Good morning, ladies and gentlemen, and welcome to PTC's Third Quarter Fiscal Year 2014 Results Conference Call. [Operator Instructions] As a reminder, ladies and gentlemen, this conference is being recorded. I would now like to introduce James Hillier, PTC's Vice President of Investor Relations. Please go ahead.

Thank you, Gail. Good morning, everyone, and thank you for joining us on today's third quarter fiscal 2014 earnings call. As a reminder, today's call and Q&A session may include forward-looking statements regarding PTC's products, our anticipated future operations or financial performance. Any such statements will be based on the current assumption of PTC's management and are subject to risks and uncertainties that could cause actual events and results to differ materially. Information concerning these risks and uncertainties is contained in PTC's most recent Form 10-K and 10-Q on file with the SEC. Unless otherwise indicated, all financial measures in today's call are non-GAAP financial measures. A reconciliation between the non-GAAP measures and the comparable GAAP measures is located in the Q3 2014 press release and prepared remarks documents on the Investor Relations page of our website at www.ptc.com. With us on the call this morning is Jim Heppelmann, our President and CEO; Jeff Glidden, our CFO; and Barry Cohen, EVP of Strategy. With that, I'll turn the call over to Jim.

Thank you, James Hillier. And good morning to all of you who are investing your time here this morning to join us for our third quarter of fiscal 2014 earnings call. The past 90 days have been very exciting for us and I'm pleased both with the strategic development and with the business results that we announced yesterday evening. On the strategic side, we're very pleased to announce the Axeda acquisition. With 1 foot in the Internet of Things or IoT world, and the other foot in service lifecycle management or SLM world, Axeda is a great complement to both ThingWorx and PTC. Axeda is a leading brand today in the IoT market and an almost perfect fit to PTC in terms of what they contribute as compared to what we already have. When we factor in the scale that we will gain from Axeda in terms of technology, employees and IoT expertise, customers, partners and revenue, we believe that this acquisition will reinforce PTC's clear leadership position in this dynamic new world of smart connected products and the Internet of Things. The Axeda deal will be synergistic for us on multiple levels. Because we don't really have a pre-established fiscal year '15 financial baseline to compare against, it's hard to say in a meaningful way how many pennies of accretion it might add. But I can tell you, first, that there's a meaningful cross-sell opportunity because our research indicates that Axeda customers generally love ThingWorx and vice versa. On the cost side, the synergies are significant in the near term. In the case of the Axeda acquisition, it's about cost avoidance rather than cost-cutting, but either way, we accomplished the same results. To clarify this point, when we acquired ThingWorx 6 months ago, we had a roughly 40 person…

Thanks, Jim. As Jim cited, over the last 90 days, we've made great strides strategically. In addition, we delivered strong financial performance in Q3. Total revenue increased 7% to $337 million and non-GAAP EPS increased 19% to $0.53. License revenue was up 16%, our support business increased 7% while our service revenue declined slightly. As a result of this favorable revenue mix shift, we delivered gross margins of 75% and we expanded our operating margins to 24.2%. On a geographic basis, our business in Japan was very strong with revenue increasing 16%. You'll recall that as we closed out Q2, we have cited that some business in Japan have slipped from Q2 but that we were confident that these deals will close in Q3. And I'm pleased to say that our team in Japan clearly delivered on that promise. Our European business increased 13%, year-over-year, while our revenue in the Americas and Pac Rim were flat with the prior year. The highlight in the quarter was our CAD business with revenue increasing 10% year-over-year. As discussed previously, we are in the midst of a new product cycle, now with more than 65% of our customer base upgraded to Creo, and we're seeing customers purchase both new Creo modules, as well as additional seat capacity. We continue to have very good cash collections from our customers and generated $106 million in cash flow from operations, up 26% year-over-year. Cash increased to $304 million and we purchased $60 million worth of PTC stock during Q3. Looking ahead to our outlook for Q4 and the full year, we expect an improving economy and outlook in the U.S. and Europe but continued softness in Asia Pacific. We expect Q4 revenues to be in the range of $340 million to $355 million and we expect to deliver non-GAAP EPS of $0.59 to $0.63 per share. For the fiscal year, we expect revenue to be in the range of $1.330 billion to $1.345 billion and expect our non-GAAP EPS to be in the range of $2.10 to $2.14. Again, we appreciate you joining us today and now, I'll turn the call back over to James Hillier.

Thank you, Jeff. Gail, you can now open up the call to questions, please.

[Operator Instructions] Our first question is from Mr. Jay Vleeschhouwer of Griffin Securities.

Jim, Jeff, as part of giving the new metrics that you alluded to in connection with the subscriptions, would you expect to give a billings guidance or bookings guidance as one of your competitors in the industry has begun to do? And secondly, with respect to managed services, could you talk about how widely adopted that is, thus far, in terms of Windchill customers making use of that and is it the sort of offering that you could offer beyond just Windchill into any other functional areas that you have? Then I'll follow-up.

Okay. So let me say, first on the billings guidance. We're actually, Jay, in the process of, right now, trying to figure out our own answer to that question. So I don't want to preempt with a guess here. But what I will tell you is I think PTC has always had high marks for transparency, so we're going to try to find the right set of metrics that we feel like provide sufficient transparency to our businesses operating but I don't want to get into the detailed answer for fear that I might give you one that 90 days from now we change. But I can say, we'll be appropriately transparent. Then on the managed services piece, that business, right now, has relatively low adoption across our entire customer base. Perhaps, higher adoption in some pockets, maybe higher adoption in new business, higher adoption in medium-sized companies but if you were to look at our entire customer base for PLM, if we had high adoption that business will be many times larger than it is right now. But that's it. It's a nice business and it's got a nice growth rate. We have some managed services business also in SLM, particularly the Servigistics piece. When we acquired Servigistics, they had a managed services business. So if you look at our full managed service business, it's probably 60% PLM and 40% SLM right now, and again on the SLM side as well, we think that there's quite a bit of growth opportunity, though as customers purchase the Servigistics software there is a higher rate of adoption of managed services on the initial purchase. I think that's just consistent with where SaaS and managed services in general has the most traction. It tends to be a little more traction, let's say, in back office systems, in of course CRM being the high point.

Right. Just a follow-up on your fiscal '15 preliminary comments and then an IoT question. On '15, are you at this point able to say anything about your assumptions regarding, particularly to the CAD and PLM businesses, the relative performance you're anticipating in terms of new volume versus ARPU for either side of the business? And then on IoT having now filled out the development platform side and now the connectivity and security side, is there any important missing piece for IoT or do you think you've got the portfolio now?

Yes. Jeff, can I take the second part of this and then I'll have you...

And I'll take the first part.

To the extent you can.

Yes.

On the IoT piece, Jay, we do think we have a pretty interesting stack of technology right now. Like you said, from connectivity, which gets the data from the machine to the cloud in the application platform, which allows you to build applications that process that data in the cloud to the SLM and ALM and PLM suite, which are essentially predeveloped applications that can use and process that data, as well. We feel like we have a good stack. I think the place we could do a lot more would be in big data analytics. Because a lot of people say by running big data analytics against that data you can see things, patterns developing that you never would have necessarily been looking for, that some of these patterns are almost organic and by analyzing data and looking for these patterns, you can then begin to predict things happening as opposed to just track them as they are. So I think that's a place where you could see us continuing to build partnerships, maybe do acquisitions, I don't know. But certainly, that's a place where we could add more to the stack.

Okay, Jay, relative to 2015, what we wanted to do today was really give broad color on what we see in the overall business. We're in the midst of really assembling that plan. We do it by region, by customer, look at major campaigns both by PLM, CAD, SLM, et cetera. So I'm going to basically say we'll defer that full discussion until we release Q4 and I think we can give you good color then, but suffice to say, I think we feel very good about where we are, just completing acquisition of both Atego and now, Axeda. Those will factor in as we build that plan. So I think it will be premature to give you much more than to say we feel pretty good about the business space we're in, and some of the new growth markets. So we'll come back and give you more on that discussion when we close Q4.

Our next question is from Matt Hedberg of RBC Capital Markets.

Very nice to see continued growth in CAD, it also sounds like Axeda will be a very nice addition to the PTC family. I guess I'm wondering, in terms of the quarter, you pointed out some Q2 slipped deals from Japan closed this quarter. I was curious, was there also some 4Q license deals that got pulled into 3Q?

Yes, I mean, a couple of these deals that we took in a subscription fashion. By going subscription, we sort of lubricated the deal process and then move more quickly because we weren't necessarily requiring a significant capital expenditure upfront. So I definitely think if you look at our Q4 and maybe if I go to probably the underlying question, there's a couple of factors that have affected our license outlook for Q4. Let me just get through them. First thing is China. We're nervous there because normally we have a big Q4 in China. In fact, last year, we had a $10 million license jump from Q3 to Q4. To be frank, we don't see that happening because there's just some issues in China that we and other tech companies, particularly U.S. tech companies, are suffering through. There's a fair amount of turmoil in the government agencies and that slows things down and then there's sort of an anti-American technology sentiment, which also slows things down. So we're taking a pretty conservative view of what's going to happen in China and that takes some momentum away. The second thing is this subscription model. I can tell you, for example, one of the biggest deals that we have in the Q4 pipeline looks like it might head down the subscription path, which for all intents and purposes, will give us a goose egg as it relates to Q4 license revenue. And then the third factor is this timing thing. We feel like we're having a pretty good license year and when the dust settles, we will have had a pretty good license year. Q3 might have a been a little stronger than we expected. Maybe some of that is revenue that might have landed in Q4. But we feel like the license business is in good shape. We've got this issue in China that we hope improves but we don't feel like there's a real problem here that we're seriously worried about or anything.

That's great, very, very helpful. And also good to see the progress in Europe and Japan certainly coming off of Q2. I'm wondering, though, the Americas is a little weaker than we expected. It looks like more in the Extended PLM and SLM business. I wanted to double-click into that a little bit more, kind of, find out if that was more of a mix shift or sort of what was going on in the U.S. market?

No, I think it's really more a question of a lumpy business. I think we're thinking we're going to have a pretty strong Q4 in North America and all's well that ends well and the story in North America will end well.

Our next question is from Mr. Sterling Auty of JPMorgan. Sterling P. Auty - JP Morgan Chase & Co, Research Division: I wanted to drill into this idea of subscription. So you've mentioned it in answers to a couple of questions, including the guide here for the fourth quarter. We've watched, over the years, a number of companies from Cadence Designs to Aspen Technology to Ariba to, more recently, Adobe and Autodesk, go ahead and take the full plunge on subscription. It seems like you're starting to dip your toe in the water more and more into that subscription model, why not just jump in with both feet?

Jeff, do you want to take that one?

So Sterling, I'll just say, look, the new businesses are coming whether we acquire them or we build them, more and more flexibility for the customers with subscription models. So I think that is clear, significant and increasing portion of our business, new businesses will be subscription. I would remind you that today, 51% of our business of support is really a subscription model already. So that's very significant and I think as Jim said, we provide and we have provided in the past flexibility for both perpetual in term or a subscription type model with customers and we've offered that and we've seen more take it up. So I think it's a flexibility option for us but I want to reiterate that, today, the majority of our business is already in a subscription model and we'll work within the framework. We'll give you a better color on this as we get into -- and transparency, I think the metrics as we report for '15 become very important to look at bookings, billings and deferred revenues. So we'll give you more color on that as we get there.

Yes. Maybe, Sterling, I could a little more. And that is Jeff kind of hinted at this but I'll just go a bit deeper. We're acquiring companies that are subscription based and we either have to then embrace subscription or we need to try to convert those companies to perpetual. I think when we gave you guidance for what ThingWorx would contribute to the year, we actually assumed some amount of that would go perpetual because we thought our sales guys would drive it that way. But the customers haven't really followed. So we're looking at strong bookings but they're almost all subscription and therefore, we're not seeing near as much of these good bookings for [ph] the revenue in the near term as we might have thought they would. But we've made the strategic decision not to fight that but to embrace it because let's be frank, it would be silly for us as management to try to defend the castle of perpetual license model. So we're not going to do that. The real question is how aggressively do we embrace it in the core business? You should just assume that all this new business, this new IoT stuff will be subscription. The real question is what to do with core CAD, PLM and SLM. And as Jeff said, that's probably moving a little faster in that direction that we might have thought. And so in 90 days, we're going to come back to you with better strategy, better metrics and so forth, so that we can really tell you what we think is going to happen and then show you as bookings come in, how many of them are going perpetual, how many of them are going subscription. So you can see the full picture because today, in our earnings release, we can only paint a part of the picture and so we need to change the way that we communicate with you and we're committed to doing that in Q1 of '15. Sterling P. Auty - JP Morgan Chase & Co, Research Division: And maybe, just as a follow-up, if you can give us a little bit of qualitative insight now and maybe more quantitative at that point insight, a lot of times you see people get confused when you talk about subscription. The SaaS or hosted type of offerings or managed service type of offerings that you've got versus, to your point which you mentioned in kind of the first part of the answer, the traditional kind of just term contracts. So we can understand it, how much is going one way versus the other and what the margin impact of 1 choice over the other would look like on the business.

Yes. So the qualitative answer would be there's substantially more interest in subscription contracts than there is in hosted software in our world. For example, people might want to buy CAD in a subscription license model but they don't want to run it to the cloud, necessarily. So I think we do have managed service business, so we're covering that angle as well and that is a nice business and it's got a decent growth rate. But you know that if a megadeal flips from perpetual to subscription, it has a pretty material impact on the given quarter we closed that deal in. And today, we can't really show that to you in our reporting metrics. So we want to make sure we had a set of metrics that we can explain that, really quite frankly, to show you all the good news that today we have a difficult way of communicating to you. Sterling P. Auty - JP Morgan Chase & Co, Research Division: No, I think that's great. And last item to that is what is the lifetime value of that subscription type of contract versus perpetual? Are you getting the uplift that you normally think about when you have a multiyear subscription?

Yes, definitely. I mean, the data we have so far suggest we do much better contracts when they're in a subscription model than in a perpetual. Customers are willing to pay more because they don't have to commit to it on day 1. They don't necessarily have to pay it all on day 1. They don't have to take as much risk that they bought the wrong stuff and so forth. So we've seen that customers are -- they just negotiate with a different attitude when you're talking about subscription and they're definitely willing to pay a premium.

Our next question is from Mr. Steve Koenig of Wedbush Securities.

I have 1 for you and then 1 follow-up. So Jim, maybe qualitatively you could help us understand how would you describe the risk since you're still thinking about how fast that move to subscriptions in the core business might go? And it's not an all or nothing affair but you're obviously working on trying to size that. How would you describe the risk that, that shift could impact your preliminary fiscal '15 guidance?

Well, okay, first of all, that's a good question. I think that we thought a little bit about this in our FY '15 guidance and the FY '15 guidance sort of is a predicated on a steady-as-she-goes model, which means that the new businesses are subscription and that the old businesses are subscription, let's say, on an exception basis. So that's sort of the attitude that's factored into that guidance. Not that we would get super-aggressive and try to push the old businesses to subscription or something like that but that here and there may be a deal flips over. So I think if we were to get more aggressive, then we would want to think about implications of that. I'll tell you what we really want to do, though, is provide a way for you to sort of bridge it back and forth. That's going to be our goal as we go into next year is to give you the full picture in a bridge to help you understand just how much business did we land and how did we then recognize it, so that you get the full picture and feel good about it, sort of -- we want you to feel good about us winning business independent of which model we use to recognize it basically.

That's helpful, Jim. And then if I may ask 1 follow-up, it would be the Creo product cycles is showing some good trends right now, I'm wondering how long are the legs on that? How long can it keep boosting your revenues beyond kind of market growth here because -- the Creo product cycle, does that go into next year? do we see continuing strength or did the -- could the comps get harder, is that -- should we now expect that? Just any preliminary thoughts there would be interesting.

Yes, I mean, I think if you look at it this way, Steve, if you took the license guidance we gave you for next year and you kind of back out a reasonable number for Axeda and a reasonable number for Atego and then back out the implications of a fast-growing ThingWorx business, you'd be left with a, sort of at best, mid-single-digit organic growth number. So then if you say, okay, that's coming from CAD, PLM, SLM, I think that's a fairly conservative view of those businesses, which means we're not assuming we're going to have a 10% growth year in CAD next year. That's not to say we're not hopeful we will but we're applying some conservatism as we kind of take our aspirations and turn them into guidance. So I think we feel like the CAD business is doing really great this year and our assumptions for next year would be that it's good but not clipping along at the same level. Again, that's just the process of trying to figure out how to guide. That's not to say that we don't hope and won't push for a year next year that looks like the year we're having this year.

Our next question is from Mr. Matt Williams of Evercore.

Just had a quick one on the sort of the SLM, obviously, flat license year-over-year the subscription component probably played a part in that. So I'm just wondering, are you seeing anything in that business that -- are deal cycles elongating at all as people start to contemplate rolling in ThingWorx and that type of thing or is it really just a matter of the subscription component of Servigistics, maybe coming through a little bit stronger than you anticipated?

Yes. So first of all, I'm not going to pin that one on subscription. That's not really the issue there. I think if you look at SLM, there's 2 factors that maybe are variables in this discussion. One is that it's a lumpy business and the Servigistics side of it always has been. The second thing is sort of a bookings pipeline rebuilding process. We acquired a company that had a pipeline. We did a great job of converting that pipeline. In the meantime, we went through a process of teaching the PTC sales force how to sell that product. That process took a while. They then began building pipeline. We have quite a promising pipeline for SLM now but you might think that we went through a bit of a valley where we transition to selling from the small Servigistics sales force who had a smaller pipeline and a high close rate to the PTC sales force, which then had to build a bigger pipeline. So I feel like we've gone through a slow spot but the outlook for that business is pretty promising and had some bigger deals come in, maybe you would never have seen this but it is what it is and we feel good about the business going forward.

Okay, great. That's helpful. And maybe just 1 quick follow-up. You've got ThingWorx as a standalone offering and also down the road integrating with some of your core offerings. You've got Axeda now that will sort of match up with ThingWorx and the SLM roadmap. How should we think about just integrating all of these different components and what sort of time frame are you at least initially targeting to try and really get things into less of a standalone sale and more of, I guess, an integrated offering?

Yes, so that's a great question, Matt. Let me say, first of all, there are a number of Axeda customers who have already purchased ThingWorx and use the 2 together. So those 2, I think we can make it better but there's a workable solution for how do you use ThingWorx to build applications against data you gathered with Axeda. There's a number of customers who've already gone down that route because they love Axeda but when it comes to building applications, ThingWorx is so much better, okay. So I feel like that's a solution that's ready to go right now and we can begin cross-selling and so forth. What was going to take longer would be to connect that stuff back to our traditional suites in the SLM and PLM and ALM world but one of the nice bonuses of Axeda is they actually have some SLM and ALM connected software. For example, the idea of remote service so that you can log into a remote machine, product if you will, and diagnose what's going on and tweak some parameters and so forth. They already have an app that does that, that's quite rather used by their customer base. And then if I switch to ALM, how could I access a remote smart connected product and apply a software patch or change a software configuration or download a whole new version? They already have an app that does that pretty well. So I think that while we thought of Axeda initially as this connectivity, we actually found out that they have some very interesting SLM and ALM apps that will help us more quickly bridge from the world of IoT to the world of SLM and ALM and in my mind, that was a big pleasant surprise as we got in this acquisition. And I would just say, in closing, on that side, this Axeda acquisition, the more we looked at it the more excited we got. It was better and better strategic fit, more and more synergies. I really think, particularly coupled with ThingWorx, but even independently it's going to be one of the best deals we've done in a long time.

[Operator Instructions] Our next question is from Mr. Saket Kalia of Barclays.

It's Saket from Barclays. A few, if I may. Jim, you gave some helpful commentary on the trim in fourth quarter guidance. Looks like guidance went down by about $10 million adjusting for Atego. Is there a way to sort of break out that impact from China versus subscription versus maybe some of the intentional services off-loading that you're doing?

Yes. Let's see, if you want to work off a $10 million number, then probably 1/2 of it is China. If you're talking about all revenue, services is, quite frankly, a big piece of the balance. If you're just talking about license revenue, a picture might be a little differently but I think you asked about $10 million at the total revenue line and it's almost 1/2 China, 1/2 services would explain that.

Great, that's helpful. And then if you were to sort of look at it for -- if you're asked that same question on 2015 from a higher-level, you've got a couple of headwinds from subscription and services, is there a way to sort of think about the impact to growth to that low- to mid-single digit kind of revenue growth? How much it really would have been normalizing for those 2 factors?

Yes, I mean, I think, for fiscal '10, we probably collectively lost $10 million of mostly license revenue out of China. So I can quantify that one.

Fiscal.

Fiscal '14, I'm saying. I'm looking backwards, not forward here. I think if we look forward, we have a pretty conservative view of what's going to happen in China next year, sort of the same effect versus let's say an undisturbed business. And we hope that situation in China picks up because what we have is a situation where customers aren't buying anything from anybody and we're hopeful at some point in time, they'll go back to buying something from somebody and we hope we're that somebody. Meaning, there's some demand perhaps queuing up here, hopefully. Yes, I think on services, probably...

Flat on services.

Right.

So just continuing on Jim's earlier comment, we're looking at double-digit license growth, single-digit support growth and flat services and that's really fundamental to both the strategy and to the financial model to drive higher gross margins through that mix shift, as well as the operating margin. So I think we feel pretty good, particularly, we had a very nice license growth this quarter and driving that into the, consistently, into the double digits will be important. Now a piece of that is acquired and a piece will be organic but I think that's really the color we can give you on '15 today.

And then lastly if I could squeeze it in, just on subscription, for the contracts that you've done so far, can you just talk about the cash collections on the subscription contract versus perpetual? Are customers paying up front and what sort of the average duration, maybe a year or more? Any color there will be helpful.

Sure. So they generally range -- they're typically a minimum of 1 year, more likely to be 2 or 3 years. The billing cycle is typically that if you take, let's just take a 2 year-subscription, you can bill the first year upfront, that will be collected in normal terms at the end of that year. So you booked the total amount to 2 years. You bill 1 year, collect that within normal terms of, let's call it, 60 days. Then at the end of that first year or probably before the end of the first year, you bill it again and collect it. So I think it's an important discussion on the subscription side because I think the cash flow is very attractive and over time, we actually build greater value and greater cash flow from these kinds of transactions. And I would add that some of the monetization on these is not about seats but it's about devices or assets connected and so when you think about it, we may have a platform transaction that has, let's call it, $25,000 to $50,000 upfront but that may yield over time, go from hundreds of assets to thousands or tens of thousands, that monetization builds very, very nicely in year 2, 3 or 4. And so I think there's some significant upside on the way we monetize these, as well.

Our next question is from Ann Grackin of ChainLink Research.

So I have a twofold question. First, Jim, you we're talking about talking to a lot of CEOs about the Internet of Things. So my first question is about what do you see -- excuse me, from those conversations that point to new emerging business models for those companies and do you also see new kinds of startup companies emerging out of the whole concept of smart connected products and the Internet?

Yes. So Ann, that's great question. So definitely, when people begin to think about smart connected products, you quickly get to the point where you start to think about the business model. And there's this thing we called a service paradox. Let me just step you through that. What happens today is there's a lot of companies who make money because of an inefficient service process. The fact that you've got to make a lot of service calls and that products consume a lot of spare parts because things break, that's kind of nice if you sell service contracts and spare parts and technician hours and so forth. But if you could monitor those products remotely and change from a reactive to a proactive model and keep things from breaking and failing and so forth, the customer is pretty happy. But suddenly, you're not selling so many spare parts and you're not doing so many service calls and so forth and your service revenue is under pressure. So then people say, well, what happened is I've created an efficiency here that's being totally accrued to the customer. And I need to figure out a way to acquire some of that efficiency back to my side of the fence, to my side of the relationship and the way I could do that is by structuring the business relationship differently. Rather than selling you the asset and then selling you the parts and the technician hours, why don't I keep the asset and the parts and the technician hours and sell you use of the asset? And that leads to the so-called product as a service model. So I think that's popping up all over the place, just hundreds of examples if we sit here and brainstorm through them. And CEOs are…

Yes, like the military had performance-based logistics or powered by the hour type of contracts with some of their suppliers. So my second question is about the revenue model. You had touched on it briefly in the answer to the previous questions. So is the revenue model a base plus fees for devices or information? How are you going to charge as the user goes from 50 devices to 1,000 devices, for example?

Yes, both ThingWorx and as well Axeda have pricing models that may include some upfront cost but then scale, typically, according to the number of connected devices. So if you double the number of connected devices, you have to pay a little more. And if you double it again, you have to pay more again. So it tends to scale with the number of devices. It may also scale with the number of users. So the more users, the more devices, the more the cost of the system.

So the question around the devices then, how do you encourage the user to use the system a heck of a lot more if they keep sitting with the calculator and go, wow, if I connect 10 more devices it's going to cost me more money. I mean, that's just a tricky problem to solve.

No, actually, I think it's a straightforward problem to solve because the value is scaling even faster. If I double the number of devices I double the amount of value and I double the amount of cost but if the value is much greater than the cost, that's a good model.

And I think with that, we'd like to wrap up the call. Jim, did you want to make some closing remarks?

Yes, I did want to address one other point which I -- some people have asked about and we didn't -- it didn't happen to come up here on the questions, and that's a little bit about this tax rate if I could. So some people said, well, your tax rate was pretty favorable and that helped your earnings. And I'd say, yes, of course. But there's a couple of things you ought to be aware of. Number one, our earnings also contemplate a very significant investment in ThingWorx that we've been making far ahead of revenue. So we've been pouring quite a bit of investment in that business. And one of the great things about the Axeda acquisition is it'll take quite a bit of pressure off that going forward. The second thing is that while the tax rate was favorable in Q3, it was quite unfavorable in Q2 and year-to-date, we're actually right on plan in terms of our tax assumption. So if you look at our year-to-date EPS and our forecast for Q4, we're making that year-to-date number and hoping to achieve that forecast number at essentially the tax rate we guided for at the beginning of the year. So it's not true to say that in the big picture, taxes are helping us. In the big picture they helped us in Q3 and hurt us in Q2 and we're right on track for the year. So anyway, I just wanted to offer that because I've been asked about it a couple of times and I wanted everybody to have the same information there. And so in closing out the call here, like I said, we're pretty happy with the results in Q3. We really like this Axeda and Atego acquisition. We really feel good about this Internet of Things and smart connected products business that we built, which is completely differentiated from our traditional competitors, yet very, very interesting to all of the manufacturing companies that we all serve. So we feel great about that and we feel pretty darn good about our FY '15 plan that we put in front of you because that would represent, for us, a sixth consecutive year of strong earnings growth and we want to keep that trend going and I think we feel like we have a beat on how to do it for FY '15. So thanks a lot everybody for spending time with us here this morning and hope you have a good balance of the day. See you in 90 days, if not before. Bye-bye.

Thank you. That concludes today's conference. Thank you for your participation. You may now disconnect.