Autodesk (NASDAQ:ADSK) - Featuring the model of a modern-day transition forecast
Autodesk reported the results of its fiscal Q3 yesterday. The results significantly exceeded the expectation of the consensus with revenues of $490 million and EPS of -$.18. Prior expectations had been for revenues of $477 million and an EPS loss of -$.24. That sounds good. It really isn't all that significant.
Simply put, the purchase case for Autodesk shares has been, and remains, that the company's transition program, in which it moves its entire revenue stream from on-prem to subscriptions and to the cloud, is going to create what many believe to be incredible financial results. Note that the meaning of the word incredible is that it is not believable. Management has forecast that revenues will achieve a 24% CAGR over the next 3 years, and that free cash flow will increase by more than 4X. There is nothing in the quarterly earnings release that could be construed as impacting that argument; some observers, including myself, might feel that the results in terms of net new model subscriber adds are a positive signpost toward the company's goal. This goal has been articulated many times by management, and it remains their mantra.
I think it's worth repeating. Headlines aside, the shares are not down because of the guide the company gave for the current quarter. The specifics of the guide have no relevance to the transition. The one figure that is of significance - net sub adds - saw an increased forecast. It will be several years before headline numbers in terms of either revenues or earnings have relevance in determining the progress of this company.
The shares have sold off a bit and are down a couple of per cent as I write this. But Autodesk shares have recently achieved all-time highs in the wake of the Trump rally and the expectation that a boom in infrastructure spending will lead to stronger demand for its AutoCAD product line. I think the expectations with regards to AutoCAD seem reasonable. The shares have appreciated by 50% since post-Brexit lows and by about 20% since the company last reported results. There isn't all that much that the company might be able to do to keep up that pace - that is the basic reason behind what has been reported as a pullback.
I have written in some detail about ADSK twice now, and the shares have done well, and perhaps better than I might have anticipated in a short time frame. They are up 27% since my first article was published in April, and that compares to an 11% gain for the IGV software tech/index over the same time span. In the short term, and in the wake of its new high, I do not suppose there is much positive alpha remaining in the next few months unless more investors decide to put money on Trump-influenced names.
The shares seem unlikely to appreciate another 20% over the next quarter - unless the company gets consolidated, a low odds bet, in my opinion. I will take my position off the table for now and look for some kind of a pullback to re-establish my position. Long-term investors really don't need to do anything. Life is decent at Autodesk, and exogenous factors, including both the Trump play and the assault on pirates, seem real enough. (Management believes more than half of the population of AutoCAD seats have come from illegal sources.)
Autodesk - A company with many, many numbers
ADSK, in the midst of its transition, is a company that reports lots of numbers, and is said by some, therefore, to have many moving parts. It really has but 3 major revenue streams to consider, and is not particularly a complicated company to analyze in that regard. Given that ADSK has gone cold turkey with regard to how it offers its product, many of the numbers the company has just reported and will report in the near future do not serve to illuminate the success of its business. There is no feasible way of adding any value to the understanding of readers and investors without referring to some of the metrics that the company reports and that are the source of headline commentary in analyst reports. But for those looking for an avenue to avoid the tedium of the torrent of numbers, suffice it to say that a couple of the metrics were displeasing to some analysts and require more explanation. But with all the numbers that are produced, it is possible for a roomful of analysts to produce a "buildingful" of opinions supported by one metric or the other. As was once said of Mark Twain, reports of the company's disappointing outlook are greatly exaggerated.
Most headlines in the wake of the earnings release are talking about what is described as a "light" forecast. And the headlines look that way. But one doesn't have to go very far beyond the headlines to see that what seems bad is good - I won't quite say what seems good is bad.
In a transition from a model based on the sales of licenses and maintenance to a model based on subscription transactions, the more quickly revenues decline, the faster the transition is going. If you own, or are thinking of owning, ADSK shares, you are simply unconcerned with the amount of revenue the company is generating from legacy sources.
Yes, the forecast for traditional headline metrics looks light, but it is in fact what might be hoped for from a modern-day transition forecast. (That lead-in to this section is taken from a song, "I am the model of a modern Major General" that is part of the Pirates of Penzance, a Gilbert and Sullivan operetta of the late 1870s. For those interested in the origin of the title, which has to do with the abolition in Britain of the Purchase of Commissions in its armed services, I am happy to supply it on request - but beware, one ought to be familiar with the political situation in the UK and the rivalry between Disraeli and Gladstone of that era.)
Analysts appeared to be concerned about a relatively arcane statistic called growth in annual recurring revenues (ARR), which came to $43 million this quarter as compared to $63 million last quarter. Was that a problem? Without presenting all of the details that go to making up that particular number, the end of license sales this past quarter meant that a former expense has now been classified as a contra-revenue. Beware contra-revenues. In addition, mix swings and the specific timing of larger enterprise purchases also lead to what appears to be a "light" quarter when it comes to ARR growth. Autodesk offers many different subscription plans with different contract durations and different prices. That also had a role in how ARR is reported, but not in what longer-term revenue expectations ought to be.
In addition to ARR, usually investors are interested in the growth in the number of subscriptions. New model subscriptions, the most important number in this group, grew by 168,000, which was quite a bit above prior forecasts and expectations. The net increase in total subscriptions for the quarter, excluding those acquired when the company bought Solid Angle, was 121,000.
In the prior quarter, the growth in subscriptions was 109,000 and the growth in new model subscriptions was 125,000. The company raised the low point of the range it has used to forecast subscriptions substantially; surprisingly, no one on the call chose to focus on that metric, which many think to be the most important metric at which to look in forecasting the current business outlook for a transition company.
Subscription growth ought to be tightly correlated with ARR; well, if it isn't, then there must be an issue with annual revenue per unit (ARPU) and pricing. Again, not so fast. Mix and timing have lots to do with numbers as the transition gets underway. Not all subscriptions cover the same things for the same period. And so, the ARPU metric is subject to swings that signify little, although they engender many electrons and lots of paragraphs in analyst updates on this name. I do not believe pricing has been a particular issue for this company over the past 3 months. While overall sub growth was but a modest beat, new model sub growth was quite strong and would not likely have been a cause for the company to embark on aggressive pricing strategies.
Deferred revenue increased 26% year on year - a significantly stronger showing than was seen in the Q2 results. New model ARR was $414 million, which was up 90%+ on a constant currency basis, although the comparison has obviously lost much of its meaning with the end of subscription sales. As noted earlier, new model ARR increased by $43 million sequentially, or 12%.
Non-GAAP spending fell by 2% year on year. It grew by $8 million, or 1.5% sequentially. General and administrative and research and development costs fell year on year and were flat sequentially. Sales and marketing costs rose year on year because of the transition from what had been an expense to a contra-revenue presentation. Absent that, sales and marketing costs were more or less flat year on year. Expense management is one of those metrics with fewer ambiguities than other data released by ADSK. Expense management is perceived by most to be moving in the right direction.
Overall, if one sifts through the plethora of numbers to be considered, most analysts are actually increasing their long-term expectations for this company - and were the shares not already discounting some of that, it is a good thing.
Behind the numbers
In some ways, despite all of the moving parts, the Autodesk story when it comes down to the numbers is a pretty simple one. The company is forecasting that the results of its transition will bring it to $3.5 billion of revenue by fiscal year 2020 - essentially, calendar 2019. At that point, management is forecasting it will be generating free cash flow of $6/share. On the latest earnings call, the company forecast that it will reach $11/share of free cash flow by fiscal 2022, based on a 20% CAGR in subscriptions over the relevant time frame along with a 24% growth in ARR over that period. Just by comparison, revenues this year are forecast to be just above $2 billion and free cash flow is probably going to wind up around $150-200 million.
In a report I recently wrote about Teradata (NYSE:TDC) and its goals, I commented, "L'audace, l'audace, toujours l'audace." Some readers liked the phrase, others not so much. Still, the goals that Autodesk has are nothing if not aggressive, and from time to time, that can lead to naysayers pointing out all that has to go right for the numbers to bear out.
ADSK will reach those aggressive goals, management asserts, by expanding its TAM and because of the impact of its subscription-based revenue model, which, it is hoped, will dramatically reduce the piracy that plagues the company's AutoCAD offering. If you believe management's scenario, owning the shares is a simple decision. If you do not - well, there is little other reason to own the shares, I believe.
I do not propose to do a deep dive into all the product offerings that make up the Autodesk fleet of solutions. Many readers will know AutoCAD, which is the leading solution used by architects and space designers. AutoCAD is by far the company's most important product.
The company is also a leading competitor in the PLM space, where its solutions are sold against those offered by PTC Inc. (NASDAQ:PTC), Dassault (OTCPK:DASTY) and Siemens (OTCPK:SIEGY). Overall, a well-regarded market research source suggests that Autodesk has a combined share of 19% in the CAD space, but it is the dominant player in architecture and a much less dominant player in the PLM space.
I have written in the past about the huge bet this company has taken in developing an all-cloud PLM offering. Right now, the business model transition is what is animating the financial discussion and has the most influence on the share price. But from a business standpoint, Fusion 360 is one of the more important solutions this company has developed in a decade, I believe.
The ADSK Fusion 360 offering has been growing share quite rapidly, according to the accessible statistics, and is one of the more successful product launches that the company has had recently. I think it is relevant to comment that there is a significant difference between a subscription model - the direction that has been taken by PTC thus far - and Fusion 360. Different managements, different development strategies. I think the cloud will ultimately be the repository for the data that remains on-prem in most PLM use cases. A web publication called SolidSmack has written about Fusion360 in the terms usually reserved for trendy performers. If readers have an interest in the subject, follow the link above so that you can make your own judgments on the subject. If ADSK were to be a target for consolidation, it would be the web solutions on which it has devoted its development focus that would drive a potential acquirer.
One of the more frustrating issues with regard to modern investor communication is that conference calls can come and go without analysts asking any questions regarding the state of the business. There was some of that on the Autodesk call. About the only question that was raised with regard to the company's product strategy related to its bet on the cloud. Here is the specific commentary of the CEO, Carl Bass, regarding what is going on when it comes to technology and the positioning of this company. (Note that the company is hosting an Analyst Day on December 6th, during which it will obviously spend lots of time addressing technology, TAM and competition.)
"As I have said before, while everybody is rightfully focused on the business model transition of our existing business and how we are using it to attract new customers, what I think people will (must) be surprise(d) about as you look out the next couple of years, is the size of the cloud business that we build and we really expand our TAM. So, we are going to spend a bunch of time next week showing a little bit more, you (the questioner) were at AU (Autodesk University) and got to see it, we'll try to put kind of an investor look on it, so that people next week can understand what we are doing and what the important(ce) is, and why we are investing in the cloud and why this really is the next generation architecture."
I have said that in the very short term, ADSK shares have probably achieved a significant percentage of positive alpha that may have been available. But over the long term, the key to these shares is a combination of the business model transition coupled with a better mousetrap. I believe the company has built a better mousetrap, mainly because users and industry analysts think that to be the case. And I think the source of this better mousetrap is a defensible, new web-based architecture which is not offered by competitors at this time.
Autodesk has a set of differentiated tools in specific markets, and its lead in that regard, as compared to competitors, is likely to persist for years into the future. It is also clearly a Trump infrastructure stock. And the new business model is going to lead to greater consumption on the part of ADSK users, who will trade dollars for flexibility. Lots of things to like, but the valuation subsumes all of that, I believe.
The one thing about writing this sector is that no one can possibly use current headline financial results to arrive at a valuation. So, I won't try. How much is the worth of a stream of $600 million/year that starts in 2020 and continues to rise for some years at rates greater than 20% actually worth. I have seen analysts who say they believe in the company's transition and yet value the shares at just $71. And therein lies the problem in continuing to recommend this name in terms of its ability to produce significant positive alpha. From here on out, it will be a race between free cash flow growth and valuation. There is very little margin for error.
The current average consensus price target for ADSK is less than $70. It is very hard for share values to appreciate when the average share price target is less than the current value of the shares. Some of that is almost certainly a function of those analysts who do not believe the forecast numbers going out to fiscal 2020. And some of that is a function of analysts who say they believe the numbers, but then use very high levels of discount rates to put their thumb on the scale. But even analysts using normal estimates come out with a DPV target of just a little greater than the current value.
How much is a stream of $600 million of free cash flow worth? I would have to think that if it were to grow at 24% in the terminal year, the flow would have to be worth $15 billion, or possibly more. The shares currently have a market cap of $16 billion and an enterprise value of $15 billion. My best guess is that ADSK shares will be in share price purgatory for the next few quarters. They remain suitable for long-term investors. I believe the shorter-term positive alpha potential has been achieved, and the risks and rewards are basically balanced at current share price levels.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.