Autodesk: Is Management Credible Or Will This Remain A Play Of The Optimists Forever?

| About: Autodesk, Inc. (ADSK)


Autodesk is in the midst of a multi-year transition from a on-prem model to a SaaS model.

During the course of the transition, headline metrics will be ugly, but things like sub adds, increase in deferred revenue balance and churn can guide investors.

Autodesk derives more than two-thirds of its revenue from cyclical businesses.

The company's forecast for fiscal-year 2020 - essentially 2019 - implies valuations substantially above today's level with free cash flow generation reaching $6/share by that time and with further significant upside potential.

Is there a chance for ADSK to become more like ADBE?

Sometimes in life, great divergences turn on a thread. Once upon a time, Autodesk (NASDAQ:ADSK) and Adobe (NASDAQ:ADBE) were actually considered in the same breath. After all, they both sold software to creative people, both hobbyists and professionals. At some level, their software was quite similar - although I don't want to take that one too far. And then, Adobe developed its Creative Cloud subscription model, and from being a middle-aged boring software vendor, it has become one of the icons of the cloud world with growth of 30% or so and the ability to consistently beat forecasts.

Autodesk has gone down a different path. It obviously was badly hurt by its focus on selling products to design professionals as opposed to hobbyists. While construction and product design capabilities have come back from the depths of the Great Recession, the hobbyist market has proven to be far larger with new applications for Creative Cloud developed almost daily. Users can buy Creative Cloud for $50/month, and if a potential user is just getting started, he/she can buy Photoshop for $10/month. At this point, Creative Cloud has more than seven million users who spend close to $4 billion/year on lease payments.

There are probably not seven million architects, construction management personnel and product designers in the US. That doesn't mean ADSK doesn't have some significant business opportunities. It competes in the simulation market, which it believes has a TAM of $3.2 billion. Autodesk's most popular product AutoCAD sells for $30/month for the LT versions, which obviously puts it into the mass market category. Even its Product Design Suite costs $280/month. With that kind of price level, the company has to build a base of several million users in order to grow. Anytime a business software vendor needs several million seats to grow, it is going to be a significant struggle.

ADSK is a company within the PLM space where it has to compete with a range of competitors including PTC (NASDAQ:PTC) and Dassault (OTCPK:DASTY). The company suggests that there is a $4 billion opportunity in PLM with 10 million potential subscribers although that seems a stretch to me based on the seat data provided by DASTY and by PTC. It is one of the hundreds of companies vying for a place in the nascent IoT (Internet of Things) market space. IoT is supposed to be a gigantic market over time - at the moment ADSK competes in a relatively narrow sector of the space in which it supplies preventive maintenance for connected machines.

If I had to summarize the differences that have now emerged between ADSK and ADBE, it is that ADSK essentially provides solutions to help its customers to make things whole. Things represent a pretty broad category, and it includes buildings and other structures, parts and designs for almost everything manufactured and even solutions that help manufacturing plants run better. But things, nonetheless.

Adobe has solutions that allow its customers to create consumer experiences. The TAM for ADSK is said to be $25 billion in total. The TAM for Adobe is probably several times larger and it has a much higher market share in its market space than ADSK will ever have in the spaces in which it competes.

In this country anyway, our economy is far less about making things and far more with providing consumers with experiences. That doesn't necessarily mean that ADSK is a bad investment. It does mean that it is much easier these days being Adobe than it is being Autodesk.

You know, transition investments can be pains in the neck. Sure there are success stories like Aspen (NASDAQ:AZPN) and Adobe, but there are more than a few other companies starting with Oracle (NYSE:ORCL) and SAP (NYSE:SAP) that are having a dickens of a time in their transition to the cloud. What inning are we with this name?

The reason that analysts and investors take the trouble to deal with software companies in the transition to cloud is because of the hugely profitable business model that comes out on the other side. Over the past five years-plus, ADSK's shares have dramatically underperformed the performance of most software shares. The shares were $46 in May of 2011 and they are $58 now. The last time someone upgraded this name was back in May 2014. I wanted to think of some significant pop culture event that happened back in 2011 - but I remembered I know absolutely nothing about pop culture.

No, the real reason to take interest in these shares is the company's FY 2020 business model forecast which calls for $3.5 billion in revenues, up from $2.5 billion reported in FY 2016. Further, free cash flow, which was $345 million in FY 2016, is supposed to reach $1.3 billion or $6/share by FY 2020 and rise further to $11/share by fiscal 2023. In addition, non-GAAP margins, which were less than 10% in fiscal 2016, are supposed to reach the mid-20% range on the way to 30%-plus. If it all works out then, EPS would be close to $3.00/share in 2020 and something like $6-7/share in 2023. That's what happens with a SaaS model that features high renewal rates. Management's forecast is that company would be achieving sustained low-double-digit growth in a predictable fashion by that time going forward. As the saying goes, "nice work if you can get it."

Lots of managements are asked by lots of investors in quarterly conference calls about what inning they are in terms of the transition to cloud. It really isn't a very easy question to answer. Even a company like Adobe still has millions of customers using Creative Suite, its non-cloud offering that hasn't been available to new purchasers for years. But for what it is worth, ADSK's management has proclaimed that this year, fiscal 2017, will be the year of its inflection - by which I suppose it means it will be the first year its business is going to be without the revenue support from the sale of perpetual licenses (The end of sales of perpetual suite licenses comes in Q2 of the current fiscal year, which will further muddy the waters in terms of reported numbers).

One indication that Autodesk is still in the early innings of its conversion is that its fiscal Q4 of 2016 that ended on January 31 was finally the last quarter in which perpetual licenses were available for individual products. Q3 will be the cut-over point at which the company stops all sales of any license agreements. So it has taken the company until now to almost cut the cord, so to speak, and to wean itself off the narcotic revenue of perpetual licenses. Cutting the cord is going to produce some pretty ugly financial results this year. Revenue as reported is expected to decline by 20% and non-GAAP EPS is supposed to be a loss of $.85 to a loss of $.60. SA Columnist Eric Jhonsa described Q1 guidance as "soft" and indeed it was, but that is the inevitable result of ending sales of desktop perpetual licenses now and ending sales of suite professional licenses at the end of Q2.

There are a variety of metrics that different companies have used to talk about their progress. In the case of Autodesk, it is clear that the company's transition to cloud is in the very early innings. The company has at least 2.8 million engaged users who are not subscribers. The company obviously has millions of seats that are unconverted but for which users are still paying maintenance. In one ADSK presentation, the company showed an example of what would happen if 20% of its non-subscriber AutoCAD customers switched to the cloud. That alone would increase Annual Recurring Revenues (ARR) by 45% alone and increase subscription revenues by 21% for AutoCAD.

As I wrote a few days ago with regard to Oracle, converting the non-cloud maintenance base to subscription is one of the most difficult challenges legacy vendors face. In the case of Oracle, of course, the problem is of enormous magnitude since 52% of its current revenues and more than all of its operating profits are coming from the maintenance line. But ADSK probably gets nearly 40% of its total revenues maintenance from on-prem installed customers

ADSK has made a start in the conversion process, but at the moment, it estimates that less than 20% of its enterprise customers have moved from standard on-prem maintenance to a cloud-based pricing model. That model is called Token-Flex, and as the name implies, the model is based on tokens that users purchase when they consume a fixed amount of connect time to the software. I confess that I sometimes wonder why companies choose the figures they do to put in their presentations - but for what it is worth, the company enjoyed a 90% increase in enterprise ARR during FY 2016. How the uptake of Token-Flex pricing relates to traditional maintenance I do not know other than to assume Token-Flex is driving more revenues. For smaller users, Autodesk offers Multi-User subscriptions to replace network maintenance. In a typical 50-seat account, revenues from the transition have increased between 25% and 50%. For the smallest accounts, the transition is increasing revenues by between 50% and 100% as they transition to Single-User subscriptions.

Overall, the company is projecting a 24% growth in ARR, a 3% growth in annualized revenue per subscription and a 20% growth in subscriptions over the next four years. Currently, what this company calls its maintenance model ARR is $1.12 billion out of total revenues of $2.5 billion and of total recurring revenues of $1.38 billion. Over time, much of those revenues are going away. Self-evidently, revenue growth and more importantly earnings should be strongly positively influenced by the transition of the Maintenance Model ARR to what the company calls the new ARR, but that transition will not take place overnight or without some hiccups that are currently unknowable.

To answer the question posed above, if Adobe and Aspen are in the eighth inning of their transitions, this company has not gone beyond the second inning of its transition. There are two ways to look at that. As a stock, ADSK obviously will offer far more percentage upside as its transition proceeds. As a business, the company's reported growth in earnings, revenues and most everything else will remain hobbled until the crossover point ensues in a couple of years. One pays a huge price for successfully completed transitions as might be expected and the trick is in finding in-process transitions that are still modestly valued but have the opportunity to come out with a good result on the other side.

Look seven-year projections and four-year projections are nice - but I am not buying this name to wait several years to see if they are meeting their targets. I can't even figure out what I'm going to be doing four days from now, and ADSK, after all of its ups and downs is forecasting four years and seven years. What are the short-term metrics that will allow us to judge this company and how do you rate its chances of achieving those metrics?

I think that the simplest metric to look at is going to be net subscription adds for the next couple of years. The company is going to report that metric; it is going to tell investors if it made or missed its target for net subscriber adds and it is focusing its sales force on that metric as well. Oh, there are other metrics and certainly I would look at ARR, ARPU and of course cash flow. Do not look at traditional headline metrics such as revenue and reported EPS. They are not going to help anyone in understanding how this company is doing relative to expectations. It would be far better to see a revenue "miss" with strong net sub adds and bookings than to see a decent revenue quarter that missed on net sub adds and calculated bookings. The most important sub-headline metric is going to be calculated bookings which involve looking at the net change in deferred revenue coupled with reported revenue.

The real issue of course is going to be whether the entire investment community chooses to look at what is important now and doesn't fixate on extraneous factors. As most investors know, ADSK has some activist investors who have filed on the company. Activists include Sachem Head and Eminence. At the moment, there is a peace deal in place with the activists who own a bit more than 11% of the shares, but the standstill agreement expires the end of this coming September. The activists now have three board members in place out of an 11-member board.

One thing is true for sure and that is that this transition will be nowhere near complete by the end of September and reporting one or two quarters of either good or bad net sub adds is really not going to be enough to prove the company's strategy.

It is, I think, pretty unknowable about what the activists might do in September and what is going to influence their decision. But it is either an overhang that can't be totally disregarded (the activists decide to sell shares) or a factor that might set-up a transaction in which this company is sold.

At the moment, most analysts are dubious regarding the transition or feel that investors will not focus on signs that the transition is succeeding. The company has a mean analyst rating of 2.5 on a scale of 1 to 5, indicating that the consensus rating is hold. The mean price target is $62, or just 7% above current quotations. On the latest conference call, a prominent analyst essentially asked the question of how anyone is going to believe four- and seven-year projections from a company with a rather mixed record, to say the least. It is probably best to quote the CEO's answer in part. Carl Bass said:

"You will also be able to look at things like churn and see the renewal rates and extrapolate from there. I mean to the extent that you want to discount FY 2023, feel free to. It's way in the future... the problem we had and really the reason why we gave details around FY 2023 is because FY 2020 and FY 2021 is still so much in the steep part of the curve... And just to the extent now that traditional financial metrics are going to go down... it's just as certain they come up the other side."

What I like about the set up is that the share price, to an extent, is saying that investors either don't believe management about 2020 results or think that until the company starts printing attractive headline numbers, investors won't care. Obviously, investors in this name are investing on a speculative basis that can't be proven. On the other hand, the prospective rewards are substantial enough to take the risk if the investor believes the management. Again using ADBE as a bit of an analog, investors certainly didn't have to wait until the completion of its transition before they were well rewarded and became comfortable with the management projections.

Looking at the current environment, and looking at how users increasingly want to buy software, my guess is that the transition will succeed. I think the company is going to achieve the level of sub adds that it's forecasting of 475k-525k. I assume that management will constrain the growth of opex as it has forecast.

What is a bit harder to forecast is the global business environment both for this current fiscal year and the future. This is a company tied to a lesser or greater extent to global macro trends. Its revenues from architecture and construction are 34% of the total. Its revenues from manufacturing are 25% of the total. The company's business in China, a not insignificant part of the total, has been relatively strong. On the other hand, its results for fiscal 2016 reflected an implosion in the company's business in Brazil and Russia. I think that the principal risk to the company achieving its current FY 2017 forecast is more macro related than related to its business transition.

OK-Wrap it up. Is the valuation at a level that makes sense for a risk tolerant investor?

It should be noted that none of the traditional valuation metrics are going to make a whole lot of sense during the course of the transition. For what it is worth, the company has an enterprise value of $12 billion, and sales this fiscal year are forecast to be $2 billion. Also, 95% of the shares are held by institutions. The company is forecasting a loss this year so any kind of P/E analysis is meaningless. The company doesn't forecast cash flow or free cash flow. Deferred revenues ought to grow significantly beyond the $360 million increase achieved in FY 2016. The company's net GAAP loss will be smaller this year than last because last year included a substantial one-time tax provision although that was partially offset by a significant increase in deferred taxes. A/R showed a substantial increase last year rising by $195 million or more than 40%. Some of that increase will be reversed in the current year as a product of lower sales and lower DSO. Considering all of those factors, cash flow should increase significantly from last year's $414 million level to a level of between $600 million and $700 million. Assuming the company makes no acquisitions in the current fiscal year, free cash flow ought to be between $500 million and $600 million, clearly not at a level that would support a $12 billion-plus enterprise value. It will be two to three years at least before it is possible to do any kind of normal valuation analysis for this company.

Some Closing Comments

Autodesk is in an interesting position as an investment. On the one hand, reported headline metrics are going to look ugly for this year and at least into the middle of the following year. The company is attempting the feat of moving from all perpetual license business to all SaaS in something like a five-year span. During the course of the transition, most traditional valuation metrics are not going to be particularly helpful in establishing a reasonable share value for this company. The company's goals, however, if they are achieved, include a 50% increase in revenues over the four-year span, EPS of about $3/share and free cash flow of $6/share. I think that if there is a reasonable expectation that operating results will approach such levels by FY 2020, the shares are worth substantially more than current quotations. This company has two activist investors whose standstill agreements expire the end of September. At that point, one assumes they will either sell their shares or find a way to secure a bid for this company.

I think the model transition is most likely to work as it is congruent with the way users prefer to acquire software these days. The cloud does change almost everything and certainly will change user preferences with regards to on-prem vs. cloud within the markets that Autodesk competes.

I'm concerned, to some degree, regarding the fallout from the macro environment as this company derives more than two-thirds of its revenues from highly cyclical industry segments. But the current penchant for substantial monetary easing is likely to boost spending, at least in the architectural and construction product classes. I think that the shares make sense for risk tolerant investors in that any signs that the transition is working as planned are going to be catalysts to drive the share price long before headline metrics improve.

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.