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

Thanks, everyone, for joining us. My name is Sterling Auty. I'm the software analyst here at JPMorgan. Very happy to have with us the management team from Autodesk for our next session. And hear that, see? There was a ping in a day. I didn't follow directions. I was supposed to mute everything. So I'm the first one to violate the rule. We're happy to have both Andrew Anagnost, who is the CEO of Autodesk; Simon Mays-Smith, who's Head of Investor Relations. Before we jump into our fireside chat questions, I'm just going to turn it over to Simon just to run through some safe harbor language. Simon?

Thanks, Sterling. So we may make forward-looking statements during the course of this presentation. Please refer to our SEC filings for information on risks and other factors that may cause our actual results to differ materially from these statements. Sterling, back to you.

All right. Great. So just to warn everybody, Andrew has a little bit of construction going on at his location. So if you hear any of the chopping, the sawing, and the jackhammering in the background, just -- it shows that, Andrew, you're fully supportive of the construction industry, and we really applaud you for that.

I'm also aware of all of its challenges.

So maybe just to get us started. Obviously, a lot of investors will know you. But maybe for those that are a little bit newer to the Autodesk story, you're known as one of the leading design software companies, but help investors really understand what are the end markets that really Autodesk serves?

Yes. So we have 3 primary end markets that historically we served only on the "design or creation" side. But in all of them, we've extended from the design and creation side all the way through to the build manufacturer produce side. The end markets, in order of what I would call digital maturity, are manufacturing, design and engineering manufacturing. So we have a fairly large share in the market of design and engineering software plus manufacturing software and production management software as well that kind of works with it. Generative design, as you know, stitches together those 2 things for us. The main offerings we put into that market are Fusion, which has a highly integrated stack from design through manufacturing, which integrated all of our acquisition technology from the Delcam acquisition; and we have Inventor, which is one of our more historical products that we've had for a long time in that space. Our next end market on the digital maturity curve, okay, so I'm talking on the digital maturity curve, is media and entertainment, which you might say, "Well, how could it be that digitally mature?" Well, all the movies now are digital and special effects that are kind of tightly integrated to the thing and digital execution is critical to a lot of these films that they make. They're not as digitally mature on the production side but they're an interesting market in that special effects, digital effects, which is where we play very heavily in 3D modeling and 3D animation, is becoming a critical component of just about every bit of content you see made in some of these industries, especially on the streaming platforms. Our least digitally mature market, which is one of our biggest markets, is AEC, right? It is early still in its digital journey in becoming a fully digital end-to-end process. We are -- we have historically been the largest provider of drafting software, which was AutoCAD historically, but we have now transitioned to becoming the largest provider of building information modeling software, which is Revit. And that is the big kind of key transformation that's going on inside of AEC right now. It's becoming a model-based process, which mimics what's happening in manufacturing. And we've recently, over the last 3 to 4 years, gone into construction in a big way. So you now take the building information model and all the associated drawings all the way down to construction with Autodesk Construction Cloud. So in all of these end markets, we attacked the creation of content and the execution of content into either a built asset, a manufactured product or a produced film.

Yes. It's interesting that the least mature has kind of been the biggest segment for you in terms of your business and revenue through the years. Before I ask my next question, let me just actually take one second just to remind those that are watching, if you want to submit a question, there is a blue Ask a Question button that you can click, type in your question, it's delivered to me here digitally, and I'll weave it into the fireside chat as we go along. But just starting back up with where you left off in terms of BIM, the building information management, help investors understand what is the core fundamental difference between that and just some of the design files that you've had traditionally in AEC? So what is the functional benefit for the customer? And then what's the financial or business opportunity for you as a company?

Yes. So the key critical difference between BIM and the old kind of drafting-based processes is a 3D model-based process. So you actually model the entire building in the computer as it would appear in the real world before you try to execute it. And as a result, what that allows you to do is identify problems, issues, design flaws. You can experience the design before it's built in a way you cannot do with 2D drawings. Now historically, what did architects used to do? They used to build paper models, right? They all -- they spent a lot of time in their architecture schools building paper models. Construction firms would build them to. They build models to represent these things, physical models that they did -- it would to be time-consuming, error-prone. It captured very little detail, didn't allow you to make trade-offs between spaces and allocations. You couldn't do anything about clash detection between your mechanical and electrical components and your built structural components and things associated with that. A 3D module can do all of that. right? It's a different process. It requires somewhat more skill in terms of using the tools than a drawing-based process, but it has huge downstream returns. One, you know what you're building before it's built. You're able to predict problems before they show up. You're not fixing them up late, late in the process, which costs more. And you're able to kind of present to your clients really significant designs. And then you also have a model that acts as a record that can be used downstream. Manufacturing figured out the value of these models a long time ago. That's why most of manufacturing is driven by 3D models now that go all the way to manufacturing. AEC, we are still worldwide 20% at best penetrated on building information model. Now add to that full digital processes from design all the way through construction and you're even less than that. So the real value is design it, experience it before you build it in 3D.

So how do you take the learnings from the manufacturing world of getting comfortable using those 3D models all the way down to the manufacturing floor and drive that into construction? I've worked in that world, and technology adoption on the actual job site has always been very difficult. I think cutting edge for the longest time were push-to-talk Nextel phones. So how do you -- what have you learned in 1 side to maybe expedite or shorten the time frame to we get to that point where you're getting that BIM model usage actually on the job site?

Yes. So what -- it's important to tackle the most pressing problems first because it's easy to boil the ocean with this kind of stuff and kind of go after the problems that aren't the most significant problems that people deal with. The most significant problem right now people deal with is managing the flow of information to the construction site and making sure the right information gets to the right people at the right time and that when an issue is corrected -- detected, that issue is escalated up the chain as quickly as possible. So it's the field execution problem. This is where a lot of the money is being spent right now and it's where we're focused. But what we've built, though, is a layer of capabilities that lead backwards from the field execution and from top down, from the building information model, to bring all that stuff together in the middle. So over time, we will get the customer to build an end-to-end process. And by the way, this mimics exactly what happened in manufacturing. In manufacturing, the 3D model was originally an engineering tool to make sure that the design you were producing actually fit together, right, was strong enough and tough enough. And then a bunch of drawings were sent off to manufacturing to figure out if they could build it, all right? And then manufacturing would come back with building things. And over time, what happened was -- is the needs of the manufacturing floor or field execution started to merge with the needs of the design and engineering firms where they actually started working off the same model database. And a lot of mature manufacturers now are completely model-based in their processes. We're just going to bring the AEC business along in that journey. It's going to take a while. This isn't like you snap your finger and it happens. This industry does not change quickly. But there's a whole generation that's retiring out and a new more technically sophisticated generation that's coming in. And that generation is more savvy with some of these mobile devices and the tools that they appropriate with it. But there's one other thing, Sterling, I think it's super important to understand. And by the way, this helped drive what happened in manufacturing. The legislative, the permitting, the regulatory environment that people are building things into is changing rapidly. There are many more restrictions on how much waste you can take away from a construction site, how precise the final built asset has to mimic what was actually permitted. The process of getting permitted is getting more complex. There's more reviews, there's more regulatory requirements associated with it. And frankly, in AEC, there's the rise of the design build firm that tends to be more digital than the kind of fragmented supply chain that you see in most construction projects. And guess which people win bids more successfully than others? It's the more digital design build firm. So you've got competitive pressure, regulatory pressure and just general kind of changes in permitting processes that are driving a lot of this and creating more pressure in the ecosystem. It's still going to take time, but it's not just technology coming in and say, "Hey, tech's great." It's -- you can't survive without tech now. And last year taught a lot of people in the AEC business that.

So am I reading it right that basically what you're saying is you're seeing more vertical integration between the architectural firms and the build firms? And if that's the case, does that change any of your market opportunity in terms of the number of users or maybe the wallet that's created?

Yes. For instance, you still need the same number of people to get the work done. A matter of fact, more and more, the AEC industry has a massive capacity problem. They have more opportunity to build things than they can possibly get through, the capability they have today, which, by the way, is something important about digitization is increasing that throughput and that productivity. So design build firms don't replace the ecosystem. What they are, are just more tightly integrated representations. And what you see some larger architecture firms doing is they're forming networks with a certain set of GCs, and they're creating tighter relationships between their firm and the general contracting firms and going to prospective clients as an entity rather than the architect gets -- creates the drawings and delivers them to the owner and the owner takes the project out to bid to a GC. Sometimes they're trying to coordinate their work more because that creates a tighter bid process and a more predictable outcome for people. So it's just a shifting of the ecosystem. It's not shrinking. And if anything, we have more that needs to be done than the current system is capable of doing, which is great for our tools, by the way, because even if you see -- even if you were to put fewer people on a particular project, in a digital world, the company is executing more projects with fewer people per project.

So how important will it be to drive the shift of your tools into the cloud to kind of get to that ultimate vision? Can you do it with the tools that they have now? Or this is my shy way of bringing us into the discussion of the cloud transition that seems to be underway across your entire business.

Look, if there's anything the cloud does incredibly well, there's 2 important things. It's data flow and collaboration, especially distributed collaboration. Nothing does it better than the cloud. And when you bring the cloud to bear both in manufacturing and AEC on this notion of seamless data flow, data flow between individuals but also data flow between disciplines, architecture, engineering, general contractors, where they're all seeing the same kind of information updated and coordinated in the same way, this kind of flow -- this enables a productivity revolution in these industries. AEC, in particular, you all -- you know, Sterling, you know the productivity figures in AEC. They're terrible, okay? They're absolutely terrible. No one can afford that kind of productivity anymore. We can't afford the waste. We can't build enough. I mean, we've got an infrastructure bill coming right now. The worst thing that could happen is that infrastructure bill does not bring with it some provisions to digitize processes because instead of getting 3 projects per x dollars, you're going to get 1. And nobody can afford that because there's too many projects to build. So yes, the cloud is critical to this. It's one of the reasons why we started this journey so many years ago and kind of started integrating the cloud in our processes. Not all our customers thought we were acting in their best interest during that time. I think last year may have changed some people's minds around this. It certainly has changed a lot of people's minds that I talk to every day.

Yes. That makes sense. And given that you brought up the infrastructure bill, let's just -- since that -- top of mind, can you talk to investors -- I get this question frequently. What does that opportunity mean potentially? Obviously, nobody understands what the final format of it is. But when you think about a big increase in infrastructure spend, how does that flow through to benefit Autodesk?

Yes. So first off, we don't put this into any of the long-range model we talked about publicly, okay? Because it's a fool's errand to depend on government spending to -- for your business. However, it's clear that in its current form, if the bill passes, there's going to be an increase of activity in areas where we do business, areas where our competitors do business as well. However, I think one of the provisions that's currently in the build is digital infrastructure for construction, similar to what they did with the Affordable Care Act, where they put digital infrastructure for medical records in part of the programs. The combination of increased job activity where we're strong, road, rail and now, of course, water -- water is in the crosshairs of everybody's work right now. Just look at -- you don't live in California. I do. They need water infrastructure. But even in the south of the U.S., they need flood control infrastructure. So we've got lots of water infrastructure. So we're absolutely going to start to benefit from just increasing project capacity. How much and when, it's going to take time. These things don't surge in suddenly. They kind of build up over time. But there's also this underlying message that's going to be part of this legislation to, hey, digitize your processes. If we're going to build, we're going to build back better, and that means you're going to have to be digitized. So that's going to have another long-term implication on our business that people are going to be looking more deeply at the tech stack, especially cloud-based tools in some of their decisions. And they're going to be asking themselves, "Am I investing in a modern tech stack. Or am I investing in an older tech stack that may be good at the job that I'm doing today but won't be good at the job I'm doing 5 years from now," which I think is a tailwind for us ultimately?

Are there key examples of higher-profile infrastructure projects that have been done over the last 1, 2, 3 years that have started that digital process that you think will be used as kind of the example of driving this forward in that type of bill?

Yes. There are -- there's plenty of municipalities that are going to stand up and say, "Look, we use digital tools." Just here in the Bay Area, we partner with some of the transit authorities here, and they're doing some interesting things, all right? And they're doing lots -- they're looking at projects in different ways. These -- I would suspect there will be a competition around some of these municipalities to see how digital I am because, guess what, where is the money going to flow, right? Some of it's going to be political. Some of it is going to be who's shovel-ready and whose execution is going to be fast enough. So everyone, I think, is going to be waving the flag around their digital chops with some of these projects. And especially in the capital planning phases, we've partnered and built out the right kind of systems that allow people to get off the ground and start moving on this stuff digitally. So I think it's going to be really exciting. I think it's going to be catalytic for the industry. It just will take time. So it doesn't -- and that's the one thing I like people to know is like it's not like you flip a switch and all of a sudden, they can change their productivity model. It's going to take time, but this will catalyze it.

Yes. But I want to make sure that sometimes investors think when you talk about something taking time, that equals slow growth for a company like yours. I don't think that's the case. So maybe kind of compare and contrast the comment of taking time and the time you're seeing it grow.

Yes. As a matter of fact, what I mean is the digital -- so the digital transformation of this industry can take a decade, okay? However, they're going to be buying digital tools throughout that entire transition, all right? And what that means is they're going to front-end load some of their purchases on certain digital processes but they're going to continue to expand those digital processes. So what it means is the market has actually been opened up for new types of technology and new types of capability that we put into it. Frankly, it's been stagnant for a while except for building information modeling. But now with some of these other changes coming in and some of these other reactions to the distributed work of COVID, the requirements for more digital processes and more building information modeling for us, it's just -- it's a growth engine, right? And it's a long-term growth engine. And that's why I want people to understand is that it's not growth that something shows up in a spurt. It's growth that kind of bleeds out over multiple years, and it's good growth.

Excellent. And then I want to go back to the cloud discussion but transition over to the manufacturing. You mentioned Fusion in particular. And I think a lot of investors understand the strength that you've had in the AEC market for the longest time. But in the manufacturing side, there's a number of players that are there. Can you talk to us about kind of the Fusion kind of strategy and where you are -- where that takes you in the cloud journey and what that does for your competitive advantages?

Yes. So there's a couple of aspects of the Fusion strategy that are really important. One is the Fusion stack is cloud-native. We built it with the cloud in mind. We built it with data flow for the cloud in mind, all right, which is a critical part of where the world is going with regards to distributed supply chains, distributed project infrastructures, distributed capabilities. But it's end-to-end, a single integrated stack from design through all the way to manufacturing. So that's one thing that's completely unique to Fusion. The other thing that's important to know about Fusion is this whole notion of a highly compute-intensive cloud stack that starts doing some of the work for the end user, all right? You've heard some of the things about generative design. Those are kind of early explorations of kind of ensuring that the design can instantly go to a particular type of manufacturing capability or actually proposing designs for you. But amazing things happen when your design data is in the cloud and you're able to throw a lot of compute power at it. You can actually reduce the amount of labor required to do some pretty complex things. So compute power in the cloud is pretty powerful. The next thing is data flow in the cloud. This ability to have different teams in different places work on the same kind of data, different disciplines, mechanical engineering to electrical engineering to manufacturing engineering, all working on the same data and probably touching different parts of it at the same time but never changing a file. They're just working on this massive database that allows them to accomplish what they're doing. Most manufacturers have realized they have to move to this kind of world moving forward to manage their workflow and manage their systems. So what we've done is we believe that we've built a native new environment that targets the disruption of the cloud for manufacturing. There's really only one other competitor that's done that well so far in the space. And interestingly enough, they were -- they're a smaller player in the space than the biggest competitor in the space. But as you know, we spend a lot of time displacing our competition with Fusion. In fact, we haven't displaced, as far as I know, a single seat of Inventor with it. We displaced other people's software with it. And we do that by moving them to the cloud. This is the classic kind of tech transition manufacturing loves. Manufacturing went from 2D to 3D well before AEC. It went from 3D to parametric well before AEC. And now it's going to -- from 3D to the cloud well before AEC is doing it. And unlike last time, when it was 2D to 3D and all this was a little behind, Autodesk has been working on this for a while. That's why we have the installed base we have not because we were better or smarter than anybody else. We started earlier, and that matters.

Well, I think what's interesting is along the way, it feels like you've changed the discussion. So I think originally when the idea of concept of, "Okay, are we going to do design in the cloud, it was the simple can you take Inventor, put it in the cloud, it's not running on my local client. And geez, what about the lag that you have and the back and forth over the Internet?" But that's not it at all.

There's very little value in moving CAD to the browser. There's no incremental value to that. It's about connecting design to manufacturing, yes.

So what does the -- when you fast forward and think out into the future, does it come to the point where the algorithms, machine learning, the generative design are doing 95% of it and just human engineers are just making small tweaks or changes to parameters? So they're not doing a lot of the dragging and making this 5 centimeters instead of 4.5, et cetera, the associative design?

The idea is to remove the drudgery from the design and keep our engineers and designers framing out the design and creating the concepts and the constraints and the things that make the design important and creative. And that the computer does a lot of good grunt work. There's a lot of grunt work associated with building a 3D model that people just shouldn't have to do. And it's our belief that over time, we're going to move a lot of that to algorithms that present geometry options based on the way the designer engineer is framing the problem, and that will increase their productivity significantly, okay? And that is part of the power of the cloud. If you're going to ask me now, does that reduce the number of people that are going to be using these tools, and the answer to that question is no, it actually increases the number of people that are participating in the design ecosystem, okay? And this is something we've seen every time that we've created tools that are more accessible, more people participate. And when you're moving to a world of re-onshoring of manufacturing capability, distributed manufacturing capability, we talked about [ zoometry ] before this call. When you think about facilities where you've got manufacturing as a service, and people are streaming bits into somebody's black box factory and getting stuff out, okay? This is the future, all right? And when you can do that, when manufacturing capacity does not become the hurdle to designing and making something, guess what, more people design and make something.

Yes. And we used to talk about cheaper viewer seats versus CAD. But now to your point, it gets more of the enterprise involved in the actual value add of -- in improving the design. Now I think what was interesting is nowhere in your commentary, did I ever hear you mention PLM, PDM, but I heard a lot about data in the cloud, data collaboration, data sharing, which was the original view. Autodesk really was a laggard in kind of the PLM. Has the investments that you made in the cloud kind of changed the paradigm enough that maybe you can leapfrog a lot of that legacy?

Yes. You see a lot of those legacy systems were built because data was so hard to manage in the old paradigm, files, data and all of the stuff. When all your data is in the cloud and your information is disaggregated from a file and to a set of cloud data graphs and APIs and components, all of a sudden, managing the life cycle of that information becomes a lot more seamless, managing the clashes. So you don't buy a different PDM system anymore. What you do is you turn on the management capabilities inside the cloud stack. You don't buy a different life cycle system. What you do is you turn on the life cycle flow capabilities within the stack. So all the problems that PDM and PLM were designed to fix with new types of monolithic systems are much easier to address in the cloud. And it's much easier to turn them on, and it's a lot easier to implement them because you don't need to implement all these complex data graphs and stacks because, guess what, the cloud takes care of that for you, all right? The intelligence in the cloud does a lot of the work. So yes, it's a revolution. People will still be doing product life cycle management. They just won't buy a product life management system. They'll buy a stack and they'll have product life cycle management capabilities inside it. One of our competitors realizes this and is trying to protect their franchise, and my hats off to them. They're a little late, but they get it.

Yes, absolutely. All right. Again, just a reminder to investors, you can click on the blue Ask a Question. Go ahead and submit your questions there, and I'll weave them in to the fireside chat. So we talked a lot about some of the long-term trends that benefit you. And it sounds like there's a number of them that give you secular tailwinds. But I also want to spend at least a few minutes talking about more in the near term what's happened here. Where are we in the pandemic? And how should investors think about in the coming quarters what are going to be the factors that are going to impact your business?

Yes. So Look, we have seen a steady unwinding of uncertainty. Delta has brought a little bit of uncertainty back but has brought uncertainty back into an ecosystem that had already adapted to a pretty significant shock 1.5 years ago, okay? So we've got an ecosystem that understands how to adapt as uncertainty creeps back in. So I feel pretty good that the increases in usage and project demand that we're seeing is going to continue moving forward. The biggest challenge is in our supply chain right now, in our customer supply chain right now. One, AEC is having trouble hiring people. Everybody's got help white signs out in construction and everywhere. So executing on all the job -- there's a huge backlog of jobs starting to fill up. A good thing or a bad thing, right? The huge backlog means that they're going to be plunking through that backlog for a while, but it also means that people are writing checks. So they've got to be -- we got to be able to get people out there back to work, and we're seeing it, but they need more people out there working in more capacity and more productivity, by the way, which plays well for digital tools. because if they can't get more people, what do they need? They need more productivity, right? We sell productivity. In manufacturing, I think you're going to see the news, Sterling, there was just a report on -- I think it was on NPR this morning about how shipments are sitting on ships for weeks, and they can't even get them off the boats. And then there's just problems getting things on the boats in the first place. So supply chains are not able to feed the demand right now for product coming into manufacturing. It's a different problem, right? Manufacturing has its own talent acquiring problems but it also has this now supply chain throughput problem. Now if those things go on forever, that's not great because, again, people don't write the checks. However, I love seeing that backlog. And I love that, that backlog is teaching our customers, hey, the more digital I am, the more productive I can be with whatever my supply chain disruptions might be or whatever my employment disruptions might be. So it plays right into a digital future and new digital tools. But those are the things that are affecting our end markets. Right now, we're not seeing it have any effect on usage of our products out there. And I expect we've got enough runway to get through the Delta distortion right now.

All right. That's good to hear because I was going to ask you is this a people problem or a technology problem. It sounds like it might be a people problem in the nearest time frame but really -- that's the Band-Aid. The solution is really the technology.

Look, I never expected nor wanted the last 18 months we've had to kind of help people understand why digital productivity is so important, why distributed work was the future. It was always the future. The future just came on to us several years before we expected it to, okay? So yes, productivity, digital tools are absolutely a major part of this pond. They're not the only part, but they're a big part of it.

So weaving that in, over the last couple of quarters, I think there's been just a talk of acceleration. And it sounds like from what you just said in a number of those answers that things are falling into place to allow that to happen. But how about -- you talk about finding the people for AEC, finding people for manufacturing. How about your own business? Are you finding the right resources to be able to man, to staff, I should say, and drive that execution on that acceleration?

This is a -- this is another thing that this new year is teaching all of us is how we manage our employee base and our talent capacity. And the truth is there's going to be a battle for talent in all industries. And that means that not only do you need to be able to attract the best talent but you have to be able to deliver a capacity to rescale your existing talent. Gone are the days where you can just kind of assume that you can hire a bunch of capacity or capability you didn't have before. It's going to get harder to do that fast enough. So what you're going to see a lot of companies investing in over time is reskilling, retooling and upgrading their internal capacity while they try to bring in additional capacity from the outside. Obviously, the best way we bring in external capacity is acquisitions. Acquisitions are where we bring in a lot of external capacity but acquisitions are expensive. They're complicated. Hiring people is getting increasingly expensive and complicated. And I think we're going to be hiring, acquiring and reskilling. And I don't think we're unique in any way relative to anyone out there right now.

Yes. No, I would agree with that. One of the questions that has come in from an investor is just saying, with the discussion of return to office, Delta variant, et cetera, does it actually change or impact commercial construction as an industry? And can that have upstream negative impacts to your business?

Yes. So this question comes up a lot, okay? First off, commercial construction, the money being spent on commercial construction has been moving away from new buildings to reconfiguring existing spaces. So there's tons of activity in our business right now reconfiguring existing spaces. Matter of fact, we deliver tools now inside of Revit that are specifically designed at helping people reconfigure spaces in an automated kind of way because the activity and demand for that has gone up significantly. But more importantly, what you see is the money is going elsewhere, all right? It's not that people -- so yes, are you seeing lots of plans to build a 50-story urban office complex? No. Are you seeing activity going on in other sectors that offsets that product activity, like, for instance, suburban expansion, multifamily units, small suburban office parks, satellite locations for hubs, for larger companies, data centers, okay, hospitals, medical infrastructure. What you're seeing is the money is just going other places. And this is one of the great things about having a footprint that spans across so many industries. The money goes other places. And look at what's happening in infrastructure now, okay? So we're going to see more infrastructure on the horizon. There's too much work that has to be done. It almost always shifts to something else. In the commercial space, it shifted out of urban to suburban and it shifted to reconfiguring urban spaces for lower density and different types of work environments.

When you look across the 3 big pillars that you talked about, AEC, manufacturing and even media and entertainment, are there still significant functionality holes that you feel like you want to fill? And what's kind of the buy versus build approach to those?

Yes. Look, there's a lot of functionality gaps we have that we're going to address organically. There's just -- there's some basics with regards to some of our collaboration products in parts of our portfolio like Civil 3D and other types of products. that we just need to fix organically but we need to expand. We need to start building on a new environment for AEC that's built out of new types of capabilities. We acquired a team in Europe, Spacemaker team, that's really helping us build out new types of capabilities and new types of platforms. In construction, have we finished every little bit we want to finish there? No. There's -- we've acquired some of the biggest technology that we think is important but we've got to build a few things and add a few things. This doesn't go away. Same in manufacturing. There's still a few things that we'd like to do. So it's going to be, for a while, Sterling, a combination of organic investment, where we know what we have to do; and inorganic activity that fills a hole that may not be part of our core expertise or our new core expertise. So look for us to do both, which is one of the reasons why we talked about our long-term margin targets in terms of what we're doing. We're investing in top line growth, and we believe that the margin targets we put out there can absorb all manner of inorganic and organic activity to drive that top line growth.

Great. One -- an additional question came in from investors around the noncompliant users that you've been programmatically focused on for the last several years. What's left in that opportunity for you to kind of squeeze out of the system? And how does that help your growth moving forward?

Yes. So one of the things we characterized in Investor Day, Sterling, is that, that capacity, that noncompliant usage, we steadily address it in chunks. We don't -- we've never put a pedal to the metal on that. We don't drive it super hard. The goal is to bring these people into the paid ecosystem incrementally over time. Every year, we bring more in than we did in the previous year because we're getting better at it. We're getting better at identifying these people. We're getting better at engaging with them. But I think I've been consistent over the years, and I'll continue to be consistent about this. It is not meant to be a sudden accelerant to our business. It's going to be a multiyear journey of bringing these noncompliant users into the paid ecosystem in such a way that's good for us and good for them so that they actually want to buy more from us and engage in some of our bigger digital footprints and not just the footprint they might have been using in a noncompliant way. So never -- no hockey stick but a nice steady drumbeat of incremental growth from that.

Yes. That makes sense. Another one is specifically on the construction opportunity with Procore going public. There's obviously a lot of questions back and forth. And I think investors do understand that this is a massive opportunity. It's not one that's going to be filled by just one company. There's going to be a number of companies that benefit from it. But the question specifically is when you look at your offerings in the Construction Cloud, do you feel like it gives you the ample footing to compete today? Or what is the next -- what are the things that we should be looking from Autodesk in terms of additional either feature functionality or strategic moves that we'd make to be more effective in that space?

Yes. So first off, I'm super excited about what we've done with Autodesk Build and the way we've integrated PlanGrid, best-in-class mobile device technology with what we consider best-in-class cloud-based platforms with what we have done with BIM 360 Field next gen. So I think Autodesk Build in that integrated platform is best-in-class, technologically, capability-wise. It's got -- it's super competitive on the project management space. It's an excellent product, and it's a great example of what Autodesk can do when we acquire best-in-class technology and bring it together. So we believe, indisputably, we are the tech leader in this space. We think our tech is better, all right? But as you know, it's not all about tech. It's about engaging with the customers. It's about satisfying all the types of problems. We feel the expansion of build is going to be a critical part of what we're doing because field execution and project management is the right now problem for construction. But there'll be other areas where you'll see us go deeper. You'll see us go deeper into estimation. You'll see us go deeper into other particular areas that bring preconstruction planning and construction execution closer together. So look for us to close that gap between preconstruction planning and construction execution as we move forward over time. Some of that will be organic investment. Some of that will be inorganic investment.

I think one of the follow-up questions that I often get is how much of the industry is already moving into that direction that you mentioned, that close integration between design and build where BIM becomes more important, and it would seem like you would have a natural competitive advantage versus traditional kind of separation where maybe it's not as much in focus?

Yes. So up -- from the high end of the mid-market to the higher end of the market, BIM is king, okay? And more and more, the integration with the building information model and the mandated requirements from owners and other people around end-to-end flow and a single source of truth, BIM wins. Long term with BIMs everywhere, I believe, just like what you saw in manufacturing, Sterling, where you've got ecosystems built up around people's models, you have the Dassault ecosystem, the PTC ecosystem, the Autodesk ecosystem, all built around their 3D models, that's exactly what's going to happen in AEC as well. There's going to be 3D model-driven processes tightly connected to construction. And all these current notions of stand-alone "data players," I think you might remember -- you probably remember back in the day when PLM was rising. You had companies like MatrixOne. You had like 3 big, huge companies that all they did was that, they're all gone. Why is that? Because everything consolidated around a certain group's model-based processes. That's exactly what's going to happen in AEC. Who is the BIM player in AEC? It's Autodesk, okay? So as long as we keep our technological edge, keep delivering on the things that our customers need, you're going to see a massive consolidation around BIM as BIM moves deeper and deeper into the midline market. You're not going to see it in home construction like all the noise you're hearing out here. There's no BIM model at this remodeling. There's no BIM. There's no BIM here, okay, despite my best efforts to get it in there. There is some online construction collaboration that I've tried to introduce it there but no BIM.

Yes, that's going to take a while. I've worked in both the residential and the commercial side that -- the residential side has got a long way to go. Maybe with a couple of minutes that we have left, one of the other questions that came in is just your evolving use of the channel as you shift to the cloud. In other words, what do you see as the ultimate format? And how much of the business just ends up being naturally done directly, not for anything bad that the channel has done, just the natural as you're on the cloud, it's easier to work direct as we've seen Adobe kind of go through its evolution?

Well, that's one of the reasons why we told everybody out there that, look, over the next few years, look for our business to move to be 50-50 direct and indirect. Look, our partners are an incredibly important part of our business and our -- the majority of our -- 80% of our partners are -- I mean 20% of our partners are responsible for 80% of the business. Those partners are going to continue to enjoy a robust growing business for years to come, right? However, their role in that ecosystem is going to evolve and change. And one of the things I tell our partner ecosystem is: "Look, if we're building a platform for design and make, you have got to get really good at integrating that last mile for our customers, for your customers so that you can stitch together the systems to their particular needs." So look for our partners to kind of go back to the future where they were 30 years ago, where they were like customizing AutoCAD to kind of integrate new processes and innovating processes for their customers. They're going to start innovating processes for their customers on the cloud-based platforms as well. Yes, certainly, more and more of our business will become direct, it's inevitable in a cloud-based world, but our partners are going to be with us for a long time, too. Their role will likely evolve over time, though.

Excellent. I think with that, we're up against our time. Andrew, as always, real pleasure. I enjoy our conversations. Thank you for joining us. We really appreciate it.

Thank you, Sterling. It was great.

Thank you.