Autodesk Inc (ADSK) 2018 Q2 法說會逐字稿

(technical difficulty)

(Operator Instructions)

I would now like to turn the call over to David Gennarelli, Senior Director, Investor Relations.

Sir, you may begin.

Thanks, operator.

Good afternoon.

Thank you for joining our conference call to discuss the results of our second quarter of fiscal year 2018.

On the line is Andrew Anagnost, our CEO; and Scott Herren, our CFO.

Today's conference call is being broadcast live via webcast.

In addition, a replay of the call will be available at autodesk.com/investor.

As noted in our press release, we have published our prepared remarks on the website in advance of this call.

Those remarks are intended to serve in place of extended formal comments, and we will not repeat them on this call.

During the course of this conference call, we will make forward-looking statements regarding future events and the anticipated future performance of the company such as our guidance for the third quarter and full year fiscal '18; our full -- our long-term financial model guidance; the factors we use to estimate our guidance, including currency headwinds, our maintenance to subscription transition, ARPS, customer value, cost structure, our market opportunities and strategies and trends for various products, geographies and industries.

We caution you that such statements reflect our best judgment based on factors currently known to us and that actual events or results could differ materially.

Please refer to the documents we file from time to time with the SEC, specifically our Form 10-K for the fiscal year 2017, our Form 10-Q for the period ended April 30, 2017, and our current reports on Form 8-K, including the Form 8-K filed with today's press release and prepared remarks.

Those documents contain and identify important risks and other factors that may cause our actual results to differ from those contained in our forward-looking statements.

Forward-looking statements made during the call are being made as of today.

If this call is replayed or reviewed after today, the information presented during the call may not contain current or accurate information.

Autodesk disclaims any obligation to update or revise any forward-looking statements.

We will provide guidance on today's call but will not provide any further guidance or updates on our performance during the quarter unless we do so in a public forum.

During the call, we will also discuss non-GAAP financial measures.

These non-GAAP measures are not prepared in accordance with generally accepted accounting principles.

A reconciliation of our GAAP and non-GAAP results is provided in today's press release, prepared remarks and on the Investor Relations section of our website.

We will quote a number of numeric or growth changes as we discuss our financial performance.

Unless otherwise noted, each such reference represents a year-on-year comparison.

And now I'd like to turn the call over to Andrew.

Thanks, Dave.

Our strong Q2 results are a continuation of the broad-based strength across all subscription plans and types and geographies that we experienced last quarter.

As we demonstrated over the past several quarters, we are executing well and making real progress on our 2 major initiatives: growing lifetime customer value by moving customers to subscription; and expanding our market opportunity with increasing adoption of our cloud-based solutions.

These consistent results increase our confidence in the model transition and our ability to achieve our goals.

There are several areas to highlight in Q2, including the fact that total ARR grew 23% at constant currency; that we added 153,000 total subscriptions; that recurring revenue has increased to 91% of total revenue; that we've overachieved on revenue and we've coupled that with strong spend control, which has led to better-than-expected EPS performance.

In addition, the maintenance-to-subscription program or what I'll refer to later as M2S, is off to a great start.

Now let's take a closer look at our Q2 performance.

The trends we're seeing in annualized recurring revenue are clear signals that the transition is working.

Subscription plan ARR nearly doubled on a constant currency basis, driven by the strong uptake of all of our subscription plan offerings.

Subscription plan ARR now represents 43% of total ARR, and we still anticipate it will become the majority component by the end of this fiscal year.

I'd also like to share with you another piece of data that really provides us with confidence in our ability to grow ARR going forward.

If we isolate ARR growth rates for AutoCAD LT and our animation products, which are further along into the transition than the rest of the products, we saw total ARR growth in the mid-30% range for these products.

A year ago, the growth rates for these products were similar to those we're reporting for our overall business.

The same effect can be seen in our reported revenue for these products, and it's a great leading indicator as we continue to move through the transition.

We added 270,000 subscription plan subs in Q2, led by continued strong adoption of product subscription.

63,000 of these subscriptions were generated from our maintenance to subscription or M2S program.

Even when normalizing for M2S, total product subscriptions more than doubled year-over-year with triple-digit growth in each major geography, including emerging countries.

New customers continued to make up a meaningful portion of product subscription additions and represented close to 30% of the mix for the quarter.

These new customers come from a mix of market expansion, growth and emerging, converting unlicensed users and people who have been using an alternate design tool.

Some of you noticed that we started out Q3 -- our Q3 promotion targeting legacy users a couple of weeks early.

Now this led to some conspiracy theories about our Q2 performance.

You should know that the timing of these global promotions is set months in advance.

Our rationale for starting early was the result of our experience from Q3 of last year.

Starting the promo a couple of weeks early allows for greater absorption of our communications to our partners and customers, which tends to take longer in the summer months due to vacations.

The mission was to get the promo details out to the channel rather than driving subs in Q2.

And as expected, an immaterial number of subs, about 2,000, were generated from this promo in Q2, but we positioned the promo for success here in Q3.

Subscription plan subs also had strong contribution from our new enterprise customers via enterprise business agreements or EBAs.

As expected, our net EBA sub adds were not nearly as much as the seasonal strong Q1.

EBAs with our large enterprise customers have been a successful component of our transition, leading to both increased subscriptions and account value while providing increased flexibility for our customers.

This increased flexibility has led many of our EBA customers to increase their usage as they adjust to the new licensing system.

This is a win-win situation, but it does create a short-term drag on ARPS.

For example, if we isolate just the population of EBA customers from June 2016, the monthly average user for -- usage for these accounts increased 10% in the last year.

However, we don't see a corresponding revenue increase until a customer does a true-up or uplift upon renewal.

This year, we have the opportunity to renew the first wave of Token Flex EBA contracts that we signed 3 years ago.

It's early days so far, but we're seeing, on average, that contract size is increasing by over 30%.

In Q2, we renewed 2 very large EBA deals worth over $10 million each.

In one of these deals with a European-based global engineering consulting firm, the renewal contract value was 150% greater than the original value.

This company's engineers are spending roughly 5 million hours a year using Autodesk products, and the EBA contract gives them access to our entire product portfolio.

The third component of our subscription plan subs is our cloud products.

This is the TAM expansion part of our transition, and we are extending our leadership in the cloud.

Total subscriptions grew by 200% and continue to be driven by BIM 360, our construction management and collaboration tool, followed by Fusion, our cloud play-based design and manufacturing tool and opportunity, which is where our cloud-based BIM 360 has gained an early leadership position.

We're utilizing the cloud to allow our customers to take their BIM models all the way into the field, giving building owners and general contractors a digital platform for collaboration, coordination and visibility.

This is a really big deal because it's something that has been sorely lacking in the past.

Now when it comes to the world of building, we have a powerful brand and a powerful reputation.

We're making significant penetration with the biggest general contractors in the world and with building owners and doing this by driving the value of the building information model into their project ecosystem.

The result is we're both expanding BIM 360 deals after successful initial implementations and we're signing new deals with companies and organizations that we've never worked with previously.

Building owners are starting to mandate BIM 360 to gain competitive advantage and project efficiency.

One Q2 deal was an international airport that is investing in thousands of BIM 360 subscriptions as part of a multibillion-dollar renovation project.

Their goal is to use BIM on all suitable projects to inform and enhance future facilities management.

They even wrote BIM 360 into the process for all future development projects at the airport.

Another great example is with a general contractor that influenced a large state university and a large municipality to write BIM 360 requirements into their specifications and permitting process.

This is the kind of success that builds on itself over time.

Now partially offsetting the growth in subscription plan subs was the expected decline in maintenance plan subs.

As we've said in the past, we expect to see ongoing declines in maintenance plan subscriptions, and the rate of decline will vary on the number of maintenance plan subscriptions that come up for renewal, the renewal rate, and the pace of the M2S program.

A little more than half of the decline in maintenance subs was the result of the fast start to the M2S program.

In fact, nearly 1/4 of all maintenance renewal opportunities migrated to subscription.

Of those that migrated, nearly 10% upgraded from an individual product to a higher-value industry collections.

M2S is yielding some early data that is very encouraging and very interesting.

I'd like to share a few other points with you.

Specifically, some customers are doing partial conversions of their maintenance fees, but if you look at it overall, the participating accounts are growing their total subscriptions and growing their spend with Autodesk.

In other words, accounts that participate in M2S are purchasing net new product and cloud subs, and we're seeing this behavior across all geographies.

Keep in mind that it reflects only 6 weeks' worth of data, but these early figures are better than expected and bode well for the next few quarters of execution.

As I've said, we'd like these maintenance customers to move sooner rather than later as product subscriptions provide the greatest value and increased flexibility, support and access to our cloud products.

Moving to a single model makes the most sense and will immensely simplify our business and how our customers transact with Autodesk.

Now let's talk a little bit about ARPS.

For the third consecutive quarter, we experienced a small sequential increase in total ARPS.

It's worth repeating that there are many things that will influence the short-term performance of ARPS, including product mix, geo mix and timing.

Another example is that the M2S program is having a near-term impact on ARPS.

M2S is having a positive impact on maintenance due to the price increase for those that stayed on maintenance and a negative impact on subscription plan ARPS due to the discount offered for conversion.

Product subscription ARPS grew year-over-year, but if we normalize for the effective M2S, it would have grown about 10% year-over-year and had its third consecutive quarter of sequential growth.

As expected, subscription plan ARPS is being negatively impacted in the near term by cloud subs as well as the increased usage of EBA accounts.

I spoke about these influences at Investor Day last year, so it shouldn't be surprising that the more successful we are with cloud products and increased usage with our EBAs, the more it will be a near-term drag on ARPS.

Of course, the flip side of faster-than-expected M2S conversion is that the more people that take the offer here in FY '18, the smaller the price increases we will realize in FY '19.

We'll moderate that impact by continuing to focus on the upsell to collections.

Having said all that, we remain confident that overall ARPS will be positively influenced going forward by less discounting and promotions to our legacy users, a price increase for maintenance customers and the migration to higher-value products.

We still expect to see ARPS inflect up by the end of the year.

Now I'll turn it over to Scott for a few more details on the financials.

Thanks, Andrew.

Driving more users to subscription aligns with another one of the transition-related initiatives, which is to drive more direct business.

Total direct revenue for the second quarter was 29% of total revenue.

That's up from 25% in Q2 last year and just 20% 2 years ago.

We continue to generate strong growth in the volume of business we're doing with our large enterprise customers, and we're experiencing exceptional growth with our eStore.

Over 30% of our North American product subscriptions have been generated through our eStore.

We expect to continue to meaningfully grow both our direct-to-enterprise and our e-store business as we go forward.

We're pleased by the broad-based strength we're experiencing from a geographic and product family perspective.

Each major geography, including emerging markets, reported sequential revenue growth.

And for the second consecutive quarter, we experienced a marked improvement in our performance in Japan, an encouraging sign as we've made some changes there.

We also experienced sequential revenue growth in each product family.

Moving to spend management.

We've been able to execute and drive results while firmly controlling our spend growth.

Non-GAAP spend increased by 1% in Q2 as expected.

We remain committed to keeping spend flat this year and next year, although we are seeing an increased FX headwind to our expenses.

We're achieving this through targeted divestments and reallocation of those dollars to initiatives that drive our transition.

Looking at the balance sheet, reported deferred revenue grew 17% against a tough compare last year when we had the end-of-sale event for the last perpetual licenses of suites.

We also stopped selling multiyear maintenance renewals earlier this year in conjunction with the M2S program, which adds to the headwind.

Unbilled deferred revenue increased by $33 million sequentially, which would've added another 2 percentage points of growth to deferred revenue.

Total unbilled deferred revenue now stands at $63 million, and we expect it to continue to grow meaningfully as we move more of our enterprise customers to annual billing terms.

I'll also note that during the second quarter, we issued $500 million in 10-year notes, taking advantage of the current low interest rate environment, and we've redeemed $400 million of debt that was due to mature in December.

That's a good lead into our capital allocation strategy.

We repurchased 1.2 million shares in Q2 for a total of $119 million.

That averages out to $102.71 per share.

During the quarter, we rebalanced our buying strategy by putting more weight on opportunistic versus systematic buying, and we'll continue to make adjustments to this program as we go forward, recognizing our ongoing commitment to share buybacks.

Turning to our outlook.

Our view of the global economic conditions remains consistent with our view over the past several quarters, with most of the mature markets performing relatively well and little change in emerging markets.

As we look at our outlook for Q3 and the second half of the year, we expect seasonal patterns generally consistent with the last 2 years.

As I mentioned, we've made some targeted divestment that allowed us to reallocate that money, but it also created a slight headwind to reported revenue, both of which are incorporated in our annual guidance.

Q3 will be the first quarter where the year-over-year revenue comparisons are apples-to-apples comparing back to our first quarter of subscription-only sales.

But remember that our Q3 fiscal '17 results included $38 million of licensed backlog that rolled over from the end of sale of Suites perpetual licenses in Q2 '17.

Normalizing for that, the midpoint of our Q3 revenue guidance range represents year-on-year growth of 13% as reported revenues now begin to inflect upward.

Based on our year-to-date performance, we're increasing the midpoint of our guidance range for revenue, and we increased the range for net subscription adds for the fiscal year.

Overall, our strong first half results increase our confidence that the transition is working for our customers and our partners.

It also sets us up for success for the rest of the year and reinforces our conviction in our fiscal '20 targets.

We've executed well over the past several quarters, and we're looking forward to building on this success as we progress through the rest of fiscal '18 and work toward our fiscal '20 goals and beyond.

Before we open it up for questions, I'm going to turn the call back over to Andrew for a few closing comments.

Thanks, Scott.

Obviously, one big change this past quarter was that I was appointed CEO.

I'm humbled by the board's decision, and I want to thank my friend and longtime associate, Amar Hanspal, for all his work and dedication to Autodesk over the past 20 years.

Secondly, I want to assure you that we remain fully committed to the FY '20 goals around ARR, subs and cash flow that we put forth 2 years ago.

Finally, while the near-term and long-term goals remain clear, I thought I'd share with you my focus areas for driving success and achieving these goals.

First, we are hyper-focused on enhancing the subscriber experience and delivering more value to our customers.

It has to be frictionless for our customers to manage their subscription with Autodesk, and it has to be obvious to them what is driving value from subscription.

Second, we are going to continue investing in our own digital infrastructure and creating opportunities for our customers to transact and engage directly with us.

We've made nice progress in this area over the past couple of years, and I want to further enhance and accelerate these advancements.

Third, I am absolutely committed to winning the construction space and winning in the new world of digital manufacturing.

The opportunity is enormous for Autodesk and our customers.

We've done well to establish an early leadership position and we are not going to slow down.

I hope that gives you a better idea of how I'm approaching my new job.

The next era of Autodesk will not be defined by simply product or business innovation, but in their combination.

We must excel at both, all in the service of our customers.

I couldn't be more excited about the opportunity to lead such a great company, and I look forward to reporting on our progress as we go forward.

Operator, we'd now like to open the call up for questions.

(Operator Instructions) Our first question comes from the line of Jay Vleeschhouwer with Griffin Securities.

Two questions.

One, to start longer term.

Andrew, how reasonable is it to expect -- or when would you expect that your direct business would become the majority of the business and/or that the eStore would become the majority of the direct at any point so that, of course, you're having the requisite back office and infrastructure which you had just alluded to a moment ago?

And secondly, you talked in your prepared remarks around large customer, EBA customer, usage and mix issues.

Could you address that more broadly in terms of your wider customer base, in terms of what you're seeing so far, in terms of usage of products, and specifically, your confidence that, in fact, the mix will go more and more towards collections and stand-alone than we might have seen to date?

All right, Jay.

So first, thank you.

So let me address the first question, and then I'm going to ask for clarification on the second one.

So with regard to the first question, I'm not going to give you a timeframe, but I will kind of bound the problem a little bit more for you.

In terms of a steady state for us, it's probably more like a 50-50 split between our direct efforts and our partner efforts.

And we think that's the right steady-state for us, not only from serving our customers well, but also basically from the structure of the business in terms of what the right kind of profitability mix is.

Now when you look at that moving forward, the percent of the business that actually comes from direct high touch, the major account subs, probably won't change that much over time.

All of that growth is going to come from the digital direct piece.

So actually roughly speaking, digital direct and physical direct will split that 50% another 50-50, so you can kind of see a 25-25-50 type model for our business moving forward.

Timeframe-wise, I don't want to give a timeframe for that, but it's definitely going to be over a relatively narrow time horizon.

Now your second question, I just want to make sure I understood.

You were asking what is the mix of usage within our EBA accounts.

Or just clarify your question a little bit for me, so I make sure I answer the right question.

Sure.

So you've, in the past, been able to track usage of the Suites, for example, because they were instrumented for that, and I imagine you should able to do the same now under subscriptions for all products, stand-alone and collections.

The question is really twofold: one, how confident are you that customers are, in fact, going to mix up towards more collections from stand-alone in terms of initial licenses?

And then once they have a collection, that the usage of enough products will be there for them to want to continue renewing the subscription on a collection, that there's enough indispensable products, let's say, within collections that basically your ARPS must go higher over time?

All right.

Great.

So thank you for the clarification.

I always want to make sure I'm answering the right question for you, Jay.

So first off, let's just -- let me kind of break the question up into some pieces.

Right now, the run rate for collections is already up to what our historical Suites run rate is.

So we're already back to where we were with Suites in terms of collection uptake.

The other little piece of information I want to give you, and it's kind of some additional color on the M2S program.

If you heard my -- the earlier comments, we said that 10% of the customers moving from maintenance to subscription were upselling to collections.

That's 10% of the total customers moving.

If you look at those who are actually eligible, when you subtract out people that are by default going from Suites to collections or people that actually don't have the option to move from a stand-alone to collection, it's almost 25%.

So what you're seeing is not only a run rate that's up to our historical Suites level, but an acceleration in the M2S migration that's looking really solid in terms of getting people to collections.

So it's looking good right now, Jay, and it's heading directionally in the right direction.

So that's kind of the current state.

Now to your next question about how confident are we with the product usage within the collection and the things associated with that.

So there's -- I'll answer the question by bounding it in 2 ways: one, historically with Suites, we've known that it only takes 2 or more products of usage for people to really just continue with the value proposition of the Suites.

We see no reason at all why that should change in the collection paradigm.

So we expect to see the exact same behavior for the people that have chosen to move to collections.

The thing that actually gives us even more confidence is the fact that unlike Suites, we've blended more of the cloud into the collection.

So there's actually more extension from the collection into other types of capability that provide value than there was in the Suites.

That's one.

The other thing that bounds it is you're looking at our experience with EBAs and what happens when we provide more portfolio access, it just inevitably drifts over time to broader product usage.

We see, on average, quite significant increases in the breadth of products that are being used.

So I -- there's nothing right now that indicates that we're going to see any kind of different behavior than what we've seen in the past.

Your next question comes from Saket Kalia with Barclays.

First off, congrats, Andrew, on the permanent appointment.

First, for either of you, could you maybe describe the mechanics of the M2S impact on ARPS this quarter?

You said in the prepared remarks that subscription ARPS would've been, I think, about $509.

So is this the impact of the discounts to convert?

I remember kind of hearing that the discount would be -- the Suite is now in decline over time.

So is that what we're seeing?

Or did you maybe see a different mix of conversions perhaps than you were originally expecting?

Yes, Saket, we saw all of the above.

We did see -- as people converted from maintenance to subscription, you remember the attraction was they got a significantly discounted product subscription if they turned in their perpetual license that they had.

So an existing maintenance customer that converted was able to buy that product subscription at a significant discount.

That obviously pulls down the ARPS on product subs.

The faster we can make that transition move, the faster we can move people off of maintenance and over to product subscription, of course, the faster the transition works and the better off we all are.

But it does have a short-term effect on ARPS.

If you just isolated product subscriptions for a minute and said, "Let's net the M2S impact out of it", that's the statistic that we gave you in the opening commentary.

Product subscriptions by itself, without M2S, the ARPS increased 10% year-on-year, and it was the third consecutive quarter of product subscription ARPS growth.

So part of what you're seeing is just the M2S effect of kind of a bucket shift of people moving from maintenance mid-quarter up into subscription.

All 63,000 subs moved, but only a fraction of the ARR was accumulated in the product subscriptions bucket.

We also did see mix, and we always see mix a little bit in Q2.

So our historical trend is that ARPS comes down a bit Q1 to Q2 because if both cloud -- and we talked about the success we had with cloud, 200% growth in our cloud subs.

We know cloud subs have the lowest ARPS, that has a weight-averaging effect on the overall subscription plan ARPS.

The second is enterprise, where we don't get a lot of new enterprise sales during the quarter, but we continue to drive usage.

So we continue to drive net monthly active users into the denominator of that ARPS equation.

So as we accumulate more users into the denominator and don't move the numerator, that has a negative effect on the ARPS for enterprise sub, so we actually saw both.

M2S had an impact on product subs and mix had an impact on the overall subscription plan sub ARPS.

That's great.

That's very helpful.

And for my follow-up, maybe for you, Andrew.

I believe you said you're converting about 1 in 4 maintenance subs up for renewal.

Is there any change in the portion of the remaining 75% that's churning off versus, let's say, accepting the price increase, if you will?

Yes, Saket, when we first constructed this program, remember, the whole thing was to balance churn associated with price increases.

We always knew there'd be some incremental churn resulting from increasing prices.

It's very modest.

So what we're seeing is well within the bands of what we expected.

And yes, actually, in that 75%, you do see a little bit of extra churn as a result of the price increase.

But again, that's why we constructed the program the way we did to make sure that we actually minimize the churn.

But right now, it's well within the bands we expected.

Yes, and Saket, you've got the data to do the math, right?

We told you what the maintenance plan subs are and what the sub adds were.

So it was minus 117,000 maintenance sub adds for the quarter and 63,000 of those left maintenance and dropped into converted over to product subscription.

So you can see what the remainder is, and divide that by the base that would have come up for renewal during the quarter.

So it's right in line with expectations.

Which kind of reinforces the point to pay attention to the net adds.

The net adds tell the real story.

You our next question is from Rob Oliver with Baird.

Just on that 25% number, how did that align with sort of your internal expectations?

And then what sort of feedback have you gotten kind of early on from the channel as to what makes people kind of jump?

And I know Andrew, you talked about your primary goal here is kind of ease of use in the product and getting people to understand the implicit value of subscription.

So can you talk a little bit about some of the feedback you're getting, and I just had a very quick follow-up.

So first off, I want to make sure I reiterate exactly what that number is so we're all on the same page.

Remember, in the prepared remarks, we said about 10% of the total people moving from maintenance to subscription or choosing to migrate.

There's a subset of that, that were actually eligible to migrate at all.

So of the subset that weren't already on Suites that default went to collections, of that subset, almost 25% are moving.

That is above what our original expectations were with that base.

So this is a good result.

We view it positively.

Now let's be clear, though, half a quarter does not a trend make, but it's certainly a gratifying result to see.

Now right now, in terms of the value proposition the customers are buying on, I suspect the majority of the conversations are, "Hey, if I get on the collection now, wow, this is the best price I'll ever get for a collection.

I should probably get on that boat right now." And some customers are, quite frankly, in a wait-and-see mode.

How's Autodesk going to increase of value of the subscription offering?

What are they working on?

We've been very transparent with some of our plans.

But I honestly think right now, they're just looking at a strategic choice and saying, "I should go with the collection right now because that's just -- this is the best price I'll ever get."

Great, that's helpful.

And then any update, you guys, to the thoughts around piracy that you gave on last year's Analyst Day?

I think you said 1.3 million actives as you mined through the users and subs?

Any thought on an update to that number or any trends?

So first off, the number's holding.

The trends aren't changing.

We are absolutely increasing our effectiveness in targeting that piracy base.

I think I've told you guys previous, in the past, this year, it's all about increasing the quality of the leads that go into the existing machinery for piracy.

Next year, it's all about automating our communication with the customers that are actually using pirated software and supplementing the existing resources that are focused on piracy.

No change in the trends.

We're piloting the automation that's going to communicate to customers directly in the product with regards to their usage of pirated applications.

That will be rolled out sometime next year.

So we're continuing to head in the right direction there.

Your next question comes from Keith Weiss with Morgan Stanley.

Two questions, one relating to maintenance ARPS, sort of taking the other side of the equation.

You guys actually saw a nice increase there.

Is that related primarily or solely to sort of the price increases that come onboard if they don't take the S part of the [M2S] subscription?

Or are there kind of mix shifts going on there that the perhaps higher price guys tend to stay on maintenance as other guys are attritting?

And is that trend line of the increasing ARPS and maintenance, is that something that we should be looking to continue in our model?

Yes, Keith, it's the flip side of what I was talking about earlier to Saket.

So as people move from maintenance to subscription, let's say they move on July 1, which is the beginning of the third month of the quarter for us, so I'm on maintenance for 2 months, I converted on July 1 in the third month.

The way that gets recorded is I've got 2 months of maintenance revenue and 1 month of product subscription revenue, right, but the entire sub dropped down to the product subscription level.

So the impact -- there's actually impact in both places.

There's, if you will, a tail of revenue left behind in maintenance that has no sub attached to it so that drives up the maintenance ARPS, and then there's a partial quarter of revenue showing up in the product subscription side so you get this understated ARPS.

So the increase that you see in maintenance, the biggest driver of that is the flip side of what we talked about as the headwind on subscription plan.

There is -- there are a couple of other changes.

Pricing of maintenance agreements have gone up, 5% in price, and there's some mix shifts by product and by geo, but the biggest driver is just the compounding effect of maintenance to subscription.

So as we -- we'll be in this mode -- we'll be in this maintenance-to-subscription mode now for the next several quarters.

Certainly, through fiscal '19 and into fiscal '20.

And I think probably it will -- this bucket shifting will continue to make it difficult to look at product -- maintenance sub ARPS versus subscription plan ARPS.

And probably the better way to gauge our overall progress on ARPS is going to be to look at the total ARPS.

That will net out the effect of this tail of revenue left behind and a partial quarter where they landed.

Got it, that's helpful.

And if I could ask one about OpEx.

You guys went through several quarters of basically total expense decline.

This quarter, you saw some slight growth.

Is that an indication that sort of the expense reduction programs have kind of run its course, and from here, it's more about holding steady?

Or is there more room to run in terms of taking actual expense out of the equation here?

Yes, we continue to focus on that.

Now we're flat from '16 to '17, flat again this year, flat again next year.

Obviously, each year, we give a -- basically a merit increase in salary.

So we're continually looking at where we're spending money and using a portfolio management approach to sharpen the things that we have to continue to focus on to be able to live within a flat spend envelope and still drive the transition and drive the most important things.

So I talk a little bit -- when we gave guidance for the year, and you see a little bit of it in the prepared remarks.

We had a couple of divestments.

Those are -- those were expected through the year, but that's an example of the kind of things we're doing as we manage the portfolio to reduce spend in some areas so that we can afford to continue to increase in other areas and still stay flat in aggregate.

Yes, and Keith, one of the questions we always get on this is, "How are you able to continue the execution you're doing when you are looking at an expense controlled environment?" This is really all about picking the most important things to work on.

Obviously, we're doing well in our net adds so we're focusing a lot on that.

And by the way, another area where we focused on is BIM 360 in the construction space.

It's going so well that we actually shifted money during an expense-constrained environment to accelerate future development in that space.

So it's just a matter of picking your priorities.

And over time, you'll expect us -- continue to expect to see us to divest from things that just simply aren't aligned with where we need to focus over the next few years.

Your next question is from Philip Winslow with Wells Fargo.

This is actually Michael Barrett who is on for Phil.

I just wanted to circle back to the maintenance plan decline.

Obviously, 63,000 of those 117,000 directly moved over from the maintenance-to-subscription plan.

Did the -- are the rest of those all just churning off the platform completely?

Was there any shelfware in there?

Are there customers who had moved over prior to the promotion that just simultaneously had a maintenance plan in the subscription plan?

Or should we really think of that as all just turning out of the base?

Yes, Michael, I'll start, and Andrew, if you want to add color, you can.

117,000 minus 63,000, 54,000 churn on a 1.8 million sub base.

Not all 1.8 million came up for renewal during the quarter, but that's a pretty normal churn rate for our maintenance plan.

So we really saw nothing extraordinary even in the face of an announced 5% price increase.

We really saw nothing extraordinary in the churn rate within maintenance.

So there's nothing -- there's no other moving parts under the covers, if that's the question you're trying to get at.

Great.

Yes, that was what I was trying to clarify.

And then just on the move to collections, what sort of impact to subs, if any, have you seen from that move?

I know you've called out, I believe, at your Analyst Day in the past if you have a customer who has 2 subscriptions per subscriber and they move to a collection, that's minus 1 subscription even though it's ARR accretive.

Has there been any impact on total subs?

Or is it too early to tell at this point?

Well, we have lots of data from the Suites days that the effect -- this consolidation effect kind of ranged from 2% to 5%.

If you look at what's going on with the maintenance-to-subscription moves and some of the churn we're seeing as people move from maintenance to subscription, it's right in line with those exact same numbers.

We see consolidation, but it's incredibly modest.

You have to remember, people need to use the software.

They need the software for themselves individually, so when you consolidate products, yes, you do see some modest consolidation.

It's right in line with the kind of consolidation we saw with Suites.

Your next question comes from Matt Hedberg with RBC Capital Markets.

I believe you guys implemented a file format change at the end of March.

Can you talk about the mechanics of this and, if any, impact on quarterly sub adds?

And I guess, secondarily, you've done these in the past.

Can you sort of remind us historically the impact of something like this converting non-maintenance-paying active users?

Yes.

So first off, just to give you context, every 3 to 5 years, we actually do one of these file format changes because what happens is we start to accumulate technical debt inside the file format, and we have to do these things to prevent bloat.

So yes, we did one of those in this round.

What you generally see is an initial decline in the number of active legacy seats, the non-subscriber seats.

We saw that.

But then it takes a year or 2 to diffuse through the entire ecosystem.

So what you get is an initial chunk associated with the release of the product where legacy users shift up to the new product, and then over time, as people are integrating with the ecosystem, you see more and more attrition out of this non-subscriber base.

So we expect to see the exact same pattern.

It's happened historically over and over again.

So we're right on track for that right now.

That's helpful.

And then you've commented in the past, coming -- once these 3-year M2S conversions come up, there's going to be a price increase that's sort of, I think -- if it's well documented, at least you've talked about it.

Is this a once -- like a 1-renewal cycle trend to kind of get back to sort of normal discount rates or might this be a couple cycles?

I'm just sort of think trying to think about the magnitude of that price increase.

Do you want to take this?

Yes, Matt, we've also tried to articulate that.

The -- when we get to the end of the 3-year period, so someone converts from a maintenance subscriber to a product subscription, they give up their perpetual license.

They get this reduced price for product subscription, and without having to pay for all 3 years upfront, they're grandfathered at that price for 3 years.

At the end of that -- and that's for people who convert this year and next year, fiscal '18 and '19.

At the end of that 3 years, they revert to what we've been calling the terminal price, which is the price of conversion compounded over 3%, 5% increases.

So in effect, a 16% price increase from where we started the entire process.

So they will snap up to that level.

So they took a 5% increase, if they convert this year, that's roughly a 10% increase at the end of their 3-year grandfathering.

And then it's going to take time for that [because] that will create a pool of product subscribers that are at a significantly lower price than a new product subscription, and it's going to take time to migrate that pool up to the price of a net new product subscription.

That's not going to happen in 1 year or 2 years.

It's -- that would be too big a step.

Your next question comes from Heather Bellini with Goldman Sachs.

This is Mark Grant on for Heather.

Saw the disclosures around the M2S program with the subscription plan ARR and the net subs coming from that program.

One, are those metrics that you plan to disclose on an ongoing basis?

And then just second, on maintenance renewals, realizing that, that M2S program might create some noise in the seasonality of renewals this year.

Can you give us a sense of the split of contracts coming up for renewal between the fiscal 3Q and fiscal 4Q?

Yes, Mark, to your first question, we will continue to provide that visibility.

I think because it's fairly disruptive between the 2 big pools of subscription plan and maintenance plan, we'll continue to provide the same metrics that we gave this quarter.

So the impact on where they landed in product subscription in terms of how much revenue accumulated there and the number of subscriptions that moved.

I still think it's going to be difficult for you to sort out at a more granular level what's happening inside ARPS because those transitions take place during the quarter.

And so the -- given -- if you go back to the example I talked about earlier, if they move at the beginning of the third quarter, there's a tail of revenue left behind on the maintenance side that will not move over.

And so it's -- I would encourage you to try to model ARPS.

If that's important to you, I'd encourage you to try to model it at the total level.

In terms of the number of maintenance agreements that come due in the second half of the year, we haven't given that kind of seasonality metric historically, but what you see is as with most enterprise software companies, we do see a bigger second half than we do in the first half.

And so you can expect to see it be slightly more opportunities for conversion in the second half of the year than what we would have had in the first half.

Your next question is from Sterling Auty with JPMorgan.

I apologize if you covered this in the prepared remarks and I missed it.

But of the users that chose to make the move over to subscription from maintenance, can you give us a sense of what products were most popular that actually chose that switch?

Well, so, first off, if they're on a Suite, it's kind of a no-brainer for you to do this.

So we saw a lot of people on Suites making the move to subscription and taking up the offer pretty quickly.

And then it kind of follows from there to products like AutoCAD, naturally, is the next biggest product up there.

So it's pretty straightforward.

Okay.

And then Scott, on to a different area.

On the OpEx side, I know you guys -- your hedging program, but just want to make sure, how, with the recent moves that we've seen included with your hedging program, is that impacting the outlook here for the back half?

Yes, it's a great question.

The interesting part about FX and particularly a subscription model where a lot of the revenue that we report in a given quarter was actually sold in a prior quarter, right?

So as we open the quarter, we've got a very high percentage of the reported revenues for Q3 sitting on the balance sheet at exchange rates that are commensurate with when those were actually sold.

So there's a lag effect in revenue really independent of the hedging.

Whenever it's sold, we have the hedge benefit but then it goes in deferred, so the benefit of the weaker dollar will take a lot longer to show up in revenue simply because a lot of reported revenue's coming off the balance sheet before the FX rate is changed.

We'll see it much more immediately on the expense side.

We hedge the net exposure but we'll see the impact of the FX hit much more immediately in -- on the spend, both COGS and OpEx side.

Your next question comes from Gregg Moskowitz with Cowen and Company.

So the cloud subscription growth again was very impressive, and it sounds like you're really starting to see BIM 360 construction and related activity ramp up as well.

On the other hand, I believe the comps on cloud subs make it a little tougher later this fiscal year given the acceleration that we saw in that year-ago period.

So just kind of curious how we should generally think about cloud subs growth over the balance of the year.

You're right.

In BIM 360, we're seeing a beautiful -- a really good acceleration of clouds growth.

We see nothing that would indicate that, that's going to slow down as we head throughout the year.

And the only thing, if I was to highlight anything, as you grow the base larger, the base you have to renew gets larger as well.

So we'll be watching that very closely.

But the momentum, there's really nothing out there right now that would slow the momentum, particularly on BIM 360.

Okay, perfect.

And then with respect to EBAs, I know Q2 obviously not a seasonally strong quarter for you there, but your unbilled deferreds were up significantly again on a sequential basis.

How is the EBA uptakes tracking relative to your expectations, both in terms of number of customers signing on as well as the degree of ARR spending increases?

Yes, Gregg.

The EBAs are tracking right in line.

EBAs are always heavily weighted into the second half of the year.

And so if you go all the way back to our Investor Day last year when I made sure everyone understood that we were going to move most -- many of our EBAs from billed upfront to annual, I said I thought the impact on deferred revenue by the end of the year would be $300 million or more of unbilled deferred that would accrue, and that's still our expectation on the unbilled deferred by year end.

So expect to see the unbilled deferred grow more heavily in the second half of the year as we sign more of those EBAs.

But in general, the EBA business is tracking right in line with our expectation.

Your next question comes from Ken Wong with Citigroup.

When looking at your sub guide, I guess the midpoint is around 650, and you guys are already kind of halfway there.

And correct me if I'm wrong, but usually in the second half, you add a lot more.

Can you help us understand some of the dynamics that are going into your guidance and the thinking for the second half?

Yes, Ken.

We shifted the entire range up by 25,000.

So prior guide on subs was 600 to 650.

We shifted it up to 625 to 675.

And a lot of that's based on the strength that we've seen in the first half of the year.

We've touched a little bit on this in the opening commentary.

It was a very strong first half.

Our net sub adds coming from the enterprise business outpaced our expectations in Q1.

The legacy promo continued to be strong, and cloud in the quarter we just closed, our cloud sub adds grew 200% year-on-year as Andrew said.

So we had a very strong sub add first half of the year.

The second half of the year is right in line with our expectations.

So it's not a -- there's nothing -- there's no difference in what we expected in the second half built into that guidance.

What you see in the guidance range shifting up is the strength of first half.

Got it.

And then maybe in terms of just renewal rates.

I know in the past few quarters kind of across the different subscription lines, maintenance, product, et cetera, you guys have seen always kind of close to peak, if not peak.

How are those tracking in Q2 here?

Our renewal rates are tracking exactly the place -- the way we would expect them to track right now.

So we're right in line with everything -- all of our expectations.

And Ken, the place to look to get a feel for that is, again, look at the net sub adds.

And we've given you the visibility of maintenance, and you can get a real sense of what the renewal rates look like there because we told you how much of the 117,000 that went away were conversions over to product subs.

You can do the same math on subscription plan and look at what the net adds are there.

And that gives you a sense of not just what the volume is -- or the combination of what the new volume is and what the renewal rates are.

And so -- and back to your earlier question, at the midpoint of our guidance by the way, 625 to 675, so the midpoint is 650.

That's 21% growth year-on-year.

So it's right in line with our expectations.

Your next question comes from Ken Talanian with Evercore ISI.

So you mentioned earlier that you had reinforced conviction in the fiscal '20 targets.

Given the better-than-expected results, you think it's reasonable to assume that there's upside to those targets?

Ken, so I'll start, and I see Andrew smiling.

I know he can't wait to add some color to this.

One of the things that we've done -- those targets, of course, we put out a little more than a year ago at this point, and the goal in setting that out was not to say, "This is guidance, and we're going to update it every quarter." It was to say, "This is where we see ourselves headed as we work through this transition." One of the things we spent a lot of time on back in December at our last Investor Day was talking about the various steps we had taken, the leverage we have pulled to, in effect, derisk that set of targets, and to provide some buffer between where we think we're going to land and what those targets are, and we continue to do that.

We continue to pull those levers and continue to drive for those targets.

I don't think it would be wise for us to try to continually update what that number looks like.

It's good to see you asking questions like this.

I mean, I think right now, all of us being on the same page that it's looking more and more likely that we're heading in exactly the right direction for our FY '20 targets, I think, is a good place to be.

There's no reason for us to revise those targets right now.

Okay.

And I guess just a more specific question.

Can you talk about the mix of products that you're seeing transact through the eStore?

Yes, when you look at the eStore, the eStore really dominates at the low end of our offerings.

We actually do see almost every type of -- product type that we sell going through the eStore, but if you look at what the majority of it, a lot of it's going to be LT and then the next largest chunk is going to be AutoCAD.

And that's basically the majority of the eStore traffic, and that's totally what you would expect from e-commerce transactions, the lower end of our value spectrum.

Great.

Just one quick follow-up to that one.

Do you know if you're seeing any pirates essentially transact through the eStore?

Well, you know the pirates don't announce themselves when they arrive at the eStore.

However, it's highly likely that some of them do show up there, all right, especially if they're buying LT or something like that.

But like I said, pirates don't declare themselves, but it's probably likely, especially given some of the digital-related targeting efforts we're doing with piracy that some of them show up buying there.

Your next question is from Shankar Subramanian with Bank of America Merrill Lynch.

I'm asking a question on behalf of Kash.

I have a quick question on what can be the churn rate of the current renewal base for the subscribers for next year.

So the question I have is -- so most -- some of the projects that the construction engineers use would be a short-term project, but some can be a long-term project.

So if the subscriber growth in the current year is driven by short-term projects, there's a likelihood the next year they might drop out.

But can you talk about what the mix is?

I don't know if you know this or not, but can you talk about the mix between subscribers who are using it for short-term basis versus a long-term basis?

So the pattern you're talking about, if no projects were in the pipeline at all, then I'd probably worry about that.

But for every project that drops off, another project shows up.

So that's just the wrong way of looking at it.

There's a steady state of projects within the ecosystem.

Projects retire, new projects ramp up.

So I -- there's really no material change in the run rate because of the cyclicality of a particular set of projects.

There's always new projects in the pipeline.

Got it.

And one on the digital manufacturing strategy, and you talked about the focus area for that.

Can you talk about, is there any particular segment you're focusing on, whether it's 3D printing, IoT or not?

And where are you in that strategy implementation, and when do we see the benefit of that?

Is it more beyond fiscal '19?

So the opportunity that we're dealing with right now, where the market is fully ready and the technology is fully aligned with the market rate, is in construction.

When we look at the digital manufacturing opportunity, the market is still coming to terms with, "Okay, what does it mean for us to be using solutions in the cloud when I'm in product manufacturing?

How real -- how much of the 3D printing capability is actually going to be practical for my applications?" So what you're seeing right now is in low-volume manufacturing, aircraft components, things like that, greater adoption of 3D-printed workflows and some of these more automated workflows.

That's going to eventually move down, so the bet on digital manufacturing is like the next long-term bet after the construction bet that we're playing right now.

So were actually -- what you're seeing is a series of bets that provide a very strong path for long-term growth for the company.

And we have reached our allotted time for questions, and I would like to turn the call back over to David Gennarelli.

Great.

Well, that concludes our call today.

We'll be at the Citi Conference in New York City on September 6.

For any other follow-on questions, you can reach me at (415) 507-6033.

Thank you.

Thank you for your participation.

This does conclude today's conference call, and you may now disconnect.