Fortinet Inc (FTNT) 2015 Q3 法說會逐字稿

Welcome to your Fortinet Q3 2015 earnings financial analyst Q&A.

(Operator Instructions)

As a reminder, today's conference is being recorded. Now, I'll turn it over to your host, Michelle Spolver. Michelle, please go ahead.

Thanks, John. I think before we start -- long time, no talk -- anyway, I want just to remind everybody, the disclaimer that I provided earlier -- on today's earlier call applies to any forward-looking statements we make during this call. Again, thank you everybody for calling back in. Let me -- we can go ahead and start the Q&A. -- Ken and Drew are with me as well.

Should I go ahead and --

Yes. Actually, John, before we start the Q&A, maybe Drew, go ahead and address the earlier question.

Jim -- hopefully, Jim [Fish] is on because I wanted to follow-up. Are you on, Jim?

Yes, he is.

Okay. Great. So the July 8 carryover from Meru on deferred revenue was $10 million. That was about 60% -- 60%, 40% short and long-term respectively. So 60% short-term, 40% long-term. Then again, that's got to be haircut wins -- I think it is roughly haircut in half, what you pick up, so say 45% to 50%. I don't have the exact percent, but in that range. I assume you guys understand the accounting that it gets written down. (multiple speakers) We didn't add that back on the P&L.

Okay, John, you can start the Q&A from the queue.

(Operator Instructions)

[Taz] Koujalgi, Deutsche Bank.

I have a couple of questions. I know you had some pricing changes in the last couple of quarters. The bundled pricing went up and then you raised, I believe, the pricing on the 1500D in the US in Q2. Can you just walk me through all the pricings that have happened in the last few quarters? Then, what reasons?

Quite honestly, it would be impossible to walk you through every price change because there's lots of small (multiple speakers) -- Yes. In case you are talking to somebody who said they made other changes. The only material ones are the ones that I mentioned there that are material to the overall. The bundles, we made that change in January, but again, we caution people that it takes quite a while to work its way through the channel. There's channel notifications and a lot of communication. Then as I explained on the last call, that's going to hit more on new customers than existing customers, because existing customers are going to want to -- typically want to renew at the prior rate what they bought. Then they may have agreements already in place that make it a bit harder to negotiate against. Hopefully, you get them in as you do more deals over time. Then the other things that I explained, it has a diluted impact on upfront revenue on the product side because now, on the overall deal side, you are attributing more to the bundled component, which is an amortized component over the term of whatever the subscription is. Then -- I think one of the key points to make there is again, you are creating a larger annuity stream over time. Then the second part was 1500D's. That was in the US only. That happened -- was it?

Q1.

Was it the end of Q1?

Yes. We announced them in Q1. I think, really, we said we weren't going to really expect an effect until Q2.

Yes. Then, that was the 1500D. Again, same thing, it works its way through the funnel. So there is probably a slight pickup from that -- slight reflection of that in the Q3.

Got it. Then on the US enterprise, the growth slowed down from 90% last quarter to 27%. But then if you look at the overall growth in Americas, that has slowed down that much given how much the US enterprise slowed down. So I'm guessing the exposure was really small in the US for enterprises? Or there's something else which is variable to offset that slowdown.

Our service provider business did really well in Q3 in the US. So it was -- it's a mix. Every quarter, you are going to see a mix of business. But I think in prior quarters, the service provider was doing okay -- world -- the stats that we give from a vertical perspective is worldwide. But we had a particularly strong performance from service provider in Americas in Q3.

Can you guys give us what the revenue mix -- first of all, the growth that you gave for the US enterprise was 90%, 70% and 27%. Is that revenue growth? Or is that billings growth?

Billings. It's for US only.

What is the mix of US enterprises in the US?

What do you mean the mix?

We don't -- (multiple speakers)

What percentage of billings and revenues comes from US service providers?

We don't -- that's not -- yes, we have not given that out, Taz.

What we do break out, Taz, is service provider at the 22%, government at 12%, financial services at 11%, education at 10% and retail at 6%.

That is still global. That's not even on (multiple speakers) basis.

Yes. It is global. Fair enough, yes.

Then when you talk about the US enterprise growth, you don't count service providers in that bucket; right? That's just commercial enterprises.

That's correct.

It's just commercial enterprises. I will tell you that we started doing -- we may -- we'll see what we do going forward, but we started giving it sort of on a one-off thing. It was not -- we didn't usually give that level of detail. We started doing it about two years ago, when we were talking about really getting a footing in the enterprise. We wanted to show progress on how we were doing in the enterprise.

I think a year from now or two years from now, that's going to be less meaningful. It's probably not the way we'll end up breaking out our business, but we really wanted to sort of talk about the growth that we were having and the progress we were making in that space. Quite frankly, there was very few people that actually -- we had an issue in terms of people not believing that we were even actually doing any enterprise business, so we wanted to sort of highlight that.

Got it. Then I have follow-up for Ken. Ken, you spoke about the partnership with NSX and SEI, given that one of your biggest, I guess, competitive advantage is the ASIC in your appliances. When you move, your product will -- the software defined will and the software form factor. How did that advantage carryover to the SDN or the software defined world?

I think first, all the new function are implemented software. Then we'll take a few years to move to the hardware ASIC got to accelerate. Also offload a CPU to carry some additional function. So the partnership with VMware is really, like I mentioned, it's really -- it's in a data center. If they already have some computing power, definitely they can leverage that one to draw on some of the function we have on the security.

But on the cost base -- on the performance base, the ASIC definitely have a huge advantage over the CPU. It can be 10x to 100x, more like computing advantage and also the cost advantage there. But the virtual edition environments really, they can leverage some of the free computing power of restorage. So we are very open for that approach. We're supporting that.

Also we have something called on demand, security function calculation assistance there. So we also in our partnership there. I don't think these have any -- they're probably even more helping of the -- because once they see the difference, whether their using the CPU computing part with the virtual edition of security function compared with some hardware side, they can see the difference. But also from time to time, they're also helping some of the hardware sales.

Got it. That's very helpful. Just one last follow-up. In terms of the, again, trying to explore the weakness in US enterprises. Did the price change have any effect of, I guess, pulling forward demand? Was this price change announced upfront so people bought more in Q1 and Q2? Or that's not even a factor?

No, not at all.

Yes. We don't see much change. Like I said, some view can be chunky. You can see a few million dollar deal can change the percentage a lot. So it's very normal enterprise, big enterprise get a seven figure deal.

Yes, Taz, I think that is the way to think about it, is again, we have been doing really well at the top end. Those customers -- they're testing things. They run pilots. It takes time. It just gets a bit lumpy. Zero -- certainly in the US enterprise, we've seen no issues with pricing at all. That never comes up.

Yes, also --

Our performance is -- we had a very strong certainly versus the competition already.

Also, like I mentioned, the other way to measure is really how many new big enterprise we win in. So we're very happy with that measurement and probably better that of a few other quarters.

So was expansion in enterprise an issue? If you want better -- if you did better in the more new enterprise --

Yes.

The question is, did the expansion of the existing customers -- was that weaker than what you had in the previous quarters?

No. We just feel like we are seeing a concentration of deals towards the top end. We've hired people. Again, very consistent with what we've said in the past, as you hire people, they tend to work off the top of the list -- off the top of their customer list, which continues to be the case.

We're hoping -- where we see opportunities to make investments quite honestly are in the middle and lower end. We just believe that's a different -- we've hired the people. We just want to make sure that we have the right marketing engine going there in terms of market awareness, the lead generation cycle and the sales enablement. Those things are different at -- where -- depending upon the size of the customer in the enterprise space.

Got it. Thank you.

Hendi Susanto, Gabelli & Co.

Then two questions for me. First for Drew, how should we view the gross margin guidance for Q4? It is lower than Q3 but slightly higher than Q4 last year. Then we know that we will see some positive contribution of pricing increase and Q4 being a seasonally strong quarter.

Well, look, it's always nice to over-perform. We've guided 70% to 72%. I think we usually guide 70% to 71%, quite frankly. So we've moved that up a point, reflecting the value of the bundled pricing. Again, we have -- we've cited Canada and Latin America. Who knows, maybe those two cause pricing pressure, as we talked about early -- on the last call. But we think that's the right range for now. We'll see how the longer-term trend goes.

Okay. Then one question for Ken. Ken, in the security analytics area, we are seeing many emerging new companies and then existing companies extending their footprint into that. How should we see Fortinet in the security analytics area and your aspiration?

We want to be the partner and like I said, as we have some drawing even with Splunk, also some of the customer see the huge benefit. So we definitely want to have more partnership in this area.

If I may ask further, I think, for your partnership with Splunk, I believe that it may not be able to address like many areas? So are you going to focus on certain areas with Splunk?

We are doing a lot of big data analysis on all the [sends] that sometime is working together to -- because security is something that can generate a lot of log, the sends, and also try to see what's the normal traffic, what's abnormal. So that's need a partner with some other, like a data analysis company, two working together to find the best way to protect the company.

Got it. Thank you.

Thank you.

Walter Pritchard, Citigroup.

This is Jim on for Walter again. Thanks for the clarity around that 60%, 40% number. I guess my first question would be -- it seems as though you guys really saw a deceleration. I know we've talked a little bit about it here. But what was the source of this enterprise -- US enterprise deceleration? Because -- it's not Canada or obviously, not even Brazil. But what is the source here?

Jim, I think what we've been saying is what's been true. Quite honestly, we seem to have been -- we deployed the sales team more around the top end of the business. I do think there are -- the sales cycles there are a bit lumpier, quite frankly. We are clearly in many deals, many things are out for RFP. We're looking -- we're in the deals. It's just really kind of more timing on when those larger deals close. I'd say another piece of it is when you get in some of these initial deals, they may be smaller than you would think.

They're maybe just getting an initial pilot in the door. I think as you get into some of these newer, large customers and we're seeing some of that. Overall, I would say the funnel is improving. So we can't point to anything in terms of seeing deteriorating demand in the way -- in what we see from the funnel perspective or the overall demand environment in the US. Look, I think 90% was a great number. It'd be great if we could compete that overall. But again, we kind of go back to the 41% growth overall. I know you're talking about the US, but if you look at EMEA, clearly the growth there was very good at 50%.

Got it. To kind of keep going along with this. With you guys talking about investing more in the lower end market, which seems to go away a little bit from the strategy that's been going on the last year or two --

Not the lower end. It's not the lower end, Jim, is actually the enterprise.

Mid to lower end.

When we talk about the enterprise, it is -- there's everything from mid market enterprise, which is not -- it could be a company that's $500 million to $1 billion or more in revenue. So when we say lower end, it's not investing in what weak -- what low-end meaning SMB, it's still the enterprise. It's just --

Mid and smaller enterprise.

Yes.

So think of the -- it's not SMB. We're not talking about SMB. That's not the --

Yes.

Yes, I think we do give a lot of detail than most other competitors, like we try to divide into service providers, like what's the big enterprise, what's the small -- mid enterprise. The definition of big enterprise is really using the annual revenue of a $750 million above, we treat as a big enterprise. But that may not be the most scientific way. Also sometimes it may not be the best way to line up a sales force. I think looking overall would be much better.

I think sometimes we give too much detail, which sometimes can mislead. So that's the part I think that we may need to see what's the best way we can address and explain that going forward. But also we may even change some of the definition to line up the sales more efficient in the field. Because this part could be to low or could be a -- we may do some adjustments or it could be to high. It depends on what's the best way. Because the Company now is quite different compared to a few years ago, when we started to define and building the team for -- whether the enterprise or the commercial or some other part.

But like I said, we try our best -- to give a lot of detail. But from time to time, like it could be a little misleading, like we do give the carrier. There's only a few carrier account in the US. You can see one year ago, sometimes there's up and down because certain deals can be in certain quarters these can change the number a lot. Like I said, we were happy with the enterprise performance. We win a lot of new customers. From time to time, once there's a seven figure deal fitting into one quarter or shift a quarter can be making a percentage misleading. That's the message.

Then, Jim, you probably heard us say this before, when you look back a couple of years, we were very strong in carrier, very strong in SMB. We had some enterprise presence. But the investments we made were really focused on the enterprise. We've come down better from the top than we've come up from the bottom. I think that's because if you hire -- the type of people we are hiring are out of named companies, quite frankly, where they have strong enterprise presence.

They tend to attack the top of the customer lists first. Those deal cycles can happen depending upon how testing goes with us or even a competitor. So those are things that are a little harder to control. Where we -- we've said this, where we really thought we wanted to invest going forward -- this isn't new to this quarter or this call, is that -- in the mid market it's a different motion, where it's kind of a more direct relationship one on one, testing coverage more towards the top.

Then in the marketing -- in middle, it's more kind of market awareness, the lead to closure cycle and sales enablement. We need to -- we're clearly investing in those processes and systems and approaches to go improve in those areas. I think that's something we've called out in the past. That's the truth. But, look, I hear your point on the growth, but certainly, the names we are taking down are great. We are selling really well at the top end of the enterprise. The brands are phenomenal -- the name brands. What we want to do is be able to translate that more into the middle and smaller end of the enterprise.

Well, I mean, my main question was going to be around, is the return on investing elsewhere lower than investing in the enterprise market at this point?

A fair question. I think it's too early to tell. Our thesis is that it's not. We believe that it's a higher velocity business. When you get the engine humming, it's something that actually could even be more efficient, better return.

Okay. Thanks, guys.

Saket Kalia, Barclays.

Just a few clarification modeling questions. First is, what was the headcount in the quarter? Any way to think about the hiring, excluding the employees that came from Meru?

Yes. We had 3,889 at the end of the quarter. 3,889. We don't breakout about where they come from, but --

Is a net at about 520 including Meru. I think Meru has about 300 people.

Yes. Slightly north of 300, I think. We haven't been breaking it out, but --

That's what they had? Or what they have now?

No. How many came over.

Okay.

Then we've done -- yes, I think that is in the 388; what's left, right? Yes, in the remaining. Then we've -- obviously, we've restructure some of them. Where we've found redundancies across the Company, we took care -- we grabbed those efficiencies.

Yes. Then on, just going back to product revenue really quickly. Would product revenue have followed your normal third-quarter seasonality, excluding Meru?

I don't know. I don't know that we have a product revenue seasonality --

Well, I guess flat, flattish. The last few September quarters, it's been sort of flat to up slightly. So, just --

I just looked at -- yes, I think overall revenue had been roughly -- overall, it had been kind of flat quarter on quarter, not necessarily just product but I think overall was the history, if I have that right. Billing's flat, actually -- it's billing's flat. What happens on the revenue side is the -- again, because of the price increases in the bundle, that is very different than a year ago.

So just the whole accounting action, the way -- you're recognizing more of an overall deal in deferred revenue versus upfront revenue. That would change the product. That being said, even the revenue that I have recognized in -- yes, so that plugs up the service revenue base -- yes, the services revenue. There's more of a bias on the service revenue, because of the price increase.

Right. Actually, I'm sorry, this may be a dumb question just on sort of that accounting. But if you're selling an appliance for $100 and the subscription bundle is, whatever, 15% of that -- the way that I understood it, was the price increase happened on the 15%, but the price on $100 stayed the same. Is that right? Because I guess the -- in that case, the price increase from the subscription really should just affect subscription revenue and not really product. Does that make sense?

Yes, but there is -- what happens is, the overall deal values may or may not -- you're going to get some benefit on the overall deal value, but the relative allocation of discount could shift the accounting a bit.

Oh, the allocation discount. Got it. Okay.

Yes.

Then for my last question, presumably Meru was still somewhat dilutive despite the good margin performance in the quarter. I guess the question is, when do you expect to have an integrated wireless offering and for that business to become accretive?

We do have the integrated wireless with security offer. Basically, FortiGate also has the control of function built-in. We started selling that a few years ago. The Meru just helps us get into more customer base and also add engineer resource there. Just help us to keep expanding in that area. But we do have that solution a few years ago.

I think he is asking, Ken, about an integrated product that integrates some of the technology from Meru. We talked about during the acquisition about integrate -- having a product line that will integrate our own security with Meru's enterprise wireless technology.

That's probably still like to need to take a few more months.

Yes. I would look into -- it's probably early 2016.

Got it. Then is that a good time to -- I don't want to put words in your mouth. But is that a good time to think about when the deal maybe starts to become a little bit more accretive? Or is it longer than that? How do you think about Meru from a profitability perspective going forward?

Definitely the FortiGate and plus a lot of additional service has much better margin than the traditional Meru solution.

Got it. Very helpful. Thanks, guys.

Thank you.

Catharine Trebnick, Dougherty.

Mine is more on your European numbers. It looks like it was up pretty strong. Is there any way that you can bifurcate that for us on how much of the opportunity was large enterprise versus SMB? The reason I'm asking that is, in doing some work in Europe, we came away with a big spend in Europe in the next five years is in the SMB areas would be over $30 billion. So I'm trying to get a better handle on, was it the hiring you did in Europe? Or is it more related maybe to a large or small? Thank you.

European hiring started early. If you remember a couple years ago, there was some slowdown in Europe, so we took the opportunity to starting building a team and also the management also is market is in position there already. The US, we're just starting ramp-up, the sales capacity hiring actually pretty much this year. That's still in the ramp-up stage. I do agree in Europe long term, a lot of SMB small office. They need to be (inaudible) to be managed, which we have the best part of position to take the opportunity there. So that's also what gave us a lot of potential over there. But also, APAC, we see a lot of opportunity similar like in Europe.

Catharine, I think if you go back and look at the history of the Company, there was always a strong SMB play, especially in Europe. That hasn't changed. We have certainly done better on the enterprise. I don't have an exact split to share with you. But our channels are set up. We are well-positioned on the SMB front.

Yes.

That's been -- that was our --

All right. (multiple speakers) Thank you. I appreciate it.

Sterling Auty, JPMorgan.

On the higher end solution, is there a sense that you can give us in terms of the number of subscriptions that you're seeing be utilized? When I say utilized, when you've got the mid-70% renewal rate, what are the ones that are really renewing well?

The issue that we have, Sterling, with our subscriptions is that we came from the get-go -- we came out with a bundled subscriptions offering. So over the last year, I think, maybe about last year, we introduced in ATP offering -- subscription offering. Short of that for the last six, seven years, we've had bundled offerings. All through that time, we had roughly the majority, 80% or so of customers were buying the bundle, so it's difficult.

They aren't necessarily -- with the exception of SMB, they're not necessarily using all of the services. But economically it made sense to buy the bundle and then also give them the flexibility if they wanted to turn one on later on. So there isn't a big change in terms of attach rate, which is I think what you're asking -- attach rate per subscriptions and what the trend is there. There is not a big change because from the start, customers saw the value and were purchasing the bundles pretty much all along.

Also going forward, we are probably -- sometime this quarter, we're launching another bundle called the enterprise bundle, which like in addition to the UTM bundle, we also add ATP subscription service and also the mobile security service. So that's probably give us a few more quarter we can see what's the -- in the past, we only had one bundle. 80%, the buyers, really just within our UTM bundle.

Okay. Then I want to revisit the idea of the price increase in the rest of it. I want to attack it this way. I apologize, I was jumping between calls. So maybe you described it this way. But when we think about going -- using Saket's $100 analogy. So if you sell an appliance for $100, what portion is being recognized upfront now versus what portion is going on the balance sheet? How did that compare with what it was previously? So in other words, so we get a Point A and a Point B and maybe which way it's trending?

More is being deferred. I don't have the exact percentage for you, Sterling.

Is that something that you can follow-up on? Because traditionally, when we started all the way back to the IPO, the idea was it was kind of like a 40/60 split. So there was an average, so it really helped us in terms of being able to do that modeling. I think it would be helpful if you gave us a sense of what it is on a new appliance with the pricing; where it was, so that we can say, okay, well there's the trend line. Start here and over a period of, that's subscribed a year or two, it will gradually move toward this new mix.

Yes. We could try to follow-up, Sterling. Yes, I don't have a number for you. I mean, you can look at the deferral portion as a percentage of overall billings as a trend.

Okay. Then in terms of the investment looking at the fourth quarter, what's the expectation in terms of where you might end the year in terms of headcounts? In other words, when we look at that investment in margins, how much of that is headcount? Are you looking to add 200 head? 300? Et cetera?

So far this year without Meru, we probably added about 200 -- net add about 200 a quarter. But with Meru, we do add additional 300 on top of that. Like I said, in early, we not only need to look at how the additional headcount add, but how also how do we improve in some productivity and also how to have a better process in the two to help the sales and also the marketing side. But this year, we probably will not change the amount. We have not planned the next year yet. But again, we look at it both end, both on the adding people but also improving the productivity.

Okay. Then, last question, Drew, you mentioned, lumpiness in the high-end in terms of deals, describing why quarter to quarter, you might have a strong service provider versus enterprise. But kind of curious with the pipeline here in relation to the guidance that you've given for the fourth quarter, are the opportunities skewed to the service provider versus the enterprise? Or vice versa? Just so we can get a sense heading through the quarter, what it might shake-out like?

Yes. Look, I mean the way we forecast is bottoms up. We really do look at the bottoms up forecast and pretty much go with what the team is telling us. I don't have any particular sway one way or the other on -- outside of our traditional percentages that we shared -- that we share as a percent of the overall business in the splits -- in the vertical splits, for instance. But I don't -- again, the issue with that, even, is you never know what's going to land when.

Okay, thank you.

Michael Turitz, Raymond James.

I've been jumping on and off. So help me clarify this enterprise issue. So you were 90% plus year over year last quarter in US enterprise, right?

Right.

Then it was 27% this quarter in US enterprise?

Correct.

So that deceleration, it is not just some weird comp, it is a question of the fact that US enterprise definitely slowed in some way. I understand that's because it was at the very high-end? Where you do big deals, with great big boxes, is that correct?

Well, not great big boxes. A large deal -- one of the largest deals in the quarter would have low-end boxes because it was a quarter branch deployment. They bought the branches first. The branches are lower end boxes.

Okay. So by high-end, you mean the largest companies?

Yes. We are meaning like Fortune 25, Fortune 50 companies, when in reality, it -- maybe it isn't that -- when we're talking about mid market, it might be a Fortune 600 company or something like that.

Yes. I would say Fortune 500 is large.

Yes. Okay.

But it's the largest company -- so if that's the case, if I really think about what was weaker than expected in the quarter, should I focus on that as opposed to focusing on say, the emerging markets? Is really -- it's a lumpiness at the very high-end. This quarter those biggest deals with the enterprises didn't come in.

Look, first of all, we met guidance -- we did better than we set on -- we guided on billings, so we didn't fall short on billings. We were at the high-end of the guidance on revenue. In terms of the business, again, it's a portfolio. Some porter -- some tend to do better than others. This quarter, we had excellent quarters in APAC and EMEA.

Just to get to the point here. I just do think there's a concentration -- what we've seen is a lot of effort on the high-end of the business. Those things, you're just subject more to a timing of when those things land. What were trying to do is create more balance over the rest of the enterprise beyond that Fortune 500 or whatever the number is.

I would say in terms of the economic question, it's absolutely true that Canada did negatively affect America's performance in Q3.

Yes. Again, but he's --

No. He was acting with the enterprise and not Canada. (multiple speakers)

Right. But does that --

Okay. Fair enough.

Is that significant? I mean there's lots of little things. But that sounds like it's -- would you say that is a lot less significant than it's enterprise position?

I don't know if it was a lot less significant, but it was --

It was a factor.

Yes, it was a factor.

Okay. So how did you -- have you told us how much of your business is enterprise, is characterized as enterprise business?

No, we just break out the global verticals, Michael. So, we say, like telco is 22%. Government's plus whatever --

Right. So it is very clear that of your enterprise business, it's really been skewed towards the high-end and you're trying to get it moving down mid --

Trying to broaden that. Yes, we're trying to broaden that to different motion, different go-to-market motion, different selling cycle, different market awareness, the way you go about leads and so forth -- more systematic, let's say, and approach to it.

As I look back over the past, say, four quarters and obviously it's lumpy. But would you say this was the weakest in that area of the last four quarters? The weakest of the last four quarters for US enterprise?

No, the amount of business coming from US enterprise is definitely increasing quarter after quarter after quarter.

Not sure that's a problem.

Yes. I mean, obviously -- so it's not weak but I think from a growth perspective, as the numbers get bigger, the growth is going to get smaller, 27% would still obviously -- there's a big difference between 90% and 27%. But like I said, I think in the earlier calls, a year or two years from now when these numbers get to be more sizable, that the growth was going to look a little bit different.

Look, I hate to -- okay -- I hate to come back to this, but just and I hate to use the, [myself] better say this all the time, I'm getting a lot of calls on this -- everyone has just asked me, why not break out Meru and tell us it's a couple of million, it's $5 -- what's wrong with doing that, if it will help us understand and not come to the conclusion that your business is decelerating?

We try to traded him to more like a Fortinet secure Wi-Fi solution. So that's the focus we're doing there. So that's different than what Meru traditionally is doing there. So that's -- also it's the one team. It's not a separate team. It's all merged together. So that's from -- the day one, we said, we're clearly -- so we try to offer the secure Wi-Fi solution and also will be one team. So that will be difficult to break it out.

Okay. Thanks.

Melissa Gorham, Morgan Stanley.

Just a follow-up question on the service provider business. Was there any sort of -- you mentioned how it's lumpy, but was there any sort of one-off, particularly large deals in the quarter that were driving strength that we should be aware of? In the service provider business?

There weren't -- it wasn't anything unusual. We had a few, actually, pretty good seven figure size deals in the service provider space. But it wasn't anything like an eight figure or something, very, very unusual in our business.

Yes. That's the -- a year ago, we see like whether the 4G LTE or whatever and also a security mobile device you may need to go to service provider because that's how the data traffic -- the mobile device, data traffic goes through a service provider, I doubt there's a lot of enterprise control. We see international pickup kind of before the US, now we're starting to see US also starting to pick up.

Okay. Thanks. Then just following-up on your government business. Just given that September was the Federal year end. I'm just wondering, you did get the breakout of the government vertical, but did you see any sort of Federal year-end budget flush this quarter? Any commentary you have on investments you're making to ramp that vertical?

I don't think it was anything unusual, Melissa. It was a decent business. It held its own, so to speak, overall government globally was 12%. As we've said before, we don't have high exposure to the US Federal government. What I'd probably more as where you're question is aiming, but it was -- I would characterize it as good and probably is evidenced by that number. So consistent with -- fairly consistent with a year ago.

Okay. Great, thank you.

Jayson Noland, Baird.

I wanted to ask on service provider, the top line growth there year on year was the best in years. It sounds like that was US-centric. Is that a function of Tier I telco? Or something else?

It was the Tier I.

Yes. It wasn't all US. We had strength in the US. But we did well internationally as well, but was Tier I.

Okay. Would you describe this as the start of an investment cycle? Or is it more project driven?

We may need to wait one more quarter to see more. But we do see some more positive changing. Like I said, it's like a year ago, a couple of quarters ago is more international start up first. Because a lot of our carrier there in the process upgrade to the 4G, to the LTE order since. Also even the data center try to changing from 10-gig to the 40-gig to the 100-gig. That we're starting to see is a good potential there.

Okay. Is it managed services, Ken? Or again, more broad?

Both. Both for the internal use and also the managing service.

Okay. Then last question, Drew, I think you mentioned there's some work to do on the marketing side. You've got a new CMO onboard. Is this brand-building, region, is it more focused on APAC? It's been described as an opportunity. Some more color there?

Look, it's a -- I think in terms of -- so we said three specific things: market awareness; lead generation, which really means lead to closure and how we go about that; and then sales enablement. In market awareness, I don't think we're going to do tons of branding, but I think it's just getting our name out there in articles and how we do our website, even, and how we interact online and through social media. So it's that whole presence and name presence is that. So that's part of the investment.

The lead gen cycle is how you generate leads and use your inside sales team to either direct those to our field or through the channel. We feel like there's an adjustment we want to make there. That's not only -- it's some people, but it's process and systems as well, so thinking through how that whole cycle will happen and making investment there.

Then the last piece is sales enablement. That's tying the other two things together, quite frankly. Educating the sales force and then training them with the right messaging, make sure they are loaded out there. Again, it's the top end. They are selling technically, right? So it's a technical sale. We're very good at that. I think getting the messaging across in the middle enterprise and smaller enterprise or the broader enterprise market is going to be slightly different. That's what we're trying to do is get those gears set more towards those markets.

On the regional way, is the more folks in US, not APAC. APAC, we have a market position and many countries running number one, number two. We're keeping UMS in there. But in the US, we're kind of a little bit behind with the competitor, a lot of buy in, in mid enterprise. So that's where the new marketing will be more focused in the US, mid enterprise.

Okay. Is this something we should consider for our models? Or is it more of shifting dollars around within your budget?

Well, we've given guidance. We're going to stay within the -- we guided to 14% for the year. We're again within that, we've acquired Meru and absorbed the dilutive impact of that, so to speak in that and still held. This will be within that budget. Again, we're going through the planning cycle now for next year. We'll have more on that in January.

Okay. Thanks, guys.

(Operator Instructions)

Fatima Boolani, UBS.

This is Fatima on Brent. I just wanted to drill into the scope of the mid market investment. You've had number one market share in the UTM arena for several -- many years now. So I want to better understand the scope of the opportunity that you're seeing in the mid market? How that changes your competitive landscape? I have a couple of follow-ups after that.

Yes. That's probably by the IDC data as a global number. In the US, we still need to keep on catching up. We're still falling behind in the US. Some of our competitor grow faster than us in the US. But we are also winning some international. Whether it defines as a UTM or major firewall, that is between the research firm there. But we do see, there's a lot of greenfield in the US mid enterprise.

Because the big US enterprise and also the carrier, they have a more strong technical team and more resource to RFP all the testing. The mid enterprise more depend on the marketing, the branding to make the buying decision. That's where some of our competitors are doing better, invest much more than we are in the US. So why where we are in the catch-up mode right now. So we feel the investment in the US mid enterprise will be giving us a lot of potential going forward, because we have a lot of greenfield to cover.

Just as it relates to that point, I think at last count, you had one of the largest foreign networks that I'm aware of, about 20,000 channel partners. So as you build out your internal sales and marketing functions, I think your headcount in that area is up 80% year on year. As you now focus more downmarket, if you will, how do you balance or manage over coverage or potential conflicts with partners? Because the mid market has traditionally been the bread-and-butter of the reseller community. So we'd just love to hear your thoughts on that.

I think we're starting at a headcount in the position like you mentioned there's a lot of new people on board, probably close to half of us in the US mid enterprise is really has tenure less than 12 months. We need to keep in the training and also learn our pattern with the partner. That's a lot of marketing effort put in there and how to support the partner, how to get a message to the customer. That's the focus we're doing right now. But like I said, we still behind there compared to some of our competitors. We feel there's a lot of growth opportunity over there.

Great. Just a product question, if I could. I know there is some maybe investor misconception around the fact that when you refresh your chips it causes some sort of a purchasing pause or customers sort of balk. I'm wondering if you can just remind us what chip-based upgrades are on the horizon? What the typical cadence has been? If we should even consider any changes in purchasing patterns? Or if we should be mindful of anything in that regard?

We kind of are more evenly distributed. Every quarter, when we build a new, like the 40-gig system platform there, like we early release a couple every quarter. Like the chip we had mentioned, the CP9 will be starting building the new product, more in the high-end and middle range. You can see the pattern like two or three years ago, once we released; that was Processor 6. So that's helping a lot of on the middle range, on the high-end in the performance on the interface side on the network side.

This [kind] of processor actually will be helping on the core. It will help improve a lot of performance in the application content, helping the CPU and also offload a CPU there. That's the new chip. But the way we roll out is probably just follow the same pattern as a couple porter every quarter, and I don't think we -- no -- the pattern has not tried to refresh ours in the same time. It is just a couple port I record are gradually rolled out. Depend on the like a certain technology progress in the networking, in a CPU, in the memory space.

Yes. I was going to say, Fatima, it's Michelle. If you look historically, we don't have a lot of peaks and dips that are associated with chip refreshes. When we come out with a new chip, it takes -- we have several models of our FortiGate devices, at least on the network processor and the content processor. It goes into all of our midrange and on up, versions of FortiGate. It could take two years to refresh the entire product line. Customers don't know which models are being introduced at which time, so the refresh is extremely gradual. So you don't end up seeing a lot of disruption, I think, or volatility, I guess in the product.

Yes. Because the chip is very unique. We have a huge advantage using the chip here. It's different than some other semiconductor, whether in the storage or the mobile funnel or whatever. New chip [were chin and hosting] (inaudible). Here we can pace out the way we roll out, because this is a very unique advantage, none of our competitor has. We can while keeping gradually roll out the pace that we want.

Now, last question for me, if I can speak one in. I know Meru had about 14,000 customers around the time you acquired them. Can you remind us if there's any customer overlap in terms of their customers having any Fortinet products? How you think about that cross-sell opportunity into that footprint? That's it for me, thank you.

I think we mentioned that only very low single-digit overlap. So there will be a lot of opportunity. Also 70% has come from education space. So that definitely will help us get into some new customer base. But the key is also how the train both the sales force and also the partner to sell the security, which is pretty new tool to a lot of the people or the partners there. So that's also will take some time.

That's very helpful, thank you.

Erik Suppiger, JMP Securities.

So I'm just a little surprised to hear the focus on the mid market. To date for the last couple years, your large enterprise has really been a driver of your growth or so it has been my impression. Do you think that you will continue to see the large enterprise growing faster than your overall revenues? Or do you think that more of a focus on the mid market is going to start changing that dynamic?

We are definitely keeping invest in the large enterprise. I think that interestingly enough, the carrier and the large enterprise, they're very technical. They do a lot of study and the back-up testing before making a brand decision. That's where with our product advantage, we have more advantage getting there compared to some of our competitors because their product is much weaker.

So that is where we keep winning like whether in the carrier in the big enterprises, we're starting seeing a lot of good results. The mid enterprise need a lot of marketing help and also need to be working closely with the channel partner, enabled them and also did helping with the lead and with all the process the tool. So that's where we're behind. Also the mid enterprise is also is much bigger than the large enterprise market there. So that's the part we keeping -- we feel we need to catch up.

Yes. Erik, it's true. I think Ken said it well, in terms of the selling motion. It's more -- as we've been saying, just the whole process of creating leads and handing them off either to our sales people or the channel -- just the whole market awareness where brand matters. We seem to be well-known, certainly with savvy tech companies. We do well there, but again, as Ken said, those sales cycles are just harder to call, because you grind up in testing and so forth. So that takes -- it is just harder to predict. We view it as an opportunity is the way -- without getting into specific numbers.

What we view it as is an opportunity because we really haven't hit that. That's why we're making the investment there. It's a large part of the overall enterprise market. We believe that's an opportunity. We know that if our products work in the most savvy customers in the world -- tech savvy customers in the world, they ought to work in the rest of the customers. That's what we're really trying to do is go get that messaging across and go get that go-to-market motion, so that we can exploit that space.

If I could actually just finish off with one thought is that, we talked a lot about the mid market. I want to make sure it's clear to everybody that it does not mean that we are taking our focus off of the high-end enterprise. We have made a lot of investments in that area. It's tracking very well, like we've talked a lot about our man and expand strategy and that we are landing really, really important -- great seats at very large tables. There's tons of expansion opportunity. There's tons of landing opportunity.

What we see is that in looking at the enterprise, there's a piece of the enterprise that we're not quite -- that we haven't quite focused on as much. That provides really if you look at sort of the lower end midrange enterprise, that's a lot of more -- it gives you sort of more of a run rate type business than the very high-end of the enterprise would. But we are still very -- when we talk about enterprise, we are very focused on the enterprise. We're still focused on the high-end. What we've talked about is that we have some catching up to do on the midrange.

Okay. In that midrange, do you anticipate running up against players like Palo Alto more? Is that who you envision being -- are any of the dynamics -- competitive dynamics there different from the larger enterprise?

Yes, in the US.

Yes. It's not -- it doesn't introduce any brand-new competitors because we do -- we compete with our primary competitors across the enterprise, but Palo Alto is one of the players in the mid market that we would definitely compete with.

Okay. When you talked about the growth in the large enterprise, is -- I want to just be clear, that growth rate pertains to, is it just the top 25 accounts that you're measuring when you talk about 27% and 90%? Or is that just a handful of accounts that you're measuring when you give that metric out?

No. I mean, he's asking sort of -- I can probably get back to you on it, Erik. It's not that. We don't look at it by like a Fortune 25, Fortune 50. It's the enterprise team, but not all of -- it's hard to explain a little bit. It's the enterprise team but then there's some business -- enterprise business that goes into other teams too.

Erik, I'm sorry, I'm not sure I caught -- can you repeat the question? I just want to make sure I answer -- see if we can answer it correctly.

The question is, I'm trying to get a sense for how representative the -- when you talk about your growth rate being volatile and being that 90% coming down to 27%. That's volatile. You said it's skewed by a couple or few deals. So I'm trying to understand, is that representative of a large portion of your revenues? Or is that something that where -- just it can just be a few accounts that move it one way or the other?

No. Yes, Erik -- it's basically, he's asking like do we track -- when we give a statistic for US enterprise, are we measuring it based on, say, the top 25 companies and we would characterize that?

No.

Right.

No. We're not. No. US enterprise is up from the SMB space, basically is how I would characterize it.

Okay. Then last question on the service provider front, it was strong, certainly in North America. But the service provider contribution was 22%, which was consistent with last quarter and where you've been for a little while. If the US service provider was exceptionally strong, why didn't we see the overall service provider business pickup?

Well, I think just -- again, a little bit like the large enterprise -- it's a portfolio thing. One quarter -- overall, it holds its own. But in one quarter, one [GO] may do a little better than another inside any of those spaces, including the enterprise. I think that's especially as you get to the top end of the enterprise, it gets -- it starts looking a bit like the carrier market. So -- but yes, overall, you're right. It's a consistent trend. But that can be the case if you go back over history, that would be the case. You may have one or two GO's doing better than the other one. Then it's just more timing. Sometimes it's balance, sometimes it isn't.

Okay. One, if I might squeeze in one last one. Do you think that the US enterprise -- do you think that there is a possibly a pause as they assess their cloud architectures and their changes in terms of where they're putting their workloads? Is that contributing to any change in terms of demand at all?

Erik, we haven't seen that. Again, we feel like our funnel is fine. We are getting lots of interest on our products and talking to lots of enterprises. We do hear people talk about the cloud. I don't see a big shift, certainly in the market we're looking at, shifting in that direction or stalling deals or something at that point. That doesn't seem to be the reason.

Yes. Also our cloud side of the business is also pretty small -- relatively small. So like I said -- Drew said, we don't see much change, but also it's a very small percentage of our business right now.

When we talk to customers, that's more of a visionary thing. We don't see it. Is more of -- look at the factor when they buy. But it hasn't been, as far as we know, an excluding factor so to speak or anything that has changed the timing of an opportunity.

All right. Very good. Thank you very much.

All right. Well, thank you, everyone for a rather long call. We appreciate your support. We look forward to seeing you all at the upcoming conferences and various one-on-ones and certainly in January. Thank you very much.

Okay. Ladies and gentlemen, this does conclude your conference. You may now disconnect. Have a great day.