that in Q3, we were running a similar promotion to the one that we ran in Q1. So again, we'll talk about that 90 days from now and see how many of the legacy customers we can attract. We didn't break out the number on your other question of people who converted, but it was a substantial number. I'll leave it to Scott if he wants to give a number.

Yes, the color I'd add to that is, if you think of the quarter that we just closed, we actually saw 2 end of sale surges kind of at the same time. There was -- the normal end of sale surge that -- like we saw in Q4, where it was the last chance for our customer who wanted to buy a Suite perpetual license and there are some buy ahead activity, some pull ahead of future demand for that set of customers. And by the way, we predicted that, I think, pretty accurately versus what the results were. The piece that also happened was, of course, when we saw your last perpetual license, you also saw your last crossgrade. And we saw a lot of the same behavior that we saw at the end of -- if you remember when we stop selling upgrades in our fiscal '15, so more than a year ago, we saw a big spike in demand for people who were already using our product, but decided they wanted to upgrade to a more full functioned product. They're willing to pay more for it. We saw some of that same upgrade-like behavior happen within our crossgrades. And I think that's the piece that surprised us a bit and it's a positive for ARR. It's a neutral for subscribers because you had to be on maintenance to be eligible for the crossgrade, but it's a positive for ARR and it's a positive for total revenue going forward. So that's the way I would think about the surge we saw at the end of this quarter.

And our next question comes from the line of Richard Davis from Canaccord.

So it sounds like -- I mean, you're getting good growth as you kind of pointing out, reduced piracy and also kind of changed vendors. The question that I hear, when you hear from people thinking about changing vendors is a big hurdle is just kind of the data migration and things like that. Has there been any -- have you guys done anything technologically to make that less painful? Because if I have a model set and a different vendor, I'm terrified that if I move it over to you guys, it will incinerate me or something like that. So that's really kind of -- help me understand how you can make that less fearful for these folks trying to...

Yes. Despite the advice of our legal counsel, I can guarantee you it will incinerate you, Richard. Let me just divide the markets and talk about the dynamics slightly differently. In the AEC market, which is more project based, it comes up for a building or a roadway or a bridge. There is more willingness -- there is slightly more willingness to shift with new projects as new teams come together. I think the place where you hear even greater reluctance is in manufacturing. And there we did -- we've done a number of things that we've rolled out across our product portfolio that allow customers to work in new software and access the data that they created over the last couple of decades. So they can work with the native data. They don't -- obviously, as always, we can translate it and convert it. But even more importantly, we now let customers leave their data in the native data format and work on it whether they're doing modeling or visualization or simulation. So that's been, over the last two years, a pretty big breakthrough in giving people a much smoother ramp to go from one of the legacy products to a new one. And at a certain point, one of the other things that has to happen in motivating people to move to something is the new stuff has to be that much better. If you think about it, particularly manufacturing, many of the products that people are using are more than 20 years old. And we would like to think in the industry that we can do better than having 20-year-old products be state of the art. So part of it is making the new products and the new product experiences much more compelling. The other is to reassure people about being able to migrate their data or to use all their data and access it for as long as they need to.

And our next question comes from the line of Heather Bellini from Goldman Sachs.

This is Shateel Alam in for Heather. First one is on expenses. So your expenses were down 4% this quarter and you lowered guidance for the year. Can you talk about some of the actions you've taken that are driving this spend to be lower than you're expecting? In the past, you talked about discontinuing some noncore products in trade of hardware. Just wondering if we could see more of that. Are you done with that type of product streamlining?

The two biggest places have clearly been on consolidation of the product portfolio, and the second place is in, I just say, overall in our go-to-market activities. As it refers to the product portfolio, I mean, we're on a multi-year transition to a much more streamlined portfolio, including making the end of life of certain products. But also, as you look at collections as a big simplification from suites, all across the board, we're trying to simplify the portfolio, make it easy on customers, make it easy on our partners. And so you'll continue to see the folding in of certain functionality into the products that we keep going forward, the elimination of some of the older and less important ones. And we're constantly pruning for the things that work and don't work. So I would not, at all, expect to see -- or said more positively, I would expect to see that same kind of proactive management of the portfolio for the next couple of years.

Got it, that's helpful. And then for my follow-up, new model subs, the additions of 125 was higher than we're expecting. Can you just give some color on the mix that was traditional, desktop and EBA versus new cloud offerings? And are we starting to see new cloud subs become a higher portion of the adds now?

Yes, Shateel. This is Scott. We had a record, I think, Carl mentioned this actually in the opening commentary. We had a record quarter for cloud sub adds, so it continues to move along nicely. But that new model space will continue -- did this quarter and will continue into the future to be dominated by product subscriptions, so term licenses. In fact, for the first time, we had more inventory of product subscriptions within the new model type than of enterprise coming out of the EBAs. So think of it as being dominated by product subscriptions, EBAs would be next, cloud would be next.

Great. Thank you. That’s helpful.

Thank you. Our next question comes from the line of Gregg Moskowitz from Cowen and Company.

Okay. Thank you very much and good afternoon. Carl, if you look at Asia Pac, customers there have historically moved more slowly to adapt to our subscriptions. In the Q1, however, you did reference, I think, an uptick in new model subscriptions towards the end of that period. Can you give us an update on what you saw out of Asia in Q2?

Yes. I mean, I will continue to say that parts of Asia are slower to adopt the new model. We've had some pockets of success with some of our traditional partners in some of the electronic channels that have been more positive, but overall, slightly slower adoption with a handful of countries. And as you do the analysis, it's a little bit hard to dissect exactly what's going on since up until now, the preference of the channel partners kind of shown through and that they were able to exert certain forces on the customers. Moving forward, that goes away so, hopefully, everyone will be aligned. It's a great relief to make it to August 1 and get into just the new model. And I suspect going forward, we'll see more normalization. But given that there were two choices, Asia Pac certainly lagged a little bit.

Okay, that's helpful. And then just for Scott, based on where you are at this point in the year, you're calling for about 250,000 or so net subscribers in the second half, at least, at the midpoint of the range. And I know you don't guide quarterly, of course, but maybe you can just share with the group your high-level thoughts around net adds for Q3 as compared with Q4 as you see it today. Thanks.

Yes, sure, Gregg. We always have a seasonal pattern of demand. If you think of where the subs will come into the second half of the year, there will be more than 100% of what we add will be new model, right. Maintenance will be on a continual slow, but on a steady decline. So all of the new adds will come from new model types. New model types tend to have the same seasonality as a lot of our, previously, our perpetual license did. So heavier in Q4 than in Q3 would be normal demand pattern. We also see Q4 typically heavy on EBA sales, although as you see in the last couple of years, we sell a lot of them in Q4, the EBAs will pick up those subs in Q1. So think of this as being driven -- the Q3 versus Q4 driven by our normal seasonality pattern first. And then secondarily, we do think was were some demand pulled forward from some of the end of sale of suites at the very end of Q2 that probably makes Q3 from a sub adds standpoint a little bit more challenging than what we'd normally expect with normal seasonality.

Thank you. And our next question comes from the line of Ken Wong from Citigroup.

Great. Thanks for taking my question. My first question for you, Carl, with collections now, it's been out in the market for about a month. How should we think about kind of the customers you're attracting here? Is it customers that are using a few products moving up to collections? Or are you seeing your traditional suite customers kind of recognize the incremental value with this comprehensive bundle, and those are ones that are primarily moving towards collections?

Yes, Ken, it's probably just a little too early to know for sure. And it's almost like give us the two more months. In this first month, we have certainly seen some of the suites customers move over. Some of the buying at the end of the quarter was anticipation of being able to move from suites to collections. So there's certainly some of that in there, but we'll be able to do a little bit better breakdown of the customers when we get a little bit more traction there. It's just a little bit premature.

Got it. And then earlier in your prepared remarks, you mentioned at some point convert to a single subscription offering. Is this just consistent with what you guys have said in the past where just kind of a natural bleed off of maintenance and then you eventually just be all kind of new model subs? Or is there something more specific that you guys are looking to do to try to converge that process sooner?

Two things. I think we're being more direct in a general sense, but it's probably, at this point, we don't want to give more details. But what we do know and we have talked about the complexity and the expense that comes with multiple models. And so we'll continue to look at ways to just simplify all of the offerings and merge the programs into one as it makes sense. But I think we were, at least, hinting at more proactive behavior on our part.

Yes, Ken, the only color I would add to that is you've seen us do some of this already, first, with suites and now with collections. You see us bundling together to kind of ease the ramp from -- you see us bundling the Windows-based, whether it's perpetual license or product subscriptions, with cloud to kind of ease that on-ramp. And I think you'll see us continue to do more of that to make it easier to consume, to give customers a choice within what they ought to execute, whichever product type that they like. That was one of the guiding tenants with the way we design collections, and I think you'll just see that continue.

Thank you. And our next question comes from the line of Steve Ashley from Robert W. Baird.

Hi. Most of my questions have been asked, but let me ask about Japan. Just maybe color on what is going on there. And is there, trend-wise, any hope or any color on that would be great.

I'd say flat line. It is what it is. It hasn't gotten better and it hasn't gotten worse. Scott, would you want to...

I think, that's right. I think, the year-on-year impacts in Japan are probably more pronounced than elsewhere because they were slower last year to adopt a new model type, so the falling off a cliff of going from perpetual license to new model types is more pronounced in Japan than it is elsewhere. I think that's part of what we're seeing. Part of it is just the continuation of the trend that we've seen where it's been a tough economy in Japan, and we feel some of that. And I think it's -- so no change is what I'd say, Steve. But I think part of what we're feeling from a year-on-year standpoint is just the difference in the adoption model.

And then in terms of the promotions you're running, you've been great about calling out the fact that you are running a fairly aggressive promotion less discounting than we saw in the first quarter. But in terms of digital marketing and how much wood you're putting behind the arrow, is there any difference in how visible you will be with this promotion versus the one you ran in the first quarter?

Yes, Steve. What I said, last time, we thought the promotion was generous, but not well known by all of our customers. And after having run it in that one quarter, we thought we could turn the two knobs and turn down the discounting and turn up the awareness. And as we said a quarter ago, we were extremely pleased and surprised by the results at how far back into a legacy user base we could reach. And that was with, I'd say, minimal awareness. And so we would like to promote that more heavily going forward, and we thought we could accomplish both goals of greater SRP realization and a good response by just turning up the awareness. So you should see a little bit more of it. And as we indicated in the opening commentary, you can see more of it going forward.

Thank you. And our next question comes from the line of Keith Weiss from Morgan Stanley.

Excellent. Thank you guys for taking the question. Given the fact that suites were a bigger part of sort of what was going on in this quarter, you talked a lot about a lot more new model subscription types coming from suites. I would have expected the ARPS number, the pricing side of the equation start to improve. But by my math, it's still down sequentially a little bit both on maintenance and new model subs. One -- as a two-part question, one, why has that continued to decline? And two, when should we expect that to turn back up and start being one of the drivers of ARR growth?

Yes, Keith. We talked about this I think last quarter as well. Part of what's going on there is just the math. When you do the ARPS calculation, you're taking a quarter average ARR and dividing it by a quarter end sub count, right? So if you go back and say quarter average ARR divided by quarter average sub count, you'll get a different answer on that. But I think overriding longer term, you do need to expect for some period of time that to continue to come down. And it will come down simply driven by mix. We've talked about ARPS being very mix sensitive. It will continue to be mix sensitive. And as we get further out and layer on more quarters of selling suites and collections in our higher-priced products only in the new model way, you'll begin to see that ARPS rebound. But short term, we'll continue -- that pool will be more heavily dominated by the lower-priced LT and AutoCAD, things that have been out there only as product subs for the last six months. That's the way you ought to think about it. Very mix sensitive.

And to be a little bit more quantitative about it, I would look out about three quarters before we probably saw the inflection.

Okay. And that's just because -- that's what I was going to how long it will take for the demand on the suite side of the equation or now the collection side of the equation to really ramp up and sort of normalize that mix? It felt like once you end perpetual for suites and all the new buying for those high-end products becomes new model, that should start sort of inflecting that number upwards?

Yes, you've got to layer in multiple quarters of that ARR, Keith, before it gets -- it becomes as representative of the total new model ARR as the things that have already been out there for two or three quarters selling it only in a new model way. So that's the reason you see the lag. It starts now, but there'll be a lag effect before you get that entire pool properly representing what the new sales look like.

Thank you. And our next question comes from the line of Brent Thill from UBS.

Thanks. Scott, the midpoint of the guide looks like there's a little sequential growth from Q3 to Q4. Can you just explain why that is?

One of the benefits of the shift that we're taking is we have a lot more clarity on day one of a quarter of what the reported revenues are going to look like. So as we look at both what we already have in deferred revenue for Q3 and what's sitting in backlog, we've got very high coverage of the number that we forecast for Q3 and obviously, a higher degree of certainty on that. As we get closer to Q4, we'll be in that same position on Q4, and we can evaluate if it makes sense to go up at that point.

Okay. And just a follow-up on the ARPS question. It was down 7%. I mean, should we continue to expect over the next couple of quarters that to be down mid-single digit? Is that kind of a good way to think of it? Or should we think on the higher end of this?

Yes. Let me start, Carl, and you can add color. So think of ARPS at a minimum in the two line items, maintenance separate from new model. New model will have the trend that I just talked about, to Keith on the prior question, right? It will take a couple of quarters for that entire pool of new model ARR to be representative of the new sales that are going on. On the maintenance side, I think, you'll see the ARPS there rebound a little bit quicker. We will see -- we'll continue to see higher renewal rates on suites, and the suites will become a bigger part of the pool faster. So it may be a quarter out, it may be two quarters out, but I think you can expect to see ARPS rebound a little bit more quickly on the maintenance side than on the new model side.

Yes, I would agree directionally with Scott, and I would say back to the mix, I think, as we gauge the success of legacy promotions and just the cloud services that will affect the mix as well. So we'll be able to give a little bit more clarity as we go there.

Thank you. And our next question comes from the line of Sterling Auty from JPMorgan.

Thanks. Hi, guys. Given the perpetual sales of suites were higher than you expected in the quarter, is there a way to quantify the unit volume beyond the items you gave earlier? And specifically, what I'm looking for is, is there any chance that, that has an impact on your full year 475,000 to 525,000 subscription addition number because of that pull forward?

Yes. So I mean, one is we raised the bottom of the range, which is our indication that we don't think it has any effect. We remain confident in that and I would say, increasingly confident about it. It was interesting to see, and as I tried to highlight in the remarks, as we saw the surge in the perpetual suites, what we also saw was the new model stuff also take off in the last couple of weeks. So we saw higher -- obviously, higher than anticipated volume. And I said when we give back out the extra buying of perpetual suites, it was the volume of activity that we would have expected for this quarter. It was -- it slightly exceeded our expectations. And then on top of that, we also had whatever pull forward. So I think you may see some shifting between quarters on the margin, but we're comfortable with the guidance for the rest of the year. And that's why we're willing to bring up the numbers.

Yes, Sterling, I think that's kind of a little -- sorry, just on that last one, a little more context for you around the new model sub adds of 125,000 in the quarter. If you remember back in Q1, new model sub adds were 140,000, but buried in there were two things that didn't recur in Q2. One was the promo. We didn't run that legacy-focused promo in Q2. We will do that again, as Carl said, in Q3. The second is we had a big catchup of the EBAs that rolled in, in Q1 and were a little bit more than 25,000 of that 140,000 we had in Q1. If you net those two out, Q1, without those two, is in the, call it, 90,000 range. That grew to 125,000 new model sub adds in Q2. So I don't -- there's certainly the pull ahead, I don't think, signals any weakness in where we're headed of the new model side. The only thing to bear in mind is a part of the pull ahead that we saw, a significant part, was driven by crossgrades. And to be eligible for a crossgrade, you had to already be a customer with a maintenance agreement attached. So all of those -- all of that pull ahead business was really neutral to subs. They already have a sub and they have a sub now. So it's not a big piece of that pull ahead. Really doesn't affect the subs growth.

That makes sense. And then the follow-up. Carl, the comment you said that the subscription has accelerated the same time as the perpetual. Were you suggesting that you had customers that were buying both perpetual and subscriptions? Or do you think it was more coincidental to see both accelerating at the same time?

I don't – now, I'm speaking a little out of school. We got to do a little bit more work on this. But I don't think it was customers buying both. That's just a little bit of an awkward purchase. I think it was the outreach from our channel partners’ created demand. And when people decided to buy, then they made the choice. But more likely is they chose one or the other, not both at this point.

Thank you. And our next question comes from the line of Matt Hedberg from RBC Capital Markets.

Sure. Thanks for taking my questions, guys. Scott, I've got a modeling question. Now that you're done with the license option for software, how should we think of the trajectory of the run rate of that license line going forward? I think you do allocate some of the subscription sale to license. Is that correct?

That is right, Matt. So if you look at that line, there's two components to it. One is perpetual license sales that get immediately recognized and of course, that piece goes to zero. The second is that line is actually a license and other, so there's some -- there's a grab bag of smaller items in there, some legacy products, little bit of consulting in there, so that piece, we've talked about that historically. If you go back a few quarters when we were selling perpetual licenses, that piece being about 10% of the total. That other piece will continue, the perpetual license piece goes to zero. The third piece is what you just said. Both the product subs and the EBAs, of course, when we sell it, it goes immediately into deferred revenue and gets recognized ratably. But as it comes out of deferred, a portion of it -- because if you think of a term license, the customer is buying two things. They're buying access to the license and they're buying maintenance. So a portion of that deferred revenue accretes back into the license line and a portion of it accretes back into the maintenance line. I know it's confusing. I would love to say that I'm changing that today. We are working on changing that so that it's much clearer for you to be able to see how subscription revenue comes back into a line called subscription. Maintenance is in the maintenance line. And then what's left in license and other really is mostly other at that point.

Yes. As I said, it's roughly 80% of the product subscription and 55% of the EBAs get recorded as license revenue. So it's way more than, I think, many people have in their models. And we're doing everything we can to change it and try to clarify it so that it's much clearer in terms of what, I think, everyone would just affectionately refer to as subscription. So we'll try to make this as clear possible, but hopefully, with -- there's two pieces of data. It's a little bit easy for others to model what's going on.

Thank you. And our next question comes from the line of Kash Rangan from Bank of America Merrill Lynch.

Hey, guys. Sorry about that. Yeah. Thank you so much. Sorry about that. I wasn't sure if this question has been asked earlier, but I guess, more for Scott. What is your view on the new accounting standard that is being proposed by the FASB being reviewed by the SEC? I think they call it the ASC 606 that particularly applies to software companies that are trying to move into a more ratable model. How much of it does apply to Autodesk? How much of it does not apply to Autodesk? Just curious about your thoughts. Thank you.

Sure, Kash, it applies to everyone. It's not just Autodesk or non-Autodesk. It applies to everyone who's selling software. What we've been doing on that front, and I think we talked about this when you were in town a couple of months ago, we have been very actively engaged on this as we've done everything that we have to shift to a ratable revenue model, the last thing I want is to have that shift back based on the new rev rec guidelines. So our, I guess, he's now our controller has actually been a part of the AICPA task force that's working on clarifying the guidelines for FASB on how that will work. And if you think of a lot of our offerings, they're actually hybrid offerings. There's some term license in there, but there's a lot of cloud-based offering inside there. And so we are committed to continuing to maintain ratable revenue for our product sales for both the product subscription sales going forward -- obviously, cloud is pretty straightforward, and for EBAs. And we're staying very actively in touch with the group as those rules get clarified.

And one follow-up, if I could, for Carl, and I apologize if this question has already been answered. If yes, you don't need to. Any impact that you see from fiscal stimulus if the next presidential cycle brings that forth? Do you think there's any pent-up demand for infrastructure build-out globally and then particularly, in the United States that could help you guys? Thank you.

Yes, I mean, anything that's, first, a larger infrastructure spend will be good. I would say just generally speaking, if you look across industries right now, AEC seems strong in all the countries we talked about where the economies are good. Construction in the US is really strong. Northern and Central Europe is really healthy. I mean, if there's political changes, there's more infrastructure spend, that would all be to the good side. We're not counting on it. We've heard that promise before, so -- but it would certainly be to our advantage.

Thank you. And our next question comes from the line of Monika Garg from Pacific Crest.

Hi. Thanks for taking my question. First on the ARR. The ARR growth here was close to 10% but your long-term target is 24% CAGR, right? So this quarter, yes, of course, perpetual is impacting it, but for the next quarter, as you're not filling anymore perpetual licenses, should the ARR growth start trending towards 24% CAGR you've talked about?

It will, Monika. That's not intended to be every single year and every single quarter will be 24%, obviously. I know you know that. It's a CAGR over that time frame. You will see ARR growth rates accelerate in the second half of the year, partly, because you see the divergence in the ARR growth rates between maintenance on a slight decline and new model growing 86% year-on-year at constant currency as new model becomes a bigger piece of the total, that drives it up. But also, as we go forward, there will be a higher and higher mix of new model subs. And within that, the subscriptions growth that will continue to drive new model to a significantly higher level. So you will see that begin to accelerate in the second half and continue.

Thanks. So on the follow-up side, if you look at the revenue on the platform solution PSEB segment, that has come down Q-over-Q and year-over-year. Is there any reclassification in that segment?

There is. It's a continuation of what we've seen over the last several quarters, actually, since we launched suites three years ago, three-plus years ago since we launched suites because what drives the vast majority of that PSEB segment is AutoCAD and LT. And of course, AutoCAD is included in almost all of our -- in fact, it might be in all of our suites that gets sold. And so what’s happening is it's not that there's less AutoCAD going out into the world. What's happening instead is there's less of it being sold standalone and therefore, dropping into that PSEB segment and more of it being sold as a component in a suite and in the future, as a component in collections.

Thank you. And we have time for one more question today. Our final question comes from the line of Steve Koenig from Wedbush.

Thanks for squeezing me in. I appreciate it. I apologize if this question has been asked. I just wanted to maybe understand probably a little bit your thinking behind collections and your design philosophy for it? And just my background for this question was, we were looking for an offering other than desktop subscription that might be a little bit more appropriate for the mid-market. Seeing some resistance in the mid-market to the subscriptions, more so than for AutoCAD and LTs being sold as DTS in the low end of the market. And yet collections, it looks very much like the desktop subscription. It's a larger bundle at a slightly higher price point. Maybe the question, besides your kind of your intent behind it is, what are the dividing line be in the market between your T-Flex and your EBAs versus Collections when it comes to consuming the suites in terms of kind of company side? Is that how we should think about it and what might that size be?

Yes, so the first thing I would say is, so far, we gave a lot of proof points, hopefully, tonight. We haven't seen the resistance. And as a matter of fact, much of -- many of the new products and desktop subscriptions for both suites and collections is being sold through our channel partners, which serves the broadly speaking mid-market. So we think collections are a nice way to both simplify and get people the easy ramp to the new cloud services, so what they're familiar with plus the new stuff. Let me jump to the other side of it is our EBAs -- and as you mentioned, the technology behind it, the T-Flex, is really in the less than the top 1% of our accounts. So it has a lot of room to grow before these two things run into each other. And what I would add is I would expect to see more flexibility in our product offerings going forward. Collections is really just a first step in the door of what we're going to do in terms of making sure the customers have more flexibility. And that's regardless of how it will be sold. At the same time, we've documented and we've talked about how we really want to see our EBAs go out to a broader audience and how that will grow over time. So we think there's a lot of room between the two right now, and there's actually an opportunity to introduce other offerings that have that level of flexibility from T-Flex. But for example, it can easily be sold by our channel partners.

Thank you. And that concludes our question-and-answer session for today. I would like to turn the conference back over to Autodesk for any additional comments.

Thanks, operator. That concludes our call today. If you have any follow-up questions, you can reach me, Dave Gennarelli, at 415-507-6033. We'll also be at the Citi conference in New York on September 8. Thanks for joining us.

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 great day.