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

Happy Thursday. Welcome to day 2 of Citi's Tech Conference. I'm Tyler Radke, Co-Head of Software Equity Research. Thanks for joining. And we're really excited to have Andrew Anagnost, the CEO of Autodesk. Andrew, thanks for coming out from the West Coast for our conference. I know you wanted to read an exciting safe harbor statement, so I'll turn it over to you.

Yes, my lawyers have provided me with content for this. So we may, look, make forward-looking statements during the course of this presentation. Please refer to our SEC filings for information or risks and other factors that may cause our actual results to differ materially from these statements. There, I've done my job.

All right, there you go. Well done. got that out of the way. Now we can start off with...

The harbor is safe.

Yes.

Well, thanks again for being here. I thought we'd just kick off big picture. So Autodesk has been around for decades. It's a company with very dominant market share in the architecture and engineering space, you have flagship products like Revit and AutoCAD. One of the big questions we get from investors is just sort of how to think about the growth profile? Architects are not exactly a high-growth profession, you have very high market share. So how do you sort of think about the building blocks and what that long-term growth looks like?

Yes. And architecture is somewhat an overstated percentage of our business. But the more important question is, look, you can't think of Autodesk as a design company anymore. We're a design and make company. And increasingly will also be a design, make and operate company. So the TAM that we explore now is significantly larger than the TAM we explored in the past. And if you look at some of the proof points of that, the biggest proof of where we're going is the growth in the make part of our business and the percentage of our business that is make and that it continues to grow significantly faster than the total company. So that represents this expansion in the design and make and deeper into the life cycles of all of our customers. The other thing that's really important to think about too is, when you look into the design side, the important thing is we're the 3D company, all right? There's nobody out there that delivers more 3D design tools than anybody else. And 3D still has quite a wave of adoption going on. So you've got to look at us both as a design, make, and increasingly design, make, operate company and as the 3D company.

Got it. Got it. Okay. So still a lot of growth ahead even beyond just design?

Yes, absolutely. Like I said, a lot of design still 2D, so this both in design, but the bigger driver for us long term is the design to make.

Got it. Got it. That's helpful framing. So turning to the demand side of the business, you recently reported last quarter. How do you sort of articulate what's going on in the macro environment? Obviously, there's continued challenges around interest rates. But at the same time, a lot of your companies have -- or a lot of your customers have long backlogs. There's big infrastructure projects, data center builds. So what are you seeing when you survey the land?

Yes. So you're probably aware that there's still a large construction backlog. It wobbles between 6 and 8 months every quarter. That backlog still has to clear and that's obviously driving a lot of activity in construction. Now that -- look, if slowdowns in the front-end of the business continue for a long period of time, that backlog shrinks, all right? But it would have to continue for like 3 more quarters for us to see that, that backlog can reasonably start to shrink. So there's always something going on. You mentioned data centers, right? All these data centers we're involved in that. Mid-market manufacturing, we're having a factory boom in the U.S., guess who's involved in building those factories and increasingly in outfitting the interior of those factories, that's us, all right? So there's lots of things that shift around whereas the whole sectors might be seeing some slowdowns, there's smaller sectors, smaller verticals in there that aren't.

Right, right. Got it. And I guess, specifically on the data center build, what does that look like for you? I mean, is that a combination we talked about design, make, operate. Are you seeing good traction across all those segments of the business?

Yes. So we're absolutely seeing nice growth in the make piece of the business. Obviously, we're in the design piece, but we're also seeing growth in the make piece. And increasingly, we're having conversations about digital twins associated with some of these data centers. So we're trying to get them across the entire portfolio. They are also -- data center build-out folks are some of our biggest adopters of our virtual reality tools as well. They love those tools for helping them plan out and navigate some of the work they're doing. But the make side is absolutely a big part of what's going on there.

Got it. In recent years, there's been a lot of talk about the infrastructure bill here in the U.S. and I'd be curious, just as you look back over the last few years, how has that played out relative to your expectations? And obviously, we're heading into an election cycle where there's a lot of talk from both parties about infrastructure. How are you thinking about the next few years as it relates to U.S. infrastructure?

Yes. So there's a couple of -- so one, we've always kind of tried to position this as this builds over time, and it does build over time. It takes time for the money to show up. One of the places where we've actually seen the money show up is in that little piece of the roughly $300 million that's in there for helping DOTs modernize. That's one of the ways we ended up in PennDOT and things associated with others, some of the Department of Transportation in the U.S. is because they're actually exploring new ways of executing digital processes as a precursor to spending some of this infrastructure money. Now you're also starting to see some of these contracts actually being awarded and entering the design phase. And then those will enter the construction phase and new projects will enter the design phase, but this will build over time. The good news is that infrastructure is a bipartisan issue. This isn't going away. Everybody loves infrastructure. And you're going to see the build up over time, but it's going to be a buildup, right? There's no infrastructure spike. And I hope we've been consistent about telling that story and making sure people understand that.

Yes, yes, absolutely. And as you think about the products that are exposed to that, is it -- I know you have multiple products that would sort of feed into that. But how do you feel about your competitive positioning in terms of your peers being able to take advantage of that?

Yes. So we're -- we definitely have growing strength in road and rail. That's driven primarily by our Civil 3D portfolio and some of the investments we've made there. That has increased our competitiveness. But the connection to Civil 3D in the cloud and ultimately to construction is really helping us with a lot of these efforts. And obviously, we placed a big bet in water. Water is a long existential bet in terms of the world's water infrastructure. So you're going to see us -- definitely see more activity in water infrastructure as we move forward. That was a strategic bet by us. So we're really focused on road, rail and water.

Got it. Got it. The topic of BIM mandates has been in the Autodesk conversation for a number of years. Every geography sort of has different requirements. But maybe if you could just frame how important is the progression of that around the world in terms of evangelizing and driving the digitization of the industries you serve? And where are we at in terms of that rollout?

Yes. BIM mandates are really important, all right? And they come at multiple levels. They come at the government level and then they come at the owner level, in terms of the project specifications. So we've seen BIM mandates like in the U.K. and other places, successfully drive the adoption of BIM. I mean there are firms that were not going to BIM, that are moving and have moved their infrastructure to BIM because the U.K. government mandated BIM for U.K. government projects. More and more, you're seeing owners mandating BIM for the projects because they want the output at the end of the digital twin when the project is complete. So the mandates are absolutely something that helps drive the transition. You know that we're not fully penetrated with BIM, not even close. The U.S. is where we most try to penetrate, and we're still less than 50% penetrated in the U.S. market. So there's still broad penetration left and the mandates are definitely a tailwind for that penetration.

Right. And is that sort of driving more sales of Autodesk Construction Cloud, like what products do you see kind of attached...

It drives Revit. Revit the most, and upsells to Revit is what it drives most activity on, right? And then, of course, once you get up into Revit and once owners are like interested in digital twins and all things, it starts inserting into the process. So what happens is it drives business and Construction Cloud because people want something that's connected to the BIM processes being done upfront. And increasingly, projects, especially these infrastructure projects are going to be looking at a BIM orientation more and more.

Got it. Got it. I wanted to shift and talk a little bit about Autodesk's AI strategy. So I think last November at AU, you sort of, I guess, introduced the terminology of Autodesk AI and sort of described it as more of a vision of where you're taking the company. Can you just recap sort of what is Autodesk's AI strategy? And how are you looking to sort of build out AI-enabled products over the coming years?

Yes. So if people haven't noticed yet, AI is a journey, right? And we're going to be continually investing in AI at a steady rate for years to come because it's going to become the basis of competition in multiple markets. The way we look at this is we look at it two ways. We want to deliver specific designed productivity within their existing workflows today. And we want to also work on the things that are going to change the way they work in the future. And we want to do these at different rates and speed. So the large chunk of our investment is driving productivity in the way they work today. It's very, very hard and very time-consuming to create 3D models and to create these beautiful detailed representations of things that are being built, whether they be machines or buildings. If we can reduce that time from hours to days or months to weeks, we're providing huge value. There are several slices of things that we can do for our customers that we're focused on. One of them is we've introduced a drawing dimensioning automation tool out there. It's getting heavy pickup with our customers. And that's just massive productivity for them because drawing stack creation is not a huge value-added activity. But if you can automate it with an AI that understands your dimensioning strategies and it trains on your dimensioning strategies over time, that's a huge return. There's other returns that get into the arcane ways of how people build 3D models where we can automate things that today take them a lot of time and effort or they just do flat out on. So you're going to see us do a bunch of those things. At the same time, you're seeing us roll out things like Bernini and then some of the 3D shape creation tools, some of the advanced stuff we're doing with analytics and Forma, that's frankly going to change the way they work in the future. But we're doing both of those, and the two of those will come together ultimately as the global AI strategy of the company.

Got it. Got it. And one of the themes that we hear from investors in AI is sort of the -- well, there's optimism perhaps on charging more price and higher value products, there's concerns that maybe the Q side of the equation could come down over time, right, perhaps if designers or architects are getting more efficient, does an organization need fewer of them over time. So how do you sort of balance that risk of potentially fewer seats? Or maybe it's just, hey there's just always more projects for these folks...

So the good news is we serve an industry that has a universal capacity problem, right? There is not a single segment in the industries we serve where they have enough money, people or materials to build and rebuild everything that needs to be built. They are looking for productivity. More than anything, what's driving the adoption of construction tech, it's not that somebody is waking up and saying, I want tech. It's like, I can't compete anymore, all right? Productivity is a huge issue for them. They can't -- their workforce is not going to get any bigger. So we see a pipeline of opportunity for each of our customers growing. So will there be fewer people per project in the AI-driven world, Absolutely. Will firms be executing more projects? Absolutely. So they'll be deploying their existing capacity differently, and that's going to fundamentally solve a capacity problem within the industry.

Right. And maybe also just a volume tailwind for the make business and [ operation ] Business.

Yes. Absolutely.

Got it. Got it. And obviously, one of the important parts of AI is, it starts with the data, right? You have to have good data to feed into those models. I think it was a couple of years ago, you announced the industry cloud model with Forma kind of following some of the foot steps that you did with Fusion 360. Where are we at in terms of that evolution from file-based to this industry cloud data model and when do you sort of expect that to be ready?

Yes. So we're still -- it's still on a journey. I mean the most advanced is Fusion. Fusion has a large amount of data that's in the cloud. A lot of that data is served up in a granular form to our customers when they need it, and we're capitalizing on that moving forward. In AEC, we just recently rolled out the APIs for the AEC data model, which gives customers access to granular data for that. The key goal here is, right, the basis of competition is no longer file format. The basis of competition is who connects your life cycle processes better. And our goal is to make sure that people come to our clouds to connect their life cycle processes designed through make regardless of which tools they use at each point in the process. That's going to be the basis of value creation moving forward because our customers are struggling with connecting all these things together. So we've rolled out the AEC data model. That's going to get continuously refined over the next couple of years. And then, of course, we're doing the same thing in media entertainment with some of the things we're doing with production management and onset capture of data, trying to start digital and stay digital, and help the customers do that from the get-go. So all of this is kind of on an arc of increasing. Fusion is out ahead, M&E is kind of behind them and then AEC is moving along.

Got it. Got it. Okay. One of the major company-specific changes at Autodesk over the last year has been this new transaction model rollout, which I imagine you've been talking a lot about with investors today. But we've obviously fielded a lot of questions around the financial impacts and how to sort of think about the nuances on billings, and I'm sure we'll get into that. But maybe start high level, strategically, why are you doing this? And maybe for investors who haven't been following closely, what specifically is being changed?

Yes. So I want to go back to what we said last year in terms of what we were doing and why we're doing it. So what we're doing is, we're changing the way we interact with partners from a buy-sell model to an agency model, which essentially moves the account from being on their paper to being on our paper in our systems, all right? Now why are we doing that? The fundamental reason we're doing that, and this is something we said last year is, this gives us more visibility and control to our fully loaded sales and marketing costs. Read that as when this transition is complete, we are able to drive optimization in our sales and marketing organization. And there's several things that are going to drive that optimization. One is going to be a significant increase over time in self-service because the systems and processes that we do by moving away from the old -- by the way, no cloud company would start up as a buy-sell cloud -- just you would never do this, all right? So basically, we're unwinding the last vestiges of Autodesk sold ship boxes legacy. And that's going to allow us to drive a lot of self-service into the ecosystem. And that means that some customers that used to feel like they had to go to a certain low level partner that doesn't add a lot of value to get the access of the software, they're going to go directly to Autodesk, and they're going to manage that relationship directly with Autodesk, which drives a lot of efficiency and effectiveness. At the higher end, it gives us more visibility into who our customers are, what they're using and allows us to manage cross-sell and upsell, and then also drive efficiency in all of those processes. So look for us to drive increasing productivity from the sales and marketing functions as we move through this rollout, which, by the way, translates into margin growth over time. And that's why we started this journey, to get control of that and to flush the last vestiges of kind of the historical Autodesk out of the system. And we're on our way and it's working.

Yes. And as we think about the sales and marketing efficiencies, I mean, how much of it is just reducing kind of general purpose, let's call it, marketing overhead that you can't necessarily tie to a specific customer? Or is there a big amount of just cost that were in place to sort of manage the commission structure of this reseller channel?

Look, when you have a buy-sell model and you move to an agency model, there's costs that just start to naturally go away, all right? One -- the obvious one is there are no distributors anymore. That's an obvious one. The other obvious one is that as your bigger partners get bigger, the overlap of activities that you do inside the company to support those partners, you do less and less of that. So there's less and less overlap and you're redeploying that money to other things. And the other thing again is -- and I can't emphasize this so much enough, is that the growing self-service capability. This really, really drives a lot of efficiency. And like one of the things is -- just look at some of the things that happened as we rolled out the new transaction model. One of the things that we didn't expect, which we're actually delighted about is that more customers have elected to co-term contracts. Now -- so that means if they had four contracts with Autodesk through maybe more than one partner, they basically lined them all up to one renewal date. Why is that great? That's renewal efficiency right there. Instead of wrangling four contracts, you wrangle one at renewal time. And that drives a lot of renewal efficiency. Short term, it provides some noise in the billings. But that all kind of works its way out. It's a short-term headwind, and then it's a tailwind, than long term not in the system. So all of these things drive efficiency. And that's the thing we're chasing here and actually, we're really starting to see some of the benefits from.

Right, right. Well, I'm glad you brought up co-term, because we have gotten a lot of questions on that as you had. So maybe just going back on the co-term, like what you saw last quarter? And I know there was a bunch of moving pieces in billings, and we got questions on why were some billings kind of being pushed out. But walk us through the billings mechanics.

Yes. It's fairly straightforward, but I can see how it would be confusing. We always had co-terming factored into our guide, more of it happened than we expected, which we don't think is a bad thing. We think it's a good thing. But here it is. So I'm a customer, I got four contracts with Autodesk. One terms on August, another one terms on May, another one terms on January or something, some distribution like that. What happens is you line it up to the one that say terms in December, all right? And that means some of the renewals are like renew on a shorter cycle, so you kind of like clip off a chunk of that renewal to line them up. And that billing just disappears as you term-up, but then it pops back up the next year. And then after that, it normalizes, okay? So there's a short time and you're saying, well, why do you line up to the further out, that's just standard practice. That's the way -- that's just the way it goes, and that's why you line up the co-terms. You line them all up to the later one. So it's a short-term effect. That's why when we tell people to look at the new transaction model and the movements, it's better to look at the free cash flow build through that. Though long term, we can look away from free cash flow, but through the new transaction model, the free cash flow build is the one thing that's least affected by all of this and all the noise associated with it. That's what the co-terming is, not the impact.

Right. And in terms of what you saw last quarter and what you put into the guide, you said sort of surprised the upside by the co-term and so you sort of assume that, that behavior would continue as you roll it out?

Yes. We actually put a little bit -- we basically assume that the surprise would continue in Europe, which wasn't in our original thinking. So we just basically adjusted some of the things that have taken that into account. But remember, it's kind of neutral to the business over a 2-year period.

Right. Right. Right. I'm curious with this rollout now that you've seen more and more data points this past quarter. I guess, was there any surprises in terms of customers maybe just wanting to go more direct with you than as opposed to buying the channel? Like was there any surprises in that indirect versus direct mix?

Well, I wouldn't go so far as to say there were surprises, but I think a lot of you noticed we're up to 40% direct business, which is like 10% shy of that 50% long-term goal that we put out there. So obviously, I don't think that's due necessarily to the new transaction because it's just too new, and there was a lot of pull forward where some of the partners booked renewals early to get it on their paper. So it's not necessarily related to that, but you can expect to see more of that.

Right. Right. Got it. And then just in terms of the customer response to this, was there a confusion? Was there, I guess, happiness? Were folks kind of buying into this more simplified vision? Or anything you'd call out just on the...

Yes. Well, so we learned a lot from the Australia rollout about how to deal with customer confusion. One of the biggest things that confuse the customers is like all of a sudden, they have to put Autodesk as a vendor on their vendor -- approved vendor list, whereas before the partner was their vendor. So we learned from Australia that we need to take a little bit more customer education, a little bit more partner education to make sure that we got that right. And I think we got that right. Is it like 100% satisfaction? Probably not. Okay. Is it for the vast majority of customers like a nonevent? Pretty much, okay? Because now instead of having to call a partner, which a lot of them didn't want to do to get the transaction they can or if they really like their partner, and they're working with a partner that's part of that real value-added thing, they don't even see it. Except that they had to add Autodesk as an approved vendor.

Right. Right. So you mentioned earlier just how this transaction model is really designed to unlock some of the efficiencies which -- and give you better visibility into that. Have you identified sort of the magnitude of sales and marketing dollars that could be impacted by this? Or how do you sort of think about the size of that efficiency gain over time?

So without getting into the size of the magnitude, let's just say it this way. It would have been odd if we would have announced something as big as a new transaction model without having done the work to analyze what the opportunity was when it was done, okay? That would be very arbitrary, don't you think? Go through all that pain and confusion without knowing what's on the other side. So suffice it to say, we've done a lot of work internally to know what's on the other side. And we know that we're going to be able to drive the majority of our multiyear margin growth moving forward from that sales and marketing efficiency. And that was known before we ever started the journey.

Right. Right. Okay. So it sounds exciting...

It's really exciting. And frankly, it's the last big piece of modernization. I mean what we've been doing in the last 10 years it was like we've been modernizing the company relentlessly, moving to subscription, getting off of multiyear billings upfront, moving to this new transaction model, building out the back office to enable the self-service. All of this stuff pays off in efficiency dividends moving forward that are long coming and very exciting inside the company.

Right. Right. And last quarter, I mean, despite the moving pieces and stronger transaction model rollout, the implied margins were actually much stronger than folks expected. You were able to hold that flat despite taking on more of the commission expenses. So walk us through what were the factors that enabled you to effectively raise margins. Was that to what you talked about, some of identifying that marketing spend and making those trims? Or was it other areas of efficiency?

Actually, it was just more efficiency than cost. I remember, we do -- that 40% of direct, that's good business for Autodesk, okay? So a lot of this was just a natural trajectory of the efficiency that we're in. This is -- we didn't do any -- there was no -- if you're asking, was there some deliberate cut or -- there wasn't, okay? That was kind of the rate and pace of some of the things we were doing with the business, and they turned out to be pretty strong on an underlying basis. So a lot more efficiency. Some of it has to do with like moving out the new transaction model in Europe a little bit. All of these things have little puts and takes. But ultimately, it's kind of a testament to kind of the performance we'll probably see as we finish all of these things.

Right, right. Maybe turning to the construction side of the business. The make part has certainly seen stronger growth trends as of recently. I think you did call out some M&A tailwind in the numbers, but still growth is stronger than it was last year. What's going on in the make side give us a framing for some of the company-specific drivers that could be driving this resiliency?

If you take out the M&A in construction, our momentum was consistent, okay, which is good. There was no deceleration of momentum. There was a nice consistent momentum. Of course, obviously, acceleration of momentum if you take in the payouts and stuff associated with that. What's happening here is, it's a couple of things. It's coming up the maturity curve, all right? Autodesk Build is a highly competitive product now. It's built on a modern stack. The customers at the highest end like the design through make connectivity. They like the modern stack. They like the flexibility of our business model. They can choose which -- multiple ways of engaging with us. They like that flexibility. They like the work we're doing. And that's a positive factor in some of the new account acquisition you saw us doing in some of the results we talked about. But we're also growing internationally as well. And you can expect to see that to continue because we have a robust international growth machine. And we have a good footprint in design. We've got credibility in design and make now internationally. So look for us to continue to grow internationally as another tailwind to the overall growth in that construction business.

And on the go-to-market side, what are some of the things that you're doing differently and maybe more successfully with some of the -- I think about a year ago, you did a reorg of that business. What were the major changes?

We merged the construction sales force into the main line. Now we've also merged the construction R&D into the mainline. And there's obviously some efficiencies there, and there's obviously some overlap and lack of confusion that you get when you kind of get the sales forces on it. And we're seeing that. But we're certainly seeing it in terms of partner effectiveness, where our partner effectiveness is going up over time. Now like we said, that was going to be bumpy. When we merged it, there was a little bumpy, and now we're starting to see the benefits of merging those things back into the main line. And yes, there are cost advantages associated with that. It's expensive to have a vertical team that's running everything separately.

Yes. And then on the competitive landscape, your publicly traded competitors is called out some of their own go-to-market changes. What are you seeing in terms of win rates? Are you doing things more effectively, you think, to win deals from them? Or just any comment on...

All I can say is we see our competition more and we win more on that visibility that we have. So that's what we're seeing. And we're seeing more -- we're more and more invited to deals. So people know that we are a viable alternative and that the portfolio is at a place where people can look at this and say, okay, yes, that does what I need done. So we show up more, we get called in more, and we're winning more. That's what we're seeing.

Yes. Okay. The Manufacturing business is, let's say, continue to show steady above-market growth rates. Fusion 360, I know has been talked about over the years. Help us understand what are the key drivers of the manufacturing growth? And where are we at just in terms of the monetization of Fusion 360? I know you kind of use that as kind of your landing strategy with plans to turn out pricing over time.

So what's great, all right, is we have been able to grow both Inventor and Fusion at the same time. Inventor kind of grows at the kind of market rate, which is what you would expect. Obviously, Fusion is growing faster than market rate. And that points to a couple of things. One, it points to share shift, all right? We're getting into some place where there's share shift associated with some of the manufacturing competitors we have, with some of the design competitors we have, Fusion is the better answer for some of those customers. We're also seeing ASPs increase for Fusion. We talked about higher attach rates for extensions, for example. These are all parts of the maturing business for Fusion. The next thing to start cracking with Fusion is going from small installations inside of accounts to larger installations in accounts. But that overperformance is absolutely due to Fusion.

Right, right. And there's been some -- on that point, like you've seen pretty healthy large logo wins, but obviously, maybe where you're sitting is very small as a percent of wallet. Remind investors what are sort of the use cases or products that you're targeting initially to get Fusion in the door? And then where often that easiest path of...

So one, we're firmly targeted in the mid-market with Fusion. We're not trying to take Fusion upmarket. We are firmly targeted on the mid-market. And the place where we have made the biggest initial play is on design-make integration in smaller companies. So what we've done is we've gone in and said, look, you know what, you've got a seat of TAM software X, design software Y? You don't need that, take Fusion. All right. So we go in there and we consolidate those tools onto the Fusion stack. And then what we do is we work with the company to kind of move Fusion upstream into the design. We also are graduating a huge number of rabid Fusion fans from various universities across the world. And they're starting to have an impact inside of the companies that they're going to and saying, look, this Fusion thing is more modern. It's got a better stack, it's easier to use, it goes design to make. And that's part of the core strategy, too, because the investment in education pays off over a long term. But that attack at the design-make integration end is where we're starting, and we're moving up from there.

Okay. Got it. On the Media and Entertainment business, it's not a business that we get too many questions on, obviously, but certainly...

Until it slows down.

Until it slows down and you get the writer strike and everything. But help us understand just structurally kind of what's going on in that business. Obviously, there's concerns sort of that writers and media and entertainment could be disruptive from AI. And we know that the revenue results is kind of a function of what was going on to that 2 quarters ago. So have you started to see some turnaround? How should we think about the structural growth of that?

So first off, we reevaluate every business every year to make sure that we're in the right space. Historically, media and entertainment has been neutral to our business, yet it adds value in the automotive space because the tools are used in the automated -- automotive design studios, and it adds value in the virtual world space, okay? So our goal has always been to get the synergies from that business without a drag on the business. It has been a drag on the business for a little while. And what you're seeing in there is kind of the lingering effects of what happens in a subscription model. There was definitely a slowdown in production activities, all the activities in the media entertainment space during the writers and then the actor strikes and all the things associated with showed up slowly. The recovery will show up slowly as we move forward, okay? So the buildup of the M&E business will slowly build up over time because of the nature of a subscription model. We don't see anything secular in there that is changing. There's no change in the competitive environment. There was just a big hole created. And by the way, even companies like Disney have said, well, we're only going to create like 3 streaming things. One, Star Wars, one -- they've kind of set -- I don't know if that's exactly what they said, but they kind of set goals for their streaming because they're all trying to figure out, okay, what is the mathematics of streaming. But all of that is just kind of a secular play. The impact of AI in our space is actually -- we've leaned in heavily into AI in that space and notice we've also leaned heavily into the production pipeline. So we now own the capture of the digital assets at the early part, on the set, and now we're taking them all the way to VFX. So we've shifted over to full life cycle there like we have in all of our businesses, but we're watching it. But we don't see anything different at all in the competitive environment. It's just recovering from -- people still need content and people will still differentiate on content, and we're just working through the dip.

Right. And should we think of that the growth drivers for that industry mainly being more seats? Or is it kind of data and volume and throughput driving that content?

I think the biggest growth driver is movie industry shifting to life cycle, all right? And increasing not only the depth of use of some of our tools, but also the broad use of our tools. They -- all of these studios need more efficient ways to create the content. Their actors are more expensive, their writers are more expensive, they need productivity in producing the content. Their VFX artists are going to get more expensive, okay? It's like all of the labor, even as it gets -- even as you see AI come in, the labor costs are shifting in that industry, they need full production efficiency. So we're trying to deliver them deeper efficiency, especially for their high-value streaming content that has special effects and all things associated with that.

Okay, very interesting. One of the things that's, I think, been critical to the Autodesk story over the years is, just attracting that next user, right? You talked about some rabid Fusion 360 users. You've done things over the years, whether it's your support for university programs and everything. But how do you sort of ensure that, that top of funnel is very strong? And what are some of the metrics you track to get confidence that you're still kind of gaining that next architect that's ready to use Autodesk?

First off, you have to always be out in front of the technology curve. Autodesk -- look, how many companies -- how many PC companies from the '80s that were back -- alive back then still exist? Okay? I think it's like 1%, all right? And we're one of them, okay? And I think what's important there, one of the reasons we are one of them is we're constantly refreshing this tech stack and moving it forward. Not too fast, not too far ahead of the customer. So you want to stay relevant. Things like Fusion and Forma, they keep you relevant in there. Forma in particular is something that students love and you see it replacing architecture reviews and all sorts of architecture schools are using Forma to do their architecture reviews instead of like using pasted up things on boards, which is really, really exciting to see. So first off, you just got to stay ahead on the technology curve, okay, and make sure that you're keeping everybody clear that you guys are out in front and that this is the solution you want to stick with. The second thing, obviously, is we are big players in education. We make sure that the tools are adopted, and we actually have teams that try to remove friction for teachers in the education environment so that more students are getting exposed to our tools. Remember, they don't pay anything for our tools in education. There's not a student out there that pays for our tools. And we want it to be low friction for the teachers to get those tools out in front of -- including the construction space. So we put a lot of emphasis on that. And then of course, the brand. The brand matters. People need to see Autodesk not just as a design company, but a design and make life cycle company in the cloud. And increasingly, that's how people think about Autodesk. The students have a very positive view of Autodesk. And in some ways, a much more positive view than some of the users who have been using the applications for 20, 25 years.

Right. Now that's great perspective. Speaking of users, one of the things you've talked about in years past is, the nonpaid user or noncompliant user opportunity. I think it's been a while since we've gotten an update on that. But is that still a growth driver? Should we think about the magnitude in terms of contributing to your growth as sort of consistent with what we've seen in years past? Or is it ramping up, slowing down? Just give us an update?

All right. So I've tried very hard over the years. Many times told people it's just -- it's a consistent drumbeat, okay? You don't want us to go too fast or too slow on this. Is some of the growth you see in our business, the result of moving people to compliance? Absolutely is. Absolutely is. And we're doing it at the rate and pace in such a way that we're learning how to do this on a continuous basis without creating negative relationships with our customers. And we get it right sometimes, we don't get it right sometimes. But it's just going to be this continuous drumbeat into the business, and it's going to help shore up the floor of the business. And that's what it's doing. So it hasn't gone away. We haven't cleared the decks on noncompliant users, and we won't for years.

Very steady, it sounds like. Yes. Okay. So in the last minute, I thought I would just open it up to you, and I know there's a lot of things going on at Autodesk. Anything you'd like to share with the room or any key takeaways for the audience here?

Look, I think it's important, and I said this on the earnings call, it's important to focus on why is Autodesk has done all the things it's doing and what's the payback? So 2 years ago, we did the transition to from multiyear upfront to annualized billing. What are we seeing right now? We're seeing this nice, steady free cash flow buildup that all of you can compute. It's one stacking on top of the other, stacking on top of the other for next year. Why did we do the new transaction model? Because we need to drive sales and marketing efficiency moving forward. So what you're going to see is, as we move forward and we move through it, you're going to see margin growth associated with sales and marketing productivity. It's -- we did these things all for a reason. We built a better Autodesk. And that better Autodesk is going to start showing not only consistent top line but some margin expansion there, too. Because remember, when we trend towards the lower end of the revenue growth, we make sure that we're paying attention to the margin growth. And that's always been our orientation and it will continue to be our orientation.

Right, right. Great. Well, with that, I think we're exactly at time. Andrew, thanks for being here. And thank you everyone for attending.

You landed that one, didn't you?

Yes. You ended perfectly.

Nice job, Tyler.