Autodesk, Inc. (ADSK) Earnings Call Transcript & Summary

Excellent. Thank you, everyone, for joining us this morning and afternoon over in London. My name is Keith Weiss. I run the U.S. software equity research team here at Morgan Stanley. And very pleased to have with us from Autodesk, CFO, Scott Herren. Scott, thank you for joining us. And also, Simon Mays-Smith, who's heading up IR there now on a go-forward basis. Before we get started from our side of the equation, please see the Morgan Stanley website, www.morganstanley.com/researchdisclosures, for important research disclosures about this presentation. And Simon, I believe you had a safe harbor?

Yes. Just a safe harbor statement. During the call, we may make forward-looking statements about our outlook, future results and related assumptions and strategies. These statements reflect our best judgment based on current known factors. Actual events or results may differ materially. You could refer to our SEC important risks and other factors, including developments in COVID-19 pandemic and the resulting impact on our business and operations, that may cause our actual results to differ from those in our forward-looking statements. Forward-looking statements made during today are being made as of today. If the call is replayed or reviewed later, the information presented during the call may not contain current or accurate -- current accurate information. Autodesk has no obligation to update or revise any forward-looking statements. During the call, we made quite a number of numerical growth changes as we discuss our financial performance. And unless otherwise noted, each such reference represents a year-on-year comparison. All non-GAAP numbers referenced in today's call are reconciled in our financial statements, which are available on our website at autodesk.com/investor. Thanks.

Awesome. So Scott, again, thank you for joining us.

So I want to start off the conversation kind of high level, talking about the macro environment. You guys recently reported your Q3 results, really solid quarter. You beat across sort of all of the guidance ranges, revenues, sufficient revenues, operating margins did really well. And it seems as if you've turned the corner, at least from sort of the macro environment. One, would you agree with that statement? And two, how are you guys thinking about sort of the state of play or sort of the demand environment as we head into calendar Q1?

Yes. It's a great question, Keith. And we -- as I reflect back on the year, and I'll give you some insight into the trends. And I'm not going to comment on Q4, while Q4 is in flight, right? You know we're in our fourth quarter right now. But obviously, everyone -- [ that's ] along with everyone else, took a pretty significant hit in the first quarter of this year. Our fiscal year starts February 1. I think a pretty big hit in demand, and it was really centered more in new product demand. Renewal rates, as we've talked, stayed strong, stronger even than I would have expected, given kind of the breadth of the response to the pandemic. Partial renews, didn't see a material change in partial renews either. What we've seen since then is kind of sequential improvement. We talked about this in the Q2 earnings call. It was still down. New product sales were still down, but we were seeing sequential improvement. And we saw that again through the third quarter. It's not even, of course. It's certain countries and certain geographies have done better than others. We've highlighted a couple that have stabilized, and we're seeing modest improvement now in the U.K., but still not to where it was pre-COVID. And the same with the U.S., more stability, but not seeing any significant improvement yet in the U.S. And that's on new product demand. But what I would say is, and you know this, our cloud collaboration products have done extremely well through this, BIM 360 design, Docs and Fusion, which is a cloud-based product as well.

Got it. Got it. One of the things that -- it's pretty interesting if you're looking at sort of the 2020 downturn versus what happened in '08, '09. Software, overall, we've moved from, by my analysis, like 50-50 perpetual license to recurring revenues. Now we're 80% plus recurring revenues. Companies like Autodesk are even further beyond that in terms of the recurring revenues. And the business model has worked, right? Like there's no revenue decline. You're still yielding good cash flow and good operating margins throughout the downturn. So from a business model perspective, very successful, and makes a lot of validation why we're all headed in that direction and took that [ train ] in terms of moving to subscription model. Was there any change in terms of like the fundamental demand trends in terms of -- you talked about sort of new business being down. But does the subscription model and the different pricing dynamics, does that change kind of -- you talked about a little bit with the renewal rates. But did that change like the fundamental demand trends and help keep demand more steady or less volatile through this downturn?

Yes. It's a great question. I'd say I'd highlight 2 big differences for us as I compare back to the way Autodesk performed in calendar '08 and '09. At that point, we were -- the vast majority of our revenues, as you know, were from perpetual license. And new license sales declined during the depth of the pandemic by about 40% year-on-year, 4-0. Compare that to the metric we gave you on our Q2 call that new business -- new volume sales for product subscriptions, which is what replaced perpetual licenses, we're down in the mid-teens, right? So it's a significant difference in the demand environment. Some of that is due, obviously, with the strategic nature of our product. Some of that is due to -- if you owned a perpetual license and you had a reduction in force, which many people did in 2008, you could put that license on the shelf. And as you hired people back -- once the economy recovered, you hired people back, you could consume licenses you already own. For the subscription model, when you let that subscription go and you start to hire people back, you immediately have to go pull in a new subscription to support them. So some of that is kind of due to the nature of the business model, but a significant difference in the demand for new subscriptions versus what we had seen historically. I think the second thing, and this was a little bit of a surprise to us, if you go back to '08 and '09, the canary in the coal mine, the leading indicator that, in retrospect, we figured out was the AutoCAD LT. So our very lowest-price, kind of entry-level product took the hit first back in '08 and '09. We saw a pretty significant decline in AutoCAD LT, and that presaged the rest of the way the business declined back then. When we headed into this, the slowdown in the first quarter of this year, we actually saw the reverse. We saw a more -- LT as a percent of our total mix, add up all the sales, add up to 100%, the percent that was AutoCAD LT actually increased slightly. And it's -- again, it is our lowest-priced sales to the most price-sensitive part of the market. That was a little bit surprising. I'd say it was a short-term effect because by the time we got to -- through the second quarter and into the third quarter, the mix had reverted to its historic norm in terms of the percent of sales that go to each product type. But that was a pretty significant difference versus what we've seen historically.

Got it. Got it. Got it. One more kind of like business model/macro question and then we'll get into the business drivers a little bit. As we get into a better macro environment into calendar '21, and we're seeing that business improve, one of the things I'm always cognitive of is subscription models, they -- it's a very favorable kind of dynamic for operating margins when demand and growth slows down because your amortizing revenue is off the balance sheet and you couldn't move OpEx in the real time. But you got to be a little bit cautious on the way up, right, in terms of the revenues are going to lag. But you guys need to invest sort of ahead of demand. Anything that investors should be careful about or cognitive of when we're thinking about Autodesk into the forward fiscal year in terms of that dynamic?

Yes. That's exactly why we gave you some of the color and some of the kind of the early glimpse of what our expectations are for fiscal '22 on revenue growth and on the free cash flow. So the -- if I step back, my expectation for fiscal '22 for the overall macro environment is that it's going to be a more back-end loaded year, right? We're seeing sequential improvement through Q2 and through Q3 from the shock that hit in Q1, a little bit harder to tell what Q1 of next year is going to look like. But certainly, by the second half of the year, I'm expecting a far more robust economic environment. And so what that will mean is our sales are going to be a little more back-end loaded, and of course, revenues lag sales, right, with a subscription model. So that's the reason you see the kind of the low to mid-teens guide we gave on revenue growth for the year. That back-end loaded nature hurts revenue growth in fiscal '22. Obviously, it's a tailwind to growth in fiscal '23 because all of that then accretes into the revenue stream in fiscal '23. From a spend standpoint, I expect to see margins improve next year. So -- and spending will increase as well, correct? So we're at a point in the model where we can both grow investment, grow spend in some of the key areas we've talked about investing in, and we can come back on that if you want, and expand margins, right? So margins will expand year-on-year. Some of the growth in spend will be a return of kind of windfall savings we got this year. P&E is an example that everyone points through P&E fell dramatically this year. It's not going to go back to where it was in calendar '19 next year. Part of it is it at that level, part of it is there's not going to be a willingness for our employees to travel. Part of it is there's not going to be a willingness for their customers to invite them onto the site, right? So travel is going to be a modest increase year-on-year, but it's not a huge increase year-on-year. There's some operational savings from the office shutdowns, it's nominal in the scheme of our overall spend rate. So we expect spend to go up more driven by investment and, not in a huge way, but also expect margins to expand in fiscal '22.

Got it. Got it. That's a great background. I want to shift the conversation a little bit more kind of fundamentally and really focusing on the growth drivers. And where I'm coming from, sort of as the investment analyst, you guys put forward a framework for how think about Autodesk growth on a going-forward basis. And you've put together a framework for durable mid-teens growth at Autodesk over the next several years. And from my position, if you believe in 15% as a marker, if you believe in that durability of that 15% growth and you look at sort of the margin profile that you guys are talking about in the free cash flow growth profile, stock looks really undervalued versus your peer group and versus the software group that's very expensive. So it implies that investors are a little uncomfortable with some parts of that equation, of that durable growth equation. So I want to take the opportunity to walk through with you some of those core variables in that growth equation and see if we could sort of vet out where there's more variability and which ones are you're most confident in. So I guess, maybe if we start from the kind of the bottom of the stack, if we think about user growth in your core areas, in any of your core businesses, I think that's something you guys put at like a 2% to 3% growth rate, what drives that? What drives kind of the overall kind of growth in users? Is it just kind of like this is GDP-type growth? This is -- are architects, engineers are increasing in numbers globally? Or is part of it is the expansion of your ability to get to them?

Yes. So let me go back to the first part of your comment, Keith. Well, we talked about was double-digit growth, so we didn't put a 15% on it. We just said beyond sort of [ victory ]...

I put it that way.

I know. I know. But it's important for me to put the counter out there. So think of that as double-digit growth. By the way, that's both revenue and free cash flow beyond fiscal Q3, right? So it's kind of a longer-term view of our expectations of top line growth and profitability growth. I'd say the -- to your specific question, you're right, it's a lot about just overall growth in the industry. And you know the -- if you just focus on the design piece of our business, we also have a growing make piece of our business with the design piece, which is dominated by CAD-type products. The CAD industry grows 6% to 8% a year, right? And it has for a really long time. So there's -- so some of that is exactly what you talked about, it's GDP growth, more people in the workforce, more people on the earth, right? Those people need places to live and places to work. So that's part of the growth. I think if I could take the next step on that, though, because by far, the -- and by the way, if you go back and look at that slide that talked about double-digit growth longer-term, beyond fiscal '23, we try to put those things in priority sequence for you, right? So the most important at the top and then worked our way down for the Autodesk-specific factors. The biggest factor that's going to drive that is the size of our renewal base. And we've talked about our net revenue retention rate, NRQ -- NR3 rate. Think of that as it frees a cohort of customers. Everyone who's paying us in a given quarter look at that exact same cohort 4 quarters later. How much are they paying us? Some will have turned out, some will have bought new seats. Some will have gone from -- within their portfolio, gone from AutoCAD to Revit, for example, or from Revit to an AEC collection. So there's some upsell and some seat expansion within that base. That had historically been running pre pandemic at a 110% to 120% level. So we've got an enormous renewal base. And that renewal base, again, pre pandemic, had been in that 110% to 120% growth rate. That's a big driver of growth long term. Even through the pandemic, that's still been between 100% and 110%. So we've got, from a SaaS model, basically negative revenue churn on that large renewal base, right? It's a positive number each year. That's the biggest factor that drives our growth long term. Layer on noncompliant -- the ability to monetize noncompliant, layer on the secular growth opportunity in construction, layer on what's going to be beyond fiscal '23, a burgeoning opportunity for Fusion 360. All of those factors are what gives us confidence in that double-digit growth longer term.

All right. It is going to be a very short fire-side chat if you answer all 6 of my questions all in 1 [indiscernible]

We just started talking about the [ beer ] at that point. How about that?

That'd be good. But digging into that, like, you're absolutely right. It's the dollar net expansion rate. That's B be driving that business, sort of driving the growth on a going-forward basis. Digging into those variables a little bit of what drives the average customer to expand their spending year in, year out by 10% to 20% on a go-forward basis, I guess, one thing that I -- the first thing I want to hit on is the kind of ARPU expansion, right? And there's various ways to get that ARPU expansion. One is get an individual customer to get more solutions. And I guess the best way to think about that is the move to collections, right, and then who's using a suite, who's using point products? Can you give us kind of a -- the reminder of where we are today in terms of collection adoption and how we could think about how that trends over time?

Yes. Collection adoption, we went from 0 -- so to step back, for the people who haven't followed us as closely as you have, we sold suites historically We've sold product bundles. And if you remember, we had 7 different suites. It was the classic overthinking, right? Like "hey, this one's targeted at 1 specific use case. This was targeted. And each of those had a good, better, fast version, right? So 21 different sites. We launched collections when we moved away from perpetual licenses and onto pure product subscription sales, and we simplified it to 3. If you're and in AEC, here's your collection. You're in product design and manufacture, you have a collection in the M&E. That -- so it started at 0 going back to midyear 3 years ago, and it grew rapidly. The adoption of those collections grew more quickly than the initial adoptions of suites had, right? In some ways, they were the successor to suites, but it grew quite rapidly. I'd say it's at a point of stability in terms of our overall mix right now, the percent of sales that are in industry collections. We continue to drive more value into the collections, and we continue to see people buying up into those collections, right? But if you look at the size of the overall base, especially on a seat count basis, it takes a huge swing now to -- or a huge number to swing that mix between that overall product mix and collection. So productions have been super successful, and we're continuing to add unique value into collections longer-term to drive that upsell.

Got it. So collections is 1 avenue. Just outright price increases, which you guys have talked about historically, what was like 2.5% every 2 years would be kind of the maximum that you would push through?

Yes, right.

Great. What other factors should we think about in terms of levers that Autodesk has or to drive ARPU higher?

Just sticking with price for a moment because I think that's where you want to go. There's also been adoption, right? So well -- especially in architects, moving from classic AutoCAD package for architectural design to Revit that produces the building information model. And one of the -- I think one of the really key slides that we had in our last Investor Day back in June was the one that Lisa Campbell showed that gave you an update on BIM adoption rates. So if you go back and pull Lisa Campbell's deck, there's a map of the world, and it's color-coded by the adoption rates. And even in the most -- everyone thinks Revit's such a standard, 100% of architects. That's not the case, right? BIM adoption in the most penetrated markets. It's in the roughly 50% range. So there's still a lot of room for revenue growth in customers moving from 2D AutoCAD to 3D building information model via Revit. And obviously, there's a higher-value product, there's a higher price for that.

Great. I think that's a good jumping off, pointing into 2 directions. Because the building information model, it's important both in terms of it's a higher-value solution for your end customer. It adds a lot of value for your end customer because they could really reduce the sort of the wastage in their construction process. But it's also kind of the -- I think about it as the bones or sort of the structure upon which your broader solution portfolio, going from the plan to design to build. Now this is the database that everyone is using for all those workflows on top. So maybe it's a good jumping off point to another area of kind of -- sort of durability of growth, which is the fuel construction opportunity. So this is something you guys have been building up in terms of the solution portfolio over the past couple of years. It talks a lot about our -- I don't want to use University about where this portfolio has gotten to in terms of the cohesiveness and the integrated nature. Maybe you could give us an update of kind of where are we on that sort of new construction opportunity, how are you feeling about the product and where are we in terms of adoption amongst your customers?

Yes. So just to start with, what -- the way you described the building information model is kind of the backbone, the central nervous system of our -- is exactly right, right? And Andrew has talked several times about taking the building information model straight through the construction phase. And some of the mandates, some of the -- what's driving that adoption of BIM, there are regulatory mandates in certain countries, but there's also building owners who say, "Hey, at the end of this project, I've got to operate this building for decades, right? I want the as-built version of the building information model. So it is the kind of the core that drives that. I'm super pleased with and proud of the work that Jim Lynch and our Autodesk Construction Solutions team have done. And I'm sure you spent time at AU kind of getting the updates there. It's -- some of the growth that we've had, that we had an organic business going after site construction. We obviously invested in some inorganic growth there with the acquisitions a few years ago of PlanGrid and BuildingConnected. And Jim had the unenviable task of pulling all that together, 2 founder-led companies, along with a large internal team, pulling it all together, identifying the areas of overlap and basically saying. We're going to stop doing that. And that's a hard thing for an engineer to hear when they put their heart and soul into something. He's done a great job on that. We've had to integrate that so that you don't lose fidelity when you go from step-to-step through that workflow that you described. They've done that with the launch of Project Golden Gate and the Autodesk Construction Cloud, which is the public brand for that. We're getting really good traction with it. I think it's still super early days. And obviously, the pandemic affected the new sales of the Autodesk Construction Cloud, but some of the elements are growing very rapidly. Obviously, BIM 360 design saw a huge uptick during the pandemic because people had to find out how to get their job done fully remote, right? And that's like Google Docs for a Revit design, right, for a BIM 360 design. BIM 360 Docs grew rapidly as well. One of the things, just to give you a point on how important this is to our biggest customers, we talked on the earnings call about a 9-figure deal. You probably fell off your chair when we talked about a 9-figure deal didn't you, right? Because no one thinks a lot of this because I've talked with...

Yes. That's not part of the purview of Autodesk.

Yes, exactly, which is part of why we wanted to highlight it. The reason that deal got to the size it did, part of it is -- and by the way, it was an EBA renewal in the AEC space. So we knew it was coming. We knew it was going to be quite large. They have standardized -- that company has standardized on using not just all of our design tools through the front end of the workflow, but also the Autodesk Construction Cloud elements on the back end. And so that's just another example of how strategic it is, not just as a new TAM and a new market opportunity for us, but to our existing customers to drive into that construction space.

Yes. Got it. And just one last one on this because we have like 5 minutes left, and I have a lot of topics I want to make sure we touch on. In a lot of other industries, people have been talking about how the COVID and the crisis we just came through could be a catalyst for spurring people towards digitalization. It seems like construction maybe is prime for that. Is that a correct assumption on our part that, that could really catalyze the adoption of these more digitalized ways of doing the building process on a go-forward basis?

Yes, absolutely. It has. And we've given you some of the stats on that. I mean the human toll of the pandemic has been horrendous, right? So obviously, no one wished for it. But if there's a silver lining to it, though, it's really accelerated the adoption of cloud-based digital products. I hesitate to say digital transformation because it's such a buzzword, bingo-type word anymore. But what we've seen is very rapid adoption, both really across the spectrum, but focused in on -- I talked about earlier, BIM 360 Design, BIM 360 Docs, both of those have seen an enormous growth in the number of projects started on them. And then we had that extended access program, where they could get free access for what ended up being 90 days. We converted a ton of them to paying customers, and we continue to see new project growth beyond the end of that, right? So we're seeing a really strong adoption of those tools. If you think about the impacts on the construction industry, they've been pretty severe with the way sites shut down, and this is an industry that runs on very thin margins to begin with. So one of the other things that fuels this is, hey, okay, we had to shut down. We had to find a way to get our jobs done. As those job sites have reopened, there's new safety protocols. They've got to find a way to manage those. They're using our tools to manage the new safety protocols. There's also a -- it's such an uber competitive business. When you think of these multibillion-dollar projects that are going to run on 3% to 5% bottom line margins, it's a very competitive space. And as 1 or 2 competitors start to adopt these digital tools and get an advantage, there's almost a viral effect, right? It's a different viral effect, but a viral effect, where everyone has to get onto this tool set. So I think it's -- there's no question that this has accelerated the adoption of digital tools much more quickly than they would have otherwise gone in the construction space.

Great. Got it. I want to shift gears a little bit, talk about sort of the opportunity amongst noncompliant users. And I wanted to just pick your brains to help us kind of balance. On 1 side of the equation, huge opportunity. You guys talked about 12 million-plus potential noncompliant users. On the other side of the equation, you guys don't want to be overly aggressive in how you approach that side of the equation. It's a very methodical -- you're partnering with these customers to get them into compliance. It's almost -- the way I think about it, it almost seems like those like free-to-paid user bases that other companies are going about is that there's this big pool of, not really free users, but they're kind of free users. That you have an inside sales effort that's going after to try to sort of convert these guys into being paid users, they're being -- to becoming compliant. How should -- we understand the size of the opportunity, how should investors think about the pace at which you're going to be able to convert those guys?

Yes. Some of the conversion happens exactly like you said. We identify -- through the telemetry in the product, we identify where they are right on list. There's 20 here, but there's 30 there, right? So we created the list, and there's an inside sales process. And that's going to be -- that's -- as we've gotten better at the telemetry and better at identifying, the productivity of those inside salespeople has gone up. And as their productivity has gone up, we've invested more. We put more research in. So that's going to continue for quite some time. There's a couple of other things that I don't think we've talked about enough, Keith, that is worth discussing here that we're doing. We're also -- we talked about in-product messaging and getting that rolled out. And so what that means is -- I won't use your name on, he's mine. Scott Herren logs in. And [ the thing ] comes back, says, "Hey, he's using a known track license." there's a little nag screen that pops up, right? And first time it pops up, you can dismiss it and get on with your day. We can continue to accelerate the length of time that it comes -- that it stays up. We can eliminate the ability for it to be dismissed. We can vary the size to where it can cover almost the entire screen of that nag screen, right? So what ends up happening, as we built that out and built it in first into the most pirated versions of our software, right, there's 5 titles that drive 90-plus percent of our license. That also drives conversion because, in many cases, the user who's being annoyed is not the buyer. And as that becomes more and more annoying and their productivity takes a bigger hit, they lean on the purchasing department there [indiscernible]. The second thing that we're doing here is kind of going through that process of shutting down. We can say -- so let's say, I'll use me again. I pop up the nag screen, and I'm like, "You know, this is killing me. I'm going to go sign up for an education license even though I'm not a student, right?" And then -- and for a long time, that was an eligible path where they could have, for some length of time, free access to the software. We shut that down now with having a process where you have to be certified to get access to education licenses. So we're -- Andrew described it as locking the exit doors, right? We're locking the ways that they can go get other unauthorized access to the product, trial licenses, right? We're hardening that as well. So not only are we going after it with the inside sales process that you described quite well, we're doing other things to both harden the product and make it harder to pirate, but also to -- other ways have been sending users to say, "Hey, this has got a stop. We have to go get a compliant license."

Got it. Got it. One last one for me. One last one because we're running out of time. I want to talk just about free cash flow and the sort of the path to that FY '23 target, macro kind of like, it makes it a little bit more difficult. And I think one of the investor concerns is kind of like the hockey stick up into FY '23. Can you help us garner confidence in that hockey stick in some of the elements that enable that to pop up real nice in FY '23?

Yes. Some of it -- so I'll start with what you and I have talked about a couple of times. More of that free cash flow comes from the P&L in terms of net income, right? If you step back, there's really 2 big drivers of our free cash flow. There's net income, and there's the growth in deferred revenue, right? So more and more comes from -- and you've seen that already in our results, but more and more of our free cash flow comes off the P&L. So the things that I talked about, about driving top line growth, will drive a lot of that growth from '22 to '23. Bear in mind, too, the back-end loaded nature of fiscal '22 is a headwind to our growth in free cash flow and revenue in fiscal '22, but it flips around and becomes a tailwind in fiscal '23, right? Because this back-end loaded, more of the sales that come late in the year, that cash gets collected in fiscal '23. So the back-end loaded nature drives some of that, the growth in construction that we've talked about. All the other secular drivers that we see will drive some of that growth in the P&L. There will be a small growth in deferred as well. That's the -- I've been describing it as the echo effect of the multiyear sales that we saw in fiscal '20 when we relaunched that multiyear offering. Pay for 3 years upfront, you get a 10% discount. That ramped up during fiscal '20, will all those 3-year deals come due in fiscal '23. So there will be some growth in deferred revenue that comes off the echo of what we saw in that big ramp of multiyear in fiscal '20.

Got it. Got it. That makes sense. Outstanding. Unfortunately, that takes us to the end of our allotted time slot.

Way too short.

I did -- I could talk for hours on -- such an interesting story at Autodesk. So many really compelling new product initiatives going on. It's really a fascinating story. I did want to end on a note of thank you for the partnership over time, but this is going to be our last time speaking with you as a CFO at Autodesk, so congratulations on the new role. But really, it's been a pleasure working with you. And you've been a great steward at Autodesk for the past couple of years, and -- so thank you for that. And again, thank you for joining us at the conference today.

No. Likewise, Keith, you've tracked the story since I got here 6 years ago as well as anyone, and I've always enjoyed the give and take. We actually haven't always agreed, and that's even better, right, because I think we get a [indiscernible]. So thanks for the partnership as well.

Outstanding. Thank you very much, guys. And thank you, everyone, for joining us today at the NASDAQ conference.