Fortinet Inc (FTNT) 2010 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to Fortinet fourth quarter 2010 earnings financial analyst Q&A session conference call. At this time, all participants are in a listen-only mode. Later we'll conduct a question-and-answer session, and instructions will be given at that time. (Operator Instructions.)

As a reminder, this conference call may be recorded. I would now like to turn the conference over to Mr. Ken Goldman. Sir, you may begin.

Great. Thanks. And by the way, this is purely a Q&A, so we will take your questions, and I would also add that the forward-looking statement disclaimer that I will do you a favor and not repeat fully here, but I'll just say as was discussed earlier in our earlier call applies to forward-looking statements that we will make in this call.

So with that, again, this is meant to be questions that you may not have had a chance to ask, questions that may have come up as you've looked through our numbers more, or questions related to your expected models. So we're happy to take questions as they come, in any order that they come.

Thank you. (Operator Instructions.) Jon Ho, William Blair.

My first question is on just the high end FortiGate product that you guys issued in the fourth quarter. Can you talk about the margins for this product relative to the corporate average and whether you expect that to make much of a difference?

Yes, we actually don't necessarily look at specific margins by product here. We do look at it by various customers. But in general, margins that we see, the higher-end products, are higher than in the very low end, the two-digit numbers, if you will. So we will see somewhat higher margins in the newer higher-end products versus the lower-end products.

And again, I'd just remind all, our margins really are a mixture. Our product gross margins, always tend to be right around 60%, plus or minus. And then the services business, which has combined FortiGate and FortiCare -- or FortiGuard, I mean, and FortiCare -- always tends to be more like the mid-80s.

Okay. And in terms of the federal government opportunities that you outlined, can you maybe talk about where you see opportunity to sell there and whether the government's taken, I guess, more of an open approach towards UTM solutions than in the past?

We bring a new team facing DC at these Q3 and Q4 last year, so that's where we're starting on growing the federal market. On our side of the UTM also fit in well because multifunction fit into the easy management and also kind of a faster and secure environment to the, in the federal government, certainly because they also use a multiple line of defense. The latency as a faster security is also quite important, so that's where we see the opportunity there. But we be seeing, that's the area we feel will be growing faster going forward.

Yes, the key there is many cases getting qualified into a program, and then that program will have various buys over time. So we have some of those, but honestly, there's a big opportunity to get well more than those as we -- and again, given the consolidation and it's cost effective, we do think it really fits in what the federal has to go do going forward. So we're very hopeful we'll grow the business as we -- this year and forward.

Great. And just one last question. As you guys start to sell more through the MSSP channel, are there any potential conflicts with your MAR channel customers, given the overlap?

That's an interesting question. I don't know if it's conflicts with the customers as much as we have to manage it internally. And so we work that internally as to whether that business goes through our enterprise organization, whether it could go to our federal organization, it could go through our service provider organization, and how do we handle the various relative compensation, if you will, I'd say it that way, for all these various groups? And so it isn't so much a customer issue as opposed to how we manage to support that business from our end.

Great. Thank you.

Thank you. Philip Rueppel, Wells Fargo Securities.

Great. Thanks very much, and thanks for taking my questions. I'm sorry if some of them have been answered. I had to drop off for a little bit. But as we look at 2011, are your assumptions for growth based on any major change of mix? It's been holding steady, both on a product perspective as well as service versus product mix. And is there any change in your assumption of your ability to take share versus what you think the industry's going to grow at?

I think the enterprise we see more widely separate UTM now compared to a couple years ago, because they see the ease of use, the speed of UTM also improve a lot. And that's where we see the enterprise, the big or the small, starting more start at the UTM compared with a point product in the past.

Also feel vertical market, like we mentioned, we're in the financial service and also in some other like healthcare and also the federal, they tend to have a more high-end product, faster environment, and also that with an account-based data center solution. So that's why we feel that the new product, we are now at the last quarter, most, the new 5000, the FortiGate 5000 and also the 3000, the new 3000, that's probably what we'll see out this year for the growth.

Yes, I think the way I think about it is each of our regions and each of our product categories has their own respective opportunities to grow. And so we share the percentages of the various business units, if you will. But honestly, the most important thing I always remind, certainly internally, is we want to grow the whole pie. And the reality is, each geography and each product area has comparable kinds of growth opportunities going forward.

So we don't have a stated goal in mind to grow one versus the other as much as to grow the whole holistic growth total as fast as we possibly can. And that's the way we really think about it.

Okay. And then, just from an enterprise business outside of the MSSP business, which is potentially more greenfield, when you are starting to replace existing entrenched competitors, is it more in the remote office and then you move into the core, or are you seeing some, because of the new high-end product, going right into the core of the networks at the high end?

I see so probably all size of enterprise, and most usually we replace the old firewall or some other point solution. That we see, surely not quite a, mostly from the smaller and bigger enterprise. I see all of them starting. We've got a few deals where we mentioned in the earning call, there are some bigger ones, but also the smaller ones also should be growing quite well.

Okay. And then, Bruce, from a numbers -- or I meant, Ken, from a numbers perspective -- tax rate, did you mention what the assumed tax rate is for 2011?

Yes, I actually did in a follow-up, because I did not include it in my prepared remarks. But what I did say is our guidance for this year is tax rate for pro forma is 33%. That's up a notch from the 32% we had in '010. It does take into account directly the inclusion of the primarily R&D tax credit and a few other credits here and there. So we did guide 33% for '011.

Yes, okay, great, thanks. And then the final one for me, just a follow-up on the federal business, do you sense that there -- I mean, because we're still in continuing resolution, which has affected some of your competitors -- do you sense that there is pent-up demand, including demand for UTMs in the federal government we could see if that starts to be -- new budget starts to be released, we could see some increases in that segment for you near term?

I think for us compared with some of our major competitors, we more earnestly try to do without the federal team and get order simplification and also carrying to the more -- .

New programs.

Yes. So that's where it's more important for us. That's where we have the new team lead by a few who joined in the second half of last year, starting kind of run past that space quick.

Yes, it probably has a modest impact on some of our existing programs, but I think you said it right. I think it probably affects some other folks a little bit more so because they're more entrenched and they have a larger share of their overall business in federal.

Great. That should be it for me for now. Thanks.

Thank you.

Thank you. Erik Suppiger, Signal Hill.

A couple of modeling questions. First off, on the seasonal decline in Q1, is that decline all attributable to product downtick, or would we actually see some of the service revenues tick down as well?

I think if you look at service revenues, we have assumed they're comparable to Q4, and then we've assumed maybe we do a little bit better than that. But basically, we assume comparable to Q4. And then the product revenue clearly does decline, because that is much more directly correlated to the billings, which are lower in Q1 than they were in Q4, at least in terms of our projections.

And I assume ratable is fairly flat?

Ratable is flat, and we've assumed basically flat revenue, because I'm really not too interested in it anymore, because it's going to go away over time. So in my model, I'd just assume flat to Q4.

Okay. On the tax rate question, why did the tax rate bump up from fiscal '10 to fiscal '11? It's not much, but 32% going to 33%, I would have thought maybe going in the other direction.

It bumped up a little bit. We had, with the R&D, we probably were ultra-conservative. We had some additional R&D credits we could take from, really, it's actually from prior years, a way to help us by probably 0.5 point or so. So it's really, and maybe a little bit of just being a little quasi-conservative. But we're talking about a point, and that's not very much.

Can you review the seasonality for your free cash flow through the four quarters of fiscal '11?

Well, yes, but look at '010. Clearly, we had much better free cash flow in the second half than we did in the first half. The first half, we had lower than 40. And we had about 60 in the second half. I think -- I mean, Q1 will get the benefit of a nice increase that we had in AR for another year, which we'll collect in Q1, so we'll get the benefit of that. But generally, as I looked at my plan for '011, I've assumed pretty much the relative same growth, a 15% to 25% growth that we said for each quarter. But I did highlight the fact that there is some volatility because there are a lot of moving parts in free cash flow. And so [they carry the low ball] today from quarter to quarter. But I've assumed for right now, it's comparable to what we saw in '010.

What quarter do you think would be the largest free cash flow quarter? Would it be the March quarter or the December quarter?

Again, if you go back to '010, it was pretty equal between Q3 and Q4. And what happens is, it's a little bit lagged to growth in ARs. You grow AR, you get the benefit of that the next quarter by collecting it. So I'm assuming Q3 and Q4, again, this year will be pretty good relative to Q1 and Q2.

And so typically you have a hit from Q1 to Q2, is that right? Because your Q1 benefits from a big accounts receivable in Q4?

Yes, you have a little bit of a benefit, yes. So you have a little different Q2, I think that's correct.

Okay. And in looking at the guidance, last year's operating margins, they started off very low, I think in the low double digits, and then you averaged 24%, or you got up to 24% towards the end of the year. Are you expecting, as you go through this year, that you're not going to get back to that peak level, which is implied by the average for the year in the 19% to 20% range, or how can we interpret the progression through the year?

I gave the 17% to 18% in Q1, and I think that's probably good for Q1 and Q2. You probably need to get the average in Q3, and then go out of the year with rates maybe comparable to Q4 of this past year. And so that would get you to an overall rate of 20-ish percent, roughly. Around 20.

So you said Q4 of this year might be the same as Q4 of 2010 from a margin perspective?

I think the same, maybe a point or two. It could be 25%, could be 24%.

On our side, really, we tend to be more aggressively hiring this year compared to last year and to for future growth.

Okay, very good. All right, thank you very much.

Sterling Auty, JPMorgan.

Sterling Auty. I wanted to circle back to the revenue recognition change. Just to be clear, the guidance that you gave on the call, does that include or does not include the effect of the rev rec change?

Well, right now I'll say it excludes it, but maybe later I'll say it includes it. (Laughter)

And again, I think you had mentioned the ratable piece. I just wanted any further color that you can give. First, will it go into effect for the first quarter, so when that you report the first quarter earnings, will the revenue actually going to be different than the guidance just off the bat, because of the new revenue recognition rules, or will it go into effect effective for the second quarter?

Ken Goldman; No, it goes into effect for Q1, and we'll still have the ratable you'll see, but we'll also have some revenues on new billings that will be recognized based upon the new rules. So over time, ratable will go down because we're no longer going to follow those rules, and so that will go down because you're basically aging the deferred, and what will occur is the new revenue will -- any new billings starting 1/1/11 will be based upon the new revenue rules.

And when we think about the traditional split, the 62/38 rough split in terms of what is upfront, what's deferred, I know you mentioned the larger discount type of product being impacted. But if we think from a high level, that 62/38 split ultimately will turn out to be something different, then?

What will probably occur -- again, we don't have the numbers -- but if you think of directionally, the product gross margin should increase a little bit because you'd be carving out on certain deals potentially -- it all depends on the kinds of deals we do -- potentially a little less from product into deferred. And so product gross margin may go up a tad relative to any deals that we would have had to do a larger carve-out than normal. In terms of the areas where we ratabilize, like demos in China, that is probably no difference there, because what we're going to incur upfront will be pretty comparable to what we would have incurred over time.

But in terms of the revenue piece, not only the product gross margin, but when we think about the product revenue, just on the main sales and developed areas like North America and EMEA, we always thought about, well, you recognize roughly 38%, and 62% gets deferred and recognized over 12 months. How should we think about that split under the new rules?

I don't have a number yet, but you're absolutely right, that the amount that is recognized upfront will go up modestly vis a vis what gets deferred. So you're absolutely right.

And then I know you -- I'm going to try it, and I think I know what the answer is. I know you didn't want to put a number on it, but saying that the order of materiality, are we talking about maybe a 5% or less hit, or is it, potentially could you get, not hit, but (inaudible), or could it be negative?

My guess is it will be, at this point, I would say it's less than 5%, given what I know today.

Okay, great. One other question. I know this one's always been a tough one, so I know when you sell the appliance, especially those where it takes, let's say, all of the different services included in the first year. But I always got the sense that the renewals give you a better indication of what the customers are actually using in terms of the multiple services -- firewall, intrusion prevention, antivirus, et cetera. Is there a sense of, now that in the more, let's say, established customers, what the core feature set is? Is it firewall, VP, and intrusion prevention? Is it two services, three services? How should we think about that, and is it different in higher-end customers versus, let's say, middle market?

If you look at the total shipments, I'd say probably 70% or 80% customer will buy the subscription service because the intrusion service is a function beyond the traditional firewall, VPN, and some others like one solution or the others has. And so that's where the additional service, including antivirus, including intrusion, the UIR, and also an S-band there. So that's where, like (inaudible), they argue the firewall only. That's probably only because maybe 20-some percent of total customers. And there of the customers buying the additional service, I'd say probably about 80% of them are buying, they call the bundled service. That's when all the four service all bundled together, so because we have some price incentive, and also easy to manage if they buy all the service together instead of the one by one.

So of the four service, I'd say the antivirus were, the French had asked for us another competitor, which they don't have network-based antivirus in the UTM. And the second one, I say the intrusion and the UIR are also doing well. And S-Band is working in high-speed environment quite well. But some others also go to the product we call the FortiMail. That also tends to have a strong -- and is a function there also.

Okay. And the last question for me, Ken. As you think about the different things that you've slotted in, is there an obvious next technology to add to that bundle or to that services, or is this really the right set for 2011?

I'd say that probably the two newer products we are announcing in Q4, the first is really the new generation of FortiGate, and that's the fourth generation, so we had the first generation released in 2004, and then the second in 2006, the third in 2008. Now it's the fourth generation and also the first time we're using a new Forti ASIC chip, NP4. So that's working with a much better speed and also more function integrated together. Adding to the higher speed environment, it's better feeding to the cloud data center as well. And that's, we're ahead of our competitor, both on the speed and also the reliability and the carrier environment. So that is probably the area that will be growing quite, could be quite strong.

The three (inaudible) that also, we have two models introduced in the last two quarters. That's with a 3040B and a 3950B. Why provide like a 3950 provide 120 gig server port, and also the 3040 is where 2U (inaudible) performance off of the 40-gig support, fitting quite well into the other price environment. So as I say, these two part are definitely, I see some bigger customers already behind that to try to run averaging.

Great. Thank you, guys.

Keith Weiss, Morgan Stanley.

Earlier this year, you guys had talked about in particular seeing very good win rates against Cisco. Part of it's to be some operational issues that they were having and part of it's your products performing well again. Has that opportunity continued for you into the second half? Does that dynamic still exist in the same way it did in the past, or have you seen any change in the way that they're addressing the market?

So far, in the market, we have not seen much changing now on the Cisco, both on the product and also technology on the secure side. Even I earlier mentioned Moore try to address the security, and that was into the market. But from the part side function, obviously, we don't see much has come out of Cisco yet.

Got it. And this actually came from one other one, and how do you guys, when you guys start thinking about that ramp-up in the federal business, how does the current budget negotiations going on now, the talk of perhaps more austerity measures coming in, how does that (a) impact your thinking and (b) how should we think about the potential for that to actually impact sales in the first half of 2011?

Well, in some respects, it is a positive in that it means they will be looking at how best to control costs, and we think our approach is very sensitive to costs. The other point I'd make, which probably is just as critical, is our federal business is relatively modest right now, and we think there is only upside to what we've been doing in the past. And so it's not a material business, really, for us, if you look at our overall business. So my sense is, I'm not too -- I mean, maybe at the margin would lower the potential gross somewhat, but overall, we should have a big opportunity there vis a vis what we've been doing up until now.

That's helpful. Thank you, guys.

Michael Turits, Raymond James.

I know Sterling had asked this, but just on the rev rec, two things. Did you say that you will see those new rev rec rules in first quarter already? And did you also say that you thought it would be about a bump of about 5% for the year? Are those correct?

The answer to your second part is no, and the answer to your first is yes. The question was, would it be less than 5%? And I said yes, it would be less than 5%. That in no way infers 5%. And it is effective as of Q1.

Thanks. That's why I asked for clarification. Appreciate it, Ken. And then, again, do I have it right that on the guidance -- and I was jumping back and forth between calls, and tell me again if I got the numbers wrong again here -- but is the midpoint of revenue growth about 16% and the midpoint of billings growth about 20% growth?

That's approximately correct, yes.

Okay. Last quarter you talked about the fact that you had a tough billings compare, and you actually had it flip. You thought that going into this year that revenue might grow faster than billing. Now it looks like billings will grow faster than revenue -- not that I'm complaining. But why is that?

It leads to how we see the year unfolding in terms of split. You get, not taking in account any new rules, but looking at the split of revenue and billings between products and services and how much we see deferred in terms of long-term deals. And so I think we were very pleased with our revenue growth in the second half of '010, and I think we're just being mindful that maybe we did a little bit better than we thought we would do there.

Okay. And then on the cash tax rate, I had a guy on my team ask a question on the call because I had to jump, but is it probably 2010 to 2011, was your answer that you expect to pay about the same dollar amount of cash taxes in '10 and '11?

That's correct. I said it would be comparable. It turns out as we've done some homework here is we're really getting the benefit of the deductions of giving stock option exercises in some of our primary areas, whether it's in the US or Canada and so forth. So that is really reducing significantly, and we've done some reasonably good work and some tax planning as well that's reduced significantly the amount of cash that we would be paying for our taxes, even though it doesn't necessarily have a lot of impact on our pro forma rate.

And actually, I'm not going to give you an exact number, but our GAAP rate is still expected to be somewhat less than our pro forma rate in '011 as well, as it was in '010.

And Ken, what was it again? I apologize for not having gone through this statement myself here, but where was the -- I think you nearly doubled your guidance for your free cash flow in the quarter, and obviously you had net income up cycle. But what was the real upside of the free cash flow?

You mean in terms of why we went into the year with $45 million to $50 million in free cash flow guidance and we did about $100 million? Is that your question?

I thought you gave some guidance as of last quarter for this final fourth quarter for what free cash flow would be, and it seems like you came out way above that.

Yes, we did better there, too. A variety of reasons. One is profitability was higher, and so that feeds directly in. Collections were higher as well. And I think we were maybe a little conservative on what we thought we might pay for cash taxes, and so that worked out also better. I don't have it in front of me here, what our guidance was for Q4 -- 16 -- so yes, we did well.

Okay. Good work. Thanks a lot, guys.

(Operator Instructions.) Jayson Noland, Robert W. Baird.

Thank you. Just a couple of questions. On large deals, how distributed are those large deals across verticals? I assume service provider's the most prominent vertical of large deals, but how distributed are those across others?

I'm looking to my folks here, because I don't have that data in front of me. Do you have any sense of that? So what I heard is relatively well distributed, maybe a little more sales, because I did talk to a lot of examples in our service provider space. And those deals will tend to generally be larger. So I think with the caveat that service provider will have larger deals in general, it's probably well distributed.

Okay, so it's fair to say that the mix of large deals across verticals is similar to just your revenue base in general?

Relatively, yes.

Okay. And then last question for me. Ken, you had mentioned expansion into FS, healthcare and fed gov. Is that salesperson expansion, where you have a named account manager or a team assigned to a specific part of the market or a specific customer base?

Yes, you're absolutely right. In some cases, we have people onboard. In some other cases, we need to hire some people. But it is -- in the Americas, we are definitely split out in terms of enterprise and service provider, financial services as separate segments. We're a little bit more, not exclusively, geographically based in other countries. And so there, we've actually been addressing some of these segments I mentioned as growth opportunities better. In the US, if you really go back a few years, we were more of a channel based, addressing the SMB space. We broke out first federal, we broke out service provider, we then broke out enterprise, now we've broken out financial services. So we have continued to slice off areas that we think could really, with some focused effort, we can grow quite rapidly.

And, really, I think it also reflects the fact that our products keep on moving upstream and we're able to address markets that maybe a few years back weren't as obvious to be able to address.

Thank you.

Adam Holt, Morgan Stanley.

My first question is about the enterprise business. If you look at the enterprise deals, are they typically being -- what has been your experience relative to the ability to expand existing customers versus net greenfield deals? So if you look at the larger deals that you're doing, how many of those are back to base sales versus net new sales?

It's really a mix, Adam. We don't split it out, because it's actually not that easy for us, given we go through the channel, to exactly split out new versus old. So I don't have that data. But if I think about it anecdotally, I gave you a number of examples which go both ways, both absolutely greenfield opportunities versus the additional business at existing accounts.

As I think about '011, clearly in the Americas, there are areas like financial services which in many cases for us is a greenfield opportunity, and which does take some time to penetrate. But there are other areas where we have one business with accounts, and we do get repeat business as they expand. Typically, it's the best retail that might buy for a number of their retail facilities and then increase over time. So it's actually really a mixture, but I don't have hard data yet to provide you on that.

Okay. And just one follow-up on the enterprise. Then I had a couple of model questions. Ken Xie, I believe you suggested that you are seeing a number of standard firewall replacements within the enterprise. Does that mean that people are seeing a one-for-one firewall replacement, or is there a broader use case when you look at the kinds of services and functions that are typical of an enterprise footprint?

I think maybe one of the inquiries is a function beyond a firewall. That's where we have more advantage, so that's where you see the enterprise, really, the additional UTM function beyond firewall is a lot of case. We win the deal. If they stay, we extend the firewall and that's probably sometimes the existing vendor, they maybe still can hold it there. On another side, certainly, some service provider, we found it interesting also, starting to provide certain channel to the enterprise, because we were in the service provider. Mostly they're internal and also offered -- the service is offered there. So that's why we also found that's another way to get into big enterprise.

And are you seeing any kind of a standard footprint where three, four, or five different services are typical of an enterprise customer?

Not probably. Because we sell in like 80% that the service available units is a bundle, it's a little bit difficult to see which, what are the functions. We have four functions really under ours -- the intrusion, the URL and also an S-Band. But my sense is really the antivirus function and the intrusion function, that's probably the more needed in the enterprise environment.

Okay. Just a couple of numbers questions, and Ken Goldman, sorry to drag you through this one more time, but it was a little bit confusing on just what the impact of that rev rec change is going to be. So we know it's less than 5%, we know it starts in the first quarter. Should we be thinking about that as cushion to the guidance that you gave, or is it something that is excluded and you'll report it almost as a pro forma line on a going-forward basis?

The latter. Honestly, I'm trying to be absolutely transparent with all of you. In seriousness, very transparent. And I'm making it clear that our guidance is on the old revenue. We will provide both numbers so you'll be able to see our old and new, and there will be no -- nobody's dumping these numbers.

Got it, got it. Perfect. The second question is, if you look at the range of cash flow growth for next year, so the 15% to 35% relative to the ranges that you gave for billings growth and margins, where do you see the greatest opportunity for variability towards the low end of the range of that cash flow growth versus the high end of the range?

The goal is to do well in our operating earnings, and therefore -- I guess I would say it this way. If we make all of our operating goals, we should be at the higher end of that cash flow range. If we miss something like AR on a linearity business, which affect AR, too, that's really -- our business is really driven by billings and net AR, which is really the AR net of change and deferred. The rest of it is, is we manage expenses very tightly, so I don't expect us to have a problem there.

So really, it's only a couple of numbers that I have to focus on. And so as billings grow and it's making sure we collect that in AR. We do that, we get on the higher side of the range. If our AR goes up a little bit in terms of DSO, then I'm lower into that range.

One other point is really in last year, the hiring kind of a little behind the plan, so that's where we had a few reported in-house. We tended to do more aggressive hiring this year for future growth. So that also affected account payments.

Sure, but that's effectively in our operating margins, but yes, that's true.

Just two more questions from me. I believe you suggested that your renewal rates are now above 70%. Where do you think those renewal rates could go?

Actually, not a lot higher. In my own sense, as we get to around 80-ish or so, we get to there over time, that's probably as good as we can. Because you'll have projects that will, for whatever reason, get replaced, die, get cannibalized. And so there is a number -- you have low-end deals which, frankly, is hard to follow up, less than $1,000 accounts where the renewal is so small it's hard to follow up. And so my sense is we've increased that 5 to 10 points, depending on what base you want to start, on renewals, and there's a real thing with renewals is, we used to get to a good number in the 70s, but it took us a few quarters to do that. And now we get much more real time in the current quarter. So we've moved up the time we improve our renewals to that number, and then I think asymptotically, and we're looking around here, and we think probably 80-ish or so is probably about the max.

Yes, this is Ken Xie. I feel also this is more related to the model, the way we track and renew (inaudible). So even if the same customer and using the same infrastructure when they buy a new part are replacing the old part of fully putting that, it still was not a renew. It's really a new buy. And that's where, if you look at a product, it's more like other network equipment. It's really after four or five years, it's difficult to keep up the speed of the network speed growth there. So that's where the limitation is.

Another (inaudible) also some relate to the bigger environment. So if you look at the downturn in the end of '08 and '09, sometimes the customer now tend to be more hold on to the older part longer compared to buy the new product. That also have a little bit impact on the renewal rate. So that they hold on to older parts longer sometimes, the renewal rate can be a little bit higher compared to when they have more money, they can upgrade the infrastructure quicker. So that's the way we track renewal by product, and we will not quite strictly say it's really by customer. That's the model we make.

Got it. Perfect. One more question from me and one observation. Ken, you don't usually hear the word "asymptotic" on a conference call. So that's a first for me.

It's a big word, huh?

It's a big word.

From an ex-engineer.

The last question I had was on the -- and I think Todd might have talked, asked a little bit about this on the earlier call, which is the shift -- well, I don't know if it's a shift -- but the relative growth rate in short-term deferred revenue and long-term deferred revenue. It sounds like particularly some of the larger enterprise deals that that may be shifting towards growth in long-term deferred revenue. Is that something that continues through calendar '11, do you think? Do you think we'll continue to see long-term out-pays, short-term deferred revenue?

My sense is probably not. And the reason I say that is, again, products of a stated date where we were increasing off of a smaller base, and so as the base increases on long term, it's harder to grow that as fast. And so I think -- I personally, I don't have a specific guidance for that, so I can't go through the numbers -- but as I think out loud, I think those numbers probably tend to get closer in terms of year-over-year growth, just because of the fact that long-term is catching up and it would be off of a higher base going forward.

Perfect. Thank you.

Rohit Chopra, Wedbush Securities.

Thanks. A couple of questions for you, Ken. First, CapEx is a little bit lower this year than it was last year. Just want to get your plan for 2011.

Honestly, we don't have definitive enough plans to give you guys. The reality is our net PP&E has stayed below 7-ish. Most of the stuff we buy here is capital equipment -- technical equipment, I should say -- and a little bit of infrastructure. So I think, if anything, maybe it goes up to, we do have to increase some of our test equipment. And so I think maybe it goes up a little bit.

But it's so small. You think of net PP&E of $7 million growing your business up to $325 million in revenue and $375 million, roughly, in billings, it's just a small number in the equation. And the variability is just not very high.

All right. And then my next question is -- and don't crucify me for asking this question -- but you did really well in Q4. Q1's a little bit lower, so more seasonal. Is that the new normal?

I think, to the extent you do, we did better than we expected in Q4, that would accentuate -- I'm waiting for Adam to tell me it's another big word -- the seasonality of our business. And so, yes, I think we may have a little bit more seasonality between Q4 and Q1 in terms of Q4, the drive and so forth, and then starting fresh, if you will, in Q1. So maybe a little bit of the margin was seasonality in Q1.

All right. And my last question, maybe this is for Ken Xie, but I just want to get an update on partnerships. Maybe you can talk about things that you're doing with HP or maybe some other larger partners. Anything on the horizon that you can announce, or maybe people you're working with?

I think we do partner with some other bigger players, which we don't quite have the security piece there. We tend to enhance that, the partnership, because now security is an important piece of the infrastructure. I don't have much new update about a partnership. Definitely, we also hire someone ready to enhance the partnership in the area, but definitely we see more opportunity to grow more partnerships going forward.

Thanks, Ken. Thanks, guys.

Alan Weinfeld, Kern Suslow.

There hasn't been much discussion of the current organization of the sales force. I know you've talked about vertical markets, but what percentage of the current revenue is going through the channel compared to last year?

Just to reiterate, virtually all of our revenue goes through the channel. So in some cases, they may facilitate the business, where we do all the direct selling as well, but it only facilitates their channel. There's other business which is really driven and fulfilled by the channel, and so that's where more of a turnkey system. But in general, with some minor exceptions, all of our business globally gets fulfilled through the channel.

Even some of these large enterprise deals?

Generally, yes. Generally, deals will go through the channel, even though we're doing all the work. There are a few exceptions, but not many.

That's a little different.

We try to minimize conflict with the channels, so we try to, even if there may be deals where they don't get very much of the margin, but nonetheless, we try to minimize conflict.

And it looks very consistent year over year, quarter over quarter, the product segmentation as far as the small, mid, and high, and also very consistent beyond the geographies, revenue by geography. But what is not the same is the growth year over year by geography. So I'm trying to make sense of that. I know we're not having a robust economic rebound in America, but why do you think you had that 45% growth that sticks out in America versus a much faster growing scene, A-Pac, where you had very strong 25% growth, but you just really had an amazing quarter.

Yes, I think in America there's two things that really is helping. One is we put generally a focus on growing Americas because our relative market share in Americas is definitely less than particularly throughout Europe. That's one.

Two is we split out -- again, I described this a little bit before -- two businesses, the service provider and the enterprise. And if you really go back a couple of years, they were very, almost negligible. And as we increased the billings in those areas pretty dramatically, that's now translating into revenues in America. So Americas is definitely getting the benefit of that.

The other thing I pointed is Americas is doing pretty well in Latin America, so that's -- even though that's not US, it is part of the Americas -- and so that is growing quite well additionally.

And so it was really those factors, the enterprise split-out and the focus on that, the service provider and all the things we talked about in terms of the cloud and so forth driving that business. Geographically, nice growth in Latin America. And then fourth would be coming off a relatively smaller base in terms of market share.

And you keep talking about cloud, but it seems your device is always on presence. But I guess, are you talking about delivering the services which are a dominant share of your revenue, the service for blocking on the intrusion prevention, or capturing all kinds of bad stuff in the Web and email security? Are you talking about those coming from?

It definitely would be growth in the Internet. And again, the appliances would be supported in the data centers of the various service providers as they provide that service over the Internet for their customers who are providing that -- if you think about the customers that are providing that service to their customers. So that's the way it's working. And so that is -- what's really driving it is the proliferation of the Internet and the amount of commerce and so forth -- e-commerce, whatever -- going through the Internet.

Thank you.

(Operator Instructions.)

Going once, twice. So I'm looking, and I don't think we see any more questions. So unless anyone has, if they do a star-one real quickly, I think we will finish the call on that. And I assume, operator, you don't see any more questions? So, Alan, correct?

Yes, sir, I'm showing no further questions.

Okay, so why don't we just finish it up there. I thank you all for listening to the first call. And again, I know we go through a lot of detail. Perhaps we can find a way to prune that at some point here, but we do think it's good and it's transparent relative to providing data. And then thank you for, on this call, where it's much more of a, obviously, a Q&A session. So thanks until we see you -- we'll probably see many of you at the RSA coming up in a couple of weeks. So we look forward to that, and we'll have a major booth there, and Ken and I will be there for at least two of the days.

Okay, thank you.

Thank you all.

Ladies and gentlemen, thank you for participating in today's conference. This concludes our program for today. Now all disconnect, and have a wonderful day.