ADSK - Autodesk Inc. (ADSK) UBS Global Technology Conference (Transcript)
2023-11-28 19:46:02 ET
Autodesk, Inc. (ADSK)
UBS Global Technology Conference
November 28, 2023 3:35 PM ET
Company Participants
Andrew Anagnost - CEO
Conference Call Participants
Rich Hilliker - UBS
Presentation
Rich Hilliker
All right. Well, hi, everyone. My name is Rich Hilliker, and I'm a software analyst here at UBS. We're really excited that you all could join us, and we're very grateful that Autodesk is here again year after year, and we've got Andrew Anagnost, CEO. Thank you for being here. Excited to be up here together and having a great discussion.
Andrew Anagnost
Yes. Thank you, Rich. Pleasure.
Question-and-Answer Session
Q - Rich Hilliker
Awesome. Well, why don't we dig in kind of hot topic, you reported earnings just last week, so I thought we could start there. A key focus for investors has been on free cash flow. And I think you've been pretty clear in your messaging that the path of free cash flow would be nonlinear kind of as we work through next year in part due to the shift to the one-year billings. But I was wondering if there's a little bit of noise here, there's been a little bit of discussion. What would you say were the most important takeaways for investors from last week's -- from last week's earnings, but also in regards to free cash flow?
Andrew Anagnost
Yes. So first off, it was a strong quarter. We did beat and raise. So it's important to do that. We also kind of set some guardrails on next year. And I think it's important to kind of clarify some of those things and get on the same page here. We didn't guide revenue. We put a floor out there, and we kind of talked about some aspects that are not macro related, so we weren't guiding to a macro, we even guide at all. We had a 9% floor out there, primarily driven by a 1% impact from EBAs not renewing from the pandemic like they did this year. It was a great EBA year, next year. It's going to be good EBA next year, just not quite as strong. So you average the 2 years, you can kind of get what usually happens.
So that was one important thing people need to take away from there. It was just a floor just to help people understand that the EBA business and some of the deceleration this year has a pull-through. The other thing we talked about was the $200 million in free cash flow generated this year that was because of multiyear contracts built up front, right? And we wanted to make sure that people weren't counting that as part of the buildup in the next year.
And the great thing is, right now, we're now in a position where the current Street consensus is solidly in the middle of where we expect to be. And that personally for next year that that number that's out there. That's a number that I can personally stand behind, just like I stood behind flat margins for this year. So I think that was a great outcome. And I think people should take a lot of data in that, that we're sitting there next year solidly in the middle of where we expect to be.
And the thing I want people to remember too is because our multiyear contracts have been coming up at about the same rate that they've come up historically, there is an inevitable mechanical buildup of free cash flow after next year. We hit the bottom in free cash flow this year as part of the free cash flow transition. Next year, we're already talking about where we're going to end up. And then because of all these multiyear contracts signed this year and signed next year, there's a nice mechanical buildup of free cash flow, an inevitable mechanical buildup of free cash flow as we head into the outlying years. Those are important takeaways from the conversation, even though there was some kind of worry and concern that we were signaling something around macro, we were not.
Rich Hilliker
Right, right. Absolutely. And I think your point on the fact that you're still selling multiyear agreements, it's just the billings timing, the term, right? You look at it upfront and multiyear, it's annual billings, but that kind of leaves that mechanical kind of build-up of cash flow that you mentioned.
Andrew Anagnost
Yes. Exactly.
Rich Hilliker
Got it. Another big item that I took away from the call. I think you signed your largest EBA deal ever in the quarter. So I was wondering if you could maybe just give us any more color on that deal, whether it's an industry, the size, the length? Any other color that you could give us.
Andrew Anagnost
I can tell you that it's in the AEC industry. And look, what's important about these big EBAs that you're seeing is there are a clear sign of our customers moving to life cycle solutions, design, make solutions in the cloud and exploiting the power of our Autodesk platform services to integrate some of their processes internally. All these large EBAs have an expansion element in them, where they're moving away from buying design solutions to us to buy these life cycle solutions, design and make, revit, construction cloud, preconstruction planning, all of these things. So these deals are all inherently those kind of deals.
Rich Hilliker
Got it. Can you give us a sense of this customer's journey with Autodesk? Are they -- have they been an existing customer? It sounds like if it's the largest delever, they're kind of -- they have to be at least in some capacity standardizing, really kind of like you said, capturing the entire life cycle here.
Andrew Anagnost
Existing customer, obviously, existing customers started with us is using our design tools. but then started building out using more and more of our portfolio and using more and more of our capability and then also trying to invest in our platform services because those are part of these deals as well in order to integrate processes more tightly within their internal capabilities as well. So it's a journey from design all the way to construction solutions, all the way to then integrating data flows and data solutions on top of our cloud APIs. It's the whole picture of where we're taking the company.
Rich Hilliker
Got it. So it sounds like data is a big -- also a big picture.
Andrew Anagnost
And it has a big impact on what happens in the mid-market because customers of this size have a huge impact on what types of solutions get adopted downstream and what kind of solutions end up spec for particular projects.
Rich Hilliker
Okay. Got it. Maybe last one on that. To the extent you can comment any sense of size for another room or you can avoid that.
Andrew Anagnost
No, I going to avoid that.
Rich Hilliker
Alright. Understood. Why don't we shift over to high-level macro here. I know in your opening comments, you said your revenue guidance wasn't necessarily a macro call here. But what we have heard from other software companies --
Andrew Anagnost
Wasn't revenue guidance. It was revenue floor.
Rich Hilliker
Right, exactly. Revenue floor, right? But I guess the point of the question here is we've heard signs of continued caution around spending, around macro, just across all of software. So I'd be curious, you're coming off of AU just a couple of weeks ago, what do the conversations sound like today relative to earlier this year, maybe earlier quarters in the year?
Andrew Anagnost
Yes. So first off, we built a really resilient business. It's not just a subscription model. The business is diversified across size of company, right? One thing I want to remember about this year is that this year was a big EBA year, a solid year down market and a pressured year right in the middle. We're also diversified across industries and industries have puts and takes, and we're diversified across geographies. So let's just remember that context because puts and takes play out every year.
Now in terms of what I'm hearing it, AU, most customers are fairly bullish about next year. This year, everyone acknowledges the headwinds. They acknowledge the belt tightening that they were engaged in frequently belt tightening years are followed by ultimately investment cycles and things associated with that. They're all hyper interested in digitization, productivity and the long-term challenges their industry have are continuing to pivot them more to technology. So they're not going to be able to solve their labor shortages. They're not going to be able to solve material supply chain issues anytime soon. Their book of business, there's still a backlog in AEC manufacturing. The demand has not slowed down for some of the products they're building. In fact, they're still fulfilling some of the demand.
So most of them see digitization as a way through this. So for us, if I just -- you're probably going to ask me about some of the vertical segments. I'll let you ask me before I dive into anything else.
Rich Hilliker
That's actually the next question. I think you called out that water and construction were pretty strong this quarter. So I was going to -- exactly right. I was going to ask if you could dig into something.
Andrew Anagnost
So construction is a great example. This is another part of the power of Autodesk in that we're spread across all the types of projects that are out there. So for instance, okay, it's no secret that segments like retail, commercial office, commercial, residential, they're under pressure. Now by the way, the pressure doesn't mean that nothing is happening in those segments. They're just under pressure. Even if owner X goes bankrupt, somebody else still owns the building, all right? But counterbalancing all of that, you've got segments like health care, hospitals, elder care, right? You've got data centers. You've got a manufacturing boom in the U.S. that's contributing to factory construction. You've got infrastructure, right? You've got all of these other countervailing opportunities out there that offset some of the demand there.
So -- we -- like I said, we saw a record EBA quarter with regards to construction participation in EBAs. That's a sign of what's happening for those customers that are exposed to those growth segments. Even when you look at some of the numbers, and I know they got highlighted by some people out there that the business is slowing, yes, backlog went down. It's still like 8 months, okay? It's 8 months of backlog, okay? There's still a lot of backlog out there. So in AEC land, we're seeing a lot manufacturing, they're just trying to catch up.
Rich Hilliker
Yes, that makes total sense I guess given this kind of complex demand of Mosaic you sort of painted different areas -- pockets of demand, different areas, maybe some tension. I guess some -- how have strategic road maps for customers evolved? Is there like a common -- couple of common threads you could call out? And then kind of a related question is, any sense of customer budgets for 2024 relative to 2023?
Andrew Anagnost
So look, every customer I talk to is highly focused on their data integration and their ability to turn data from a liability to an asset for their particular business and what they're trying to do. So they're super interested in what we're doing with the platform services. They have very specific requests around APIs, so you can see where their minds at. Some of our biggest customers with some of the biggest engagements with us are hyper focused on connecting data across projects and cycles, if they get that right.
So that -- sorry I've been coughing for 3 weeks. I don't anybody freak out. It's -- this is just what 3 weeks on the road does to you. It leaves you with a lingering cough. So they're really excited about the possibility of integrating their data and their processes. Most of the customers I talk to see a continuing trend of investing technology as they head into next year. Nobody is talking about pulling back. They absolutely all acknowledge some belt tightening this year. But looking forward, and I'm talking to some pretty big and influential customers, they're looking to continue to invest in technology. And they have very specific requests from us.
Rich Hilliker
I think that's consistent with some of the things we heard at AU as well. A lot of customer budgets were seemingly increasing next year in pockets of demand like you described. So -- that's really helpful color. We talked a little bit earlier about insulating the business and you've done a lot of work. There's been a number of moving pieces over the years, different chapters of transitions. I guess maybe, could you talk maybe about some of the -- what you view now as the most critical steps that you took to insulate the business relative to any other maybe macro hiccups or challenges that we face here. Maybe you could call those out. And how does that factor into your 2025, 2026 planning as you think about this more resilient business, more insulated business.
Andrew Anagnost
So first off, the resilience starts with some consequential decisions we made as a company, right? The most consequential one is we were going to move into life cycle solutions in the cloud, all right? So -- not only does that immediately diversify you away from design, it puts you into a high-growth segment connected to data flow in the cloud and the rise of custom API solutions and platform services that people can use. So you immediately set yourself up into a different vector.
And as a result of that consequential decision, you start to do a bunch of things that immediately change the category of your business. If you believe that in the future, most of your business is going to come from the cloud, you don't sell perpetual licenses. You don't do buy-sell transactions, and you certainly don't bill multiyear contracts upfront, right? You're a subscription company, you provide multiyear access but billed annually. And you transact directly with the customers. So all of these things build a more resilient, more -- a stronger business that's diversified across design and make in the various industries and geographies that we serve. That is a really important part of how we build a company that can flex as opportunity ships to very different areas, and we can do buy and -- upsell and cross-sell in new and interesting ways.
So yes, that affects how we invest moving forward and which adjacencies we choose to invest in. Remember, we made a consequential decision around investing in water. Water is everywhere now. It's an issue for everything. How water moves, floodwater management, water treatment, water purity. This -- you hear it in the news all the time. Water is not going away as a challenge for people. And that was an investment that further diversified us into yet another segment that was going to see future investment. And we'll continue to do that as time goes on.
Rich Hilliker
Absolutely. One of the consequential decisions right now that we're all kind of talking about is the transactional model, and we've talked about kind of rolling that out in Australia, and there's the potential global rollout as you see how it kind of progresses. I guess maybe could you talk about some of what you're experiencing so far, customer feedback, partner feedback, any pain points, things that you're really focused on? And maybe what that rollout could potentially look like timing-wise?
Andrew Anagnost
Yes. So I mean, first off, let's make sure we anchor on the why here, okay? So this is like the last big transition in managing the company. If you were to start a cloud company today, you would never sell your product through 2-tier buy sell distribution. I mean everybody would think you were like something was wrong, all right? So this notion of bringing the transaction to us and the partners helping with the design make integration services and extension is a fundamental decision. And it not only allows us to own the customer record, but it allows us to understand the customer from a usage perspective and an account perspective, which means that we can manage upsell and cross-sell internally, much more fluidly, we can do closed-loop marketing with these customers, and we can help our partners manage upsell and cross-sell more fluidly, which partners struggle with upsell and cross-sell in general. They're very good at reach, not so good at demand generation.
So we now have control over those things. So let's just put that in context as we move forward into this new model, we're going to get control of those things, which I think is important. Customers -- for customers, this is going to be virtually invisible. It just means they can transact directly with us if they want to or through their partner or they continue to transact to the partner of choice and get the services that they expect from that partner. It will also give them pricing stability. So we're not hearing anything from customers on these things.
It's a big lift for some of our partners. And the way we managed all of this is we started over 18 months ago. And as you recall, we introduced Flex, the Flex model, which is the volume consumption model we have. We introduced it first in Australia as an agent model. It's basically the new transaction model. We tested a few things out here. We learned a few things like it's always important to be on your customers approved vendor list. We learned a few things that partners wanted as we rolled flex out globally. We learned that they want to be able to true up contracts to the same renewal date and do some of these things. These were like priority requirements for them.
We also learned a few things about the process and the user interface that made it easy. So we did all of that with Flex first, and we actually have been able to test it at scale because Flex is scaling nicely. Now we're testing in Australia, so far it's already out there, no major problems. Now we're looking for problems. We're trying to find them. We still got the rest of Q4. There's a lot of volume and renewals that will come into Australia. If there's going to be a problem with scale, we're going to see it. And if there's something consequential that really is like, oh, it's a showstopper, then we'll adjust our schedule, but so far, we don't see it. We're looking for it, we're hunting for it, but we've deliberately tested this out. So it's a big change for our partners. They have to get used to this model. Customers less so, all right, because they can go through their partner of record, and we've prepared for this.
And by the way, more importantly, sorry, have to say this, it's neutral to free cash flow, right? It could be slightly accretive, but let's just say for the record, it's materially neutral free cash flow. It doesn't change anything about next year. It doesn't change anything about the buildup after that. It mechanically increases our revenue, which at constant profit decreases the percent of profit, but it's purely just moving cells in a spreadsheet for the same business with the same factors and strength that it has today.
Rich Hilliker
Got it. That's helpful. And I know you talked about like a long-term optimization of this effort. I guess -- so you hit a little bit on the financial side of it. Could you talk about maybe like what you can do with this added visibility? You talked for just a second here on closed loop marketing as one of the examples. But -- how are you going to harness this extra visibility, this extra ability to engage with your customers?
Andrew Anagnost
Yes. I think the best example is to talk about what EBAs have done for us and for our customers and how we can take -- what we did with our enterprise business agreements and bring it down market to our partners and to customers that want to interact directly with us. Why are we so successful EBAs? We have absolute full visibility to what's going on in an enterprise business agreement. We know what they're using. We know how much of it they're using. We know where they're having success. There's a customer success representative dedicated to an EBA, sometimes a team of customer success representatives dedicated to a particular EBA customers, and we have intimate knowledge of how they're using the Autodesk portfolio and where they're successful and where they're not.
That knowledge allows us to, one, either help them be more successful with what they have and spread usage to more of the company or it allows us to do cross-sell and upsell to new solutions that we believe are going to be advantageous to that. And imagine the power of taking all of that down market into the rest of our business where we have the intelligence, the dashboards that we don't have the day, by the way, because 70%, roughly 65% to 70% of the contracts that come into the company, we don't own the record. So we can't match account fully to usage. We have to kind of cobble it together. In the new world, it's all there. Our reps will have it, and our partners will have access to some of this information so that they can see where the customers are. So all of a sudden, you're applying that same engine of increasing penetration of design and make solutions or life cycle solutions, increasing use of APIs and doing upselling cross-sell available to everybody, right?
And then at the very bottom, if you're self-serve, you can see what you're doing, and we can actually serve up to the self-serve customer and you should be looking at this. you should be using this. You should be considering this. FYI customers like you are using more of this than you are. That's really powerful. So it allows us to bring really high-touch upsell, cross-sell to medium touch of upsell cross-sell to low to no touch to every segment of our customer base.
Rich Hilliker
Got it. You can really help partner with them even more strongly greater visibility.
Andrew Anagnost
And it allows us to manage the business as new accounts in, growth in existing accounts.
Rich Hilliker
That makes sense. I guess sticking with similar motion here, you've talked about direct sales for a while and how that would continue to grow. And I think that this transactional model kind of leads us even further in that direction. So I guess how would you say that like direct and digital are becoming even more important to Autodesk in terms of a new customer acquisition perspective. How does that shift now?
Andrew Anagnost
Well, look, the digital channel is responsible for three quarters of our new customer acquisition, right? So it is a massive new customer acquisition engine and it will continue to be. And if we can bring more solutions into that, what we can do is get more beachheads further down into the market, where maybe we might not have reach or visibility. And a lot of that business often translates either into future business for us or future business for some of our partners. So we want to make that digital acquisition engine even more effective and more efficient, closed loop marketing, all the activities that are associated with that. Things we can already do there, we'll be able to do more effectively in the future. So yes, that will absolutely have consequential impacts on our ability to do that, not just down market digitally, but in the middle as well.
Rich Hilliker
Got it. Okay. Sticking to go-to-market, but I kind of back to those intentional changes that you've been making. I wanted to just quickly hit on the billings shift one more time. You guys highlighted the impact of cash flow and RPO. So I think that you've been messaging this for a while now. I'm wondering, from a customer perspective, does that create any disruption for them? Does it change how they engage with you? Is it just kind of a formality and any changes from that.
Andrew Anagnost
It's mostly a net positive for customers. Very few customers wanted to buy multiyear contracts from us upfront. A few would do it because they had some money in a quarter, and they just wanted to lock it in and get done with it. Most likely ability to lock in price and pay annually. And we believe, over time, we'll be able to increase the percentage of multiyear contracts because people will be chasing these lock-ins more in the future. So for a customer, this is neutral to positive, right?
For a partner, I mean, there were some impacts, partners love those upfront -- those upfront deals. For a partner, that's cash now and cash now is good. So they've had to adjust to it. It's a little bit of a billings transition for them as well. But for our customers, it's really neutral to positive.
Rich Hilliker
Okay. And does this change any sort of like ability to I don't like deal making, right, and discounting. Does this -- how does that impact that picture as you think about?
Andrew Anagnost
Yes. I mean, particularly with the new transaction model, I mean, we'll probably do less discounting. All right, because we won't be trying to use discounts as a way to motivate the channel to certain types of activity. We'll do more of these things with direct and targeted marketing activities to a larger segment of customers. We'll be able to segment the customers better and target them with specific things that are right for them rather than doing kind of these blanket discounts and things that are historically kind of how the company operates. And not that we won't do discounts, there'll probably be a lot more targeted and a lot more deliberate in how they're done. So it definitely changes the way we make deals and we engage with the customer.
Rich Hilliker
Totally. More targeted and more deliberate. Got it. Why don't we shift over and talk a little bit about products, given successful AU just a couple of weeks ago.
Andrew Anagnost
It was a really good AU.
Rich Hilliker
Packed crowd. It seems obviously [indiscernible]. So Autodesk AI, AI is not a new topic to Autodesk, but you came out with this announcement. I guess maybe like just to kind of level set, you've been working with or incorporating AI and some capacity for years. So maybe below this headline announcement, like what -- like how does this change what you've been doing? What were you doing before? How does this new announcement augment that?
Andrew Anagnost
Yes, I think it doesn't fundamentally change what we've been doing, but it does engage the customer differently in the conversation. We've been talking about AI for a decade with our customers. We were the first to roll out generative design in some of these capabilities. And we've either been trying to scare or inspire our customers with AI for quite some time. Now they kind of believe us. So putting a name on it and making it clear because Chat GPT, [indiscernible] these things, they made this stuff real for people. We always knew it was real. Most people in tech knew that we were going to be doing things with this.
So fundamentally, our road map hasn't really changed, except maybe inside the company where we're using more AI tools to work on internal productivity, which is new. But some of the tools that we were rolling out to customers like Construction IQ, or automated search capabilities or generative scheduling, these things were road map years ago. And there are things that we've been working on as vertical AI pieces moving forward.
What is fundamental here and will become more important as we move forward is the ability to train what I like to call large model models, large language model that actually significantly automate processes for creating a 3D model. It's complicated to create a 3D model. It's a lot of work. If we can take work that was months to weeks or month work that was weeks to days or days to hours, we have had a huge impact on our customers. And we have access to data, particularly in the manufacturing segment and emerging in the AEC segment that we'll be using in highly trusted ways and transparent ways to train models that could heavily automate these processes.
And that -- while it's something that we've been looking at for a while, the rate and pace has definitely accelerated and it will have meaningful impact not only to every customer using our existing products, but there's also the opportunity to deliver custom models to individual customers.
Rich Hilliker
Got it. Okay. Large model models, love it. Sound like tongue twister. So data is obviously essential to AI and unlocking all in the whole.
Andrew Anagnost
Can't have AI without data.
Rich Hilliker
That's right. And you've talked in the past about how Autodosk has a data advantage. I'm wondering if you can kind of just spell that out for everyone, just a little bit further, both in terms of your breadth and your ability to compliantly access that data I know that's a big topic right now.
Andrew Anagnost
Yes. So look, we're very clear about how we use the data in our terms of use and tried to simplify these things and make it transparent. We'll be even more transparent in the future. We're super dedicated to that because we believe engaging with our customers on this topic in a trusted way will make the ecosystem better and will make our customers better. And we have to continue to engage in an ongoing way, respecting their data and they're trusting us to do things with it that makes them more productive. So first and foremost there.
The thing that's unique about Autodesk is especially manufacturing, we were very early to the cloud. I mean, Fusion has been in the cloud for about a decade. As a result, we have a large accumulation of cloud-based data and design. The ramp-up to the cloud in AEC was a little slower, but post pandemic, it is definitely much more rapid than it's ever been in the past, particularly in construction as well, but even in design. As a result, we have a very rapidly building set of data that we can use to train large model models and do some of these things. So we have to do it responsibly. We have to do it appropriately. We'll probably start manufacturing and move down to other segments. And we hope as we move forward that, especially as we start to create custom models for customers that they trust us enough that they let us participate with more and more interesting data that we wouldn't normally have access to, to train and increase the potential for the whole ecosystem.
I think that will come with time, but it has to be built on a foundation of trust. And frankly, most of our customers know that as individuals, they will never have enough data to trade a model that meaningfully changes the rate and pace of how fast it takes -- how long it takes them to develop a complete model, detailed model of an entire project. So they know they have to work with somebody, and we want them to work with us.
Rich Hilliker
That makes sense. And I think kind of continuing on that thread of collective trust and data, I believe you made some announcements recently. I think it was through Forma with industry partnerships around data. Can you just tell us a little bit more about that, your vision here, why this is valuable to both customers, these partners and also Autodesk?
Andrew Anagnost
Yes. So we're doing something that is kind of different for Autodesk in the past. Autodesk 15 years ago would keep closed file formats, Raven to drive upgrades and all of these things. And now what you're seeing with a lot of this platform services, new applications like Forma is you're seeing very open data APIs. People can get into the data to fairly granular levels and extract information they want or integrate them tightly with a competitive product. This is core to what we're trying to do. We are not trying to get the data in, lock it into our system and not allow people to connect it to other processes. That's just not the future. The future is you have to have enough APIs to create enough openness for the data so they can connect it wherever they need to connect it.
Even in Forma, it's natively built to be extensible. So we have third parties integrating their IP into Forma as add-ons. We have a great third-party partner, TestFit that integrated some of their automated parking solutions right into the Forma tool. So it's in the Forma window, you can not only do building placement and wind analysis, light analysis, but you can also do parking capacity analysis. And that's the future, building these large ecosystems out and they have to be open.
Rich Hilliker
Well, I think we could probably talk for a whole another half hour. I've got more questions, but we're out of time. So we want to thank you so much for being here, Andrew, and the whole Autodesk team. It was great to be up here and thanks again.
Andrew Anagnost
Thank you, Rich. Take care.
For further details see:
Autodesk, Inc. (ADSK) UBS Global Technology Conference (Transcript)