Fortinet Inc (FTNT) 2015 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome your Q2 2015 earnings financial analyst Q&A.

(Operator Instructions)

As a reminder, today's conference is being recorded. And now I will turn over to your host, Michelle Spolver.

Hi, everyone. Thank you for calling in again. Ken Xie and Drew Del Matto are accompanying me on this call. Before we start -- it's an all-Q&A session. Before we start, I have to remind you that -- remember the disclaimer I provided during our 1.30 call earlier applies to any forward-looking statements we make during the second call. With that, John, you can go ahead and open up the call for questions.

(Operator Instructions)

Gregg Moskowitz, Cowen.

A couple of follow-ups. Drew, obviously your growth in seven-figure deals was very good. Can you say if they were any kind of mega deals in the quarter, sort of in that $5 million or above type range?

Not a lot. I can't think of any off the top of my head. Maybe one.

It is not skewed like that. The customers tend to buy over time, I think is how I would more characterize it, instead of front-loading purchases.

Okay, perfect. Was also wondering if you guys are looking to bring on a new CMO, and if so, if that was likely to be an internal or external hire? Any color there would be helpful.

Yes, we are actively recording interview with a CMO candidate. We also will keep advising in the marketing and we're hiring other level. Actually, you can see the investment we had in the last few quarter, in the marketing, actually, is starting paying off, so it's a lot of [legal] going on right now.

Okay, great. Thanks, Ken. If I could ask one last one.

This is the second consecutive quarter in which you didn't buy back in a stock after having done so in the prior five periods. Just wondering if you could update us on your buyback philosophy. Thank you.

Sure, Gregg. It's Drew.

We are very focused on our growth. That's our strategy right now, is to invest in growth.

As we said, we'd be opportunistic on buy back. Obviously, it is harder to find that, but you look at history think our average price of the roughly 79 million shares we brought back under the current program was -- I think the average price was like $20.50. But right now, I would say again, we'll obviously remain opportunistic. Hard to define what that is, but we are clearly more focused on growth than buy back.

Okay. Thanks very much.

Sterling Auty, JPMorgan.

I wanted to drill into two areas, Drew. First, the inventory turnover, the 2.2 times, now that some of the port items and other things that were drags or concerns on inventory are behind us, is there anything that should prevent that from trickling higher? I know your stated guidance is better then two, but would be the puts and takes that would either help that number elevate in the back half of the year or keep it constrained?

Sure. As you've said, I will stick to the two or better. I think on the puts and takes side, it is really about forecast and timing and timing, if deals were to come forward or slip, I think that is what really drives it one way or the other.

We are comfortable with where we are, and that's certainly not to imply that anything pulled into the quarter or anything like that. I wouldn't want you to take that away from that. That is not the case at all. I think -- I like -- 2.2 feels like a very comfortable number for us. But again, I would guide to two or better just to -- you can't -- it is hard to be that precise on forecasting inventory needs.

Also -- Sterling, this is Ken. We may have a little different business model than some of our competitors. The most [server, understand] server PC because we custom built, I think the FortiGate is the biggest, the super-majority of the business we have. FortiGate is a moderate (inaudible) we have the customer built with [ASIC] so we deal on the chip level to the system level toward the software obligation level there.

That is where we -- arguably would need to be because the lead times is longer and because there is no ready-made system for us so we had to build out from a chip to the system to customize for the solution we have. That is where we need to keep some long inventory. And also, the inventory cost were low, what are the interest whatever, so that is where we compare to -- so we have to model compared to the opportunity lost. If we're short on the inventory, also compared to some inventory costs, so we feel two is a moderate pull, so that's where we're keeping that as the model we have.

Got you. The second topic is, I think Mike Turits asked you about directionally, just cash flow growth and should it grow in line with net income. I didn't quite catch your answer.

I guess I would be surprised if it grew in line with net income, because just looking at the billings guidance and the revenue guidance, just the contribution you should get to cash flow from deferred revenue, shouldn't that make it a much higher growth rate and cash flow than net income?

I think you're right. That is a good answer.

I was thinking of the growth in billings and basically modeling changes in working capital. I guess I maybe have perhaps misinterpreted his question, but that is how I was thinking about it.

The top-line growth being -- I thought I heard him say top line. Maybe I'm mistaken, but that is how I heard it and I meant more on the billing side. So following the billing forecast, it grows with billing. That is the point right now where cash flow out paces the P&L and you see margin compression, simply because you defer so much.

Exactly. Okay.

I see your point. I think that is right on, Sterling. Yes, that's right on. My apologies if I didn't answer that correctly.

All right. Thanks.

Saket Kalia, Barclays.

Drew, I just wanted to zero in on FortiGuard and FortiCare billings. You mentioned in the call you had a tougher comp there this quarter but you still managed to accelerate.

I guess just to dig into that a little bit more, was it a particularly good renewal quarter? Did maybe weighted average contract duration coming higher than you expected? Any color on that strength there would be helpful.

No, I -- look, just as I always remind people, an easy way to see the duration is look at the percent of long-term deferred as a percent of overall deferred. It has been ranging, I think, roughly about the same, 32%, 33%. It has not changed in a while, so it is pretty consistent duration.

In terms of renewals, no, it is not an abnormal quarter at all one way or the other. We don't allow people to pull in renewals, for instance. We wouldn't pay on it.

Okay, got it. It sounds like better attach rate and perhaps a little bit of the pricing impact maybe having an impact in --

A little bit of that, but I think good execution all around, people definitely having a nice take on the bundles. It does -- I think clearly enterprise customers are clearly buying the bundles and we're doing a lot of enterprise deals.

Got it. Not to revisit the Meru question again, but I guess just curious, is it fair to think about as the majority of their support revenue being written down post close? And then similarly, I guess as you integrate those sales teams for much more of a combined effort, how do you think about any potential expense synergies?

First of all, they have been through quite a bit of disruption over the last year. I think they have had multiple restructurings and they were still losing money. Again, we're trying to pull it into fit our model.

We are being very selective strategically in terms of our strategy, which is to basically drive a secured Wi-Fi offering. And then basically cross sell our product, FortiGate and FortiGuard and FortiCare products ultimately into their install base.

So yes, we're rationalizing what is there. Not everybody came over. We're clearly making sure that we take a step back and ensure as -- just in good hygiene as we do, we are every quarter looking at what makes sense for us and what doesn't make sense.

And we are really, as I have said, is an integrated business. We're really stepping back and looking at the whole, what does it take to be successful in the future versus going out and trying to cherry pick off things here and there. That is the right way to do an integration, make sure you do not throw the baby out with the bath water, so to speak. And make sure that you get the value out of the case you're looking for, which is an integrated wire -- an enriched integrated wireless solution.

Also, the supporting revenues of our a small percentage I think is probably 10% to 20%, is because they only have this supporting for the deployment or whatever, or operator at the back fix. They don't have any subscription like we have on the --

Yes, we're not counting on that at all. I think, look, they're going through the opening balance sheet stuff right now, the accountants, and they'll figure that out, but obviously, that, as you said, gets hair cut.

It's immaterial, yes.

Got it. Lastly if I could squeeze it in, I guess Ken, you had mentioned the new content processor over the -- coming over the next 6 months to 12 months. How should we think about that conceptually in terms of rollout by appliances. Is this something that starts at the low end and works its way up to the high end or does it start of the high end and work its way down?

It's more -- I think the network processors, most start in from the high end and then [ISOC] most starting from the low. The comp in process pretty applies to the high to the low to mid range.

It is more follow what is the next whatever the FortiGate product going to come out so we just replaced the CP8 with a CP9. That's improving the performance a few hundred percent on the content on the scan on SSL. That is probably -- the CP9 tends to be -- is a coprocessor next to the CPU. It's compared to the MP's more accelerated interface level and ISOCs more apply to the lower end.

Got it. Thanks very much, guys.

Michael Turits, Raymond James.

Couple questions now regarding things like duration, or big deals. I think what we're trying to get to is whether or not -- this was obviously huge quarter.

Is there anything that we should really think of us have been exceptional in this quarter? Another way thinking of it, is there anything, can we then think of, given what you just did, should we think of normal seasonality going forward or is it not? Is it somewhat exceptional?

Not huge deals, really. I think the overall performance on enterprise sector there, so it is really like we did last quarter, 70% this quarter [in earnings] the team starting to get on board and the more matured as the same time we were keeping investments in the market and [some other] helping.

Also, I see that the internal segmentation, that's under, that's also right now to become a sweet spot for the enterprise, because a lot of enterprise also see the need for the internal segmentation. So the internal segmentation, that is probably the one that is new, but is -- we don't have much data. I don't see any other data for [animus] yet but we fill the big enterprise really badly need this kind of solution to secure internally.

Yes, Michael, and there is just a good standard distribution across the deal side. So it is not -- we answered the earlier question. There is not a large deal or two driving the growth at all. That is not the case.

Think of it, we added 9,000 customers, I think, in the quarter. It's very consistent, it's a very run -- it seems to be a run-rate oriented business and we seem to pick up steam as the sales reps come up to productivity and they just, quite honestly, continue to over perform.

In theory, normal seasonal -- normal quarter-to-quarter seasonality should apply?

Again, you just have -- yes.

There was nothing -- no one-offs or anything this quarter that would change seasonality for next quarter. And then the other thing I wanted to highlight is, deals, while we didn't have any big mega deal or one-off deal, the deals for the $1 million range, which we've never actually given the percentage before, but we saw really good growth in the $1 million and above.

So more $1 million-plus deals than we've ever done. It was 133% year-over-year improvement. But again, there wasn't a big mega deal that drove that.

Okay. What was the status of the ERP implementation? I forgot. How much is that going to cost over what period of time and is that -- you're non-GAAPing that out form expense point of view, I believe, but where are we seeing that on the cash flow statement? Is it CapEx, or is it --

It's in CapEx, yes.

It's all in CapEx?

There is some in OpEx, too. Basically, the non-run rate portions is the pro forma piece or the non-GAAP piece.

Right. Because if you are non-GAAPing it out as the note says, some of it has to be (multiple speakers).

You basically have an investment and some consultants that goes away when the project goes away. (Multiple speakers) Design work.

(Multiple speakers) when is the project over and how much is in CapEx, how much is in OpEx, over what period?

We should be over roughly around this time next year and then I think we said $40 million to $45 million for the year in CapEx. How much for ERP?

For the year, $5 million or $6 million is in this year. In the CapEx piece.

$40 million to $45 million is your CapEx and $5 million to $6 million of it is the ERP?

Yes, $5 million to $6 million of the $40 million to $45 million is ERP in FY15.

How much is OpEx that's not being capitalized that's being non-GAAPed out?

About $5 million.

Another $5 million?

And a half. It is about 50%/50%, to think about it that way.

APAC, review for us why the growth has been so much slower in APAC? What is your China exposure and is that any part of the issue?

China is very small, probably less than a couple percent so that is where -- and also in the other part of APAC we feel we need to highly invest in growth quickly, so that is slightly behind. That is where we need to keep amassing growth in APAC. But China's a very small part of it. I do not think it has any material impact to us.

Okay. Great. That's it for me. Thanks.

Walter Pritchard, Citi.

On the G&A expense, looked like you had pretty significant uptick there. I think we're non-GAAPing out the ERP, but was there anything driving that specifically that should be highlighted?

Yes, probably. We had just a small spike in legal fees probably this quarter.

And are those -- do things continue at that level?

Hard to predict, but I would just -- I think it is probably fair to hold the percentage -- G&A percentage where it is.

Can you tell us what matter that relates to?

Just multiple -- we've been running through them. I wouldn't highlight anything particularly, Walter.

Okay. Just on the -- you highlighted your average buyback price in the low $20s. You mentioned you're using your cash for growth.

I guess the way I look at it, your funding -- you're certainly able to fund your growth organically, you're profitable, you're not burning cash from operations. So as we think about using cash to fund growth, such a substantial amount on the balance sheet, is that a cash used for acquisitions or -- it just doesn't seem necessary that you would actually hold the cash to fund growth, given profitability levels.

Look, again, we're focused on growth. That is our strategy. We could use some of it, obviously, for M&A, if that makes sense, Walter, but again, we're fundamentally focused on cash -- growth, excuse me.

Okay. Just lastly on sales and marketing spend, I think Sterling asked this on the call -- on the first call, it feels like your spending actually decelerates a bit in the fourth quarter based on your guidance. We note that your sales and marketing spend a year ago really picked up in the fourth quarter and there is a bit of -- your anniversaring that now at this point. Should we -- can you just talk us through how you think about sales and marketing through this year and is that part of what looks like a more level OpEx level between Q3 and Q4?

As we said, I think the self adjusting side of expenses obviously are the COG side and also commissions self adjust with the number a little bit. We don't -- we continue to invest and we're sticking to a the 14%.

Okay. All right. Thank you very much.

Jayson Noland, Robert Baird.

I wanted to follow up on the internal segmentation firewall, Ken. It has been a topic we have heard about from the industry for a while. It does seem like it is very nascent. It is typically described as software to monitor East-West traffic. I do not know that I have heard of an appliance, certainly not a custom ASIC-based appliance given the throughput required, but maybe if you could detail that a little bit more versus a software-only approach?

The software only also (inaudible) detection. That's really in the typing mold, is not an inland device, so they can detect. That is where so far internally they do deploy a lot of detection device, detection software on the server, but as firewall is the only device inland can take action, can do the prevention.

So that is where a lot of a bridge or whatever, the company they do have the detection like a Sony, like [Audi] deploys the latest software detection, that there is too many (technical difficulty) cannot do anything about it. The keys really have the inland device take the action to prevent an attack and also deploy internally like a security department, whether the finance or engineer or cells or whatever.

It is an internal segment, given department also protect the server. That is the relatively new and it is an inland device. It is not just the detection type device. So that is different than before.

You know we once have to match the key property of the internal network speed. You cannot slow down the internal traffic and also has to be reliable enough that don't bring down the network. So that is different than the traditional detection and also different than the perimeter security, which is connected with internet.

But internet travel is much, much slower, like we say, it tends to act slower in the internal network speed. So that is the challenge, need to be very fast, need to be inline and also need to be low latency, especially for the server.

Okay. Is it logical that a company could have more internal segmentation appliances than a perimeter-facing firewall?

Still the cost, I have to say, today, if you look at have [whether per] gig, the firewall cost is 50 to 100 times higher than the switch. So it is really, some company, (inaudible) they are starting to put more internal (inaudible) but it is also more dependant on the budget and also dependant on how they want to manage it together.

I'd say, if they have more budget, they probably definitely will -- some big enterprise we are starting to see implement also sometimes mandatory to the internal segmentation but some other enterprise not there yet. It is more budget. But this is new. This is really the (inaudible) new. We don't see it before to deploy internally and we're high-speed inland device.

That is interesting. And then a follow up on a couple of the sales exec hires. I think the gentleman from NetApp is going to be running the finance vertical. Would he have counterparts then in tech and healthcare also?

He will mostly focus on enterprise and especially to meet enterprise, all this together. That is the area he knows the best.

He was additive, he was not replacing a previous salesperson?

Yes it, you are right. We need to keep adding sales capacity and at the same time keeping improving our model and also keeping improving the marketing and the lead-in pipeline. Also growing from a today $1 billion to a multi billion-dollar, that is also we need to keep improving a lot of the process in the system on the team level.

Okay. Thanks a lot.

Fatima Boolani, UBS.

Just a few on go to market. With respect to go to market in the Americas, the approach to verticalization has been very successful. And I am curious, given the mandate to build sales capacity internationally, to what extent you're actually thinking about porting over the US model internationally and if that would make sense? And then I have a couple of follow ups.

Probably -- what depends -- in some country like some big country in Europe, we may starting into some verticalization and divide by the vertical. It really depend on which model is more efficient. Some smaller company or region, they probably based on the regional model that we did early may be better but some larger, we may start to divide into the vertical. It is difficult to say which one, what we're --

There is global coordination on the verticals, though, is a good way to think about it. They are regionally based teams, but they work on specified global accounts. I think as you get down the list in terms of size in any of those verticals, that is where you have more the regional orientation.

I agree, like (inaudible) we are starting to have some global accounts and that will be more managed by --

Yes, the big names are global.

Fair enough. I know there was also an emphasis around building out to the mid-market and channel-focused investments and you mentioned that you did make some additional hires in that space. I'm wondering what's on their road map.

Is it more marketing event, is a more pipeline building activities, additional discipline around building or improving forecasting in the mid-market? I just want to get a sense of how exactly you are building out the mid-market in your channel relationships with your recent investments.

I think it's all that you mentioned and also we're hiring at all levels, including the CMO. We actually, we're recruiting and interviewing right now. We feel the marketing would be very important for the mid enterprise in the keep investing in this area. That is where we are behind with some of our competitors.

I would just add that it is probably a slightly different sales interaction and marketing interaction, even sales engineer action -- interaction in the mid-market versus the top end were you're doing a lot of testing. I think reputation in the mid-market is going to matter more in the mid-market. We want to make sure we are developing, we are seeding that reputation and we are growing that and then making sure we're communicating with people in the right way in that market. It tends to be more like that, some of the, I would say, probably higher level of social marketing, if you will, then perhaps at the top end.

The channel, [the Palmer] play a more important role in the mid enterprise market there. At the same time, we also see, I think like I mentioned a couple of weeks ago, we announced a couple new [porter] in the mid range. There's a few more come out in the next few months. That is all helping enhance the mid-market growth.

Does that imply you're sunsetting your historical midrange SKUs? I imagine you're referring to the 400D and 900D that you announced a few weeks ago, but curious what happens to the existing lineup and if you're kind of sunsetting those then rolling the Vs into the lineup.

Yes, the Vs are probably the latest version. That is where some of the middle range is not quite using the latest MP6 yet, so where we're going to use MP6 more in the middle range and come off some new part out there so what accelerate the performance and also helping position even stronger in the mid enterprise.

Last couple from me if I may, just a regarding the bundling. You have talked a lot about that 10% price increase that was in effect as of January.

Presumably this was for the new customers that you are on boarding but I am wondering how the price increase process works for existing customers, if you are transitioning them in the bundle direction where previously they were just buying a la cart services? And if there is a dollar net extension happening at time of renewal? With existing customers.

It is probably more of a system that we have before. If you look in the UTM bundle, we offer full service, from anti-malware and the intrusion include app control and also anti-spam and the web filter. Each is about 20% and then also we had (inaudible) supporting another 15%.

If we look on the previous UTM boundary, you have full service, each is about 20% and then add 15% on the supporting, as we're only really selling for a 45%. That's way low, so we change it to 55%. It is more fixed on some things. We're not quite bundled the right before.

But like I said, we more focus on offer improving the service so that is what ATP and also the new mobile security [renew] eventually will help customers add additional value so instead try to change some of the price in the past.

Let me boil that down. I think, look, the small tailwind in Q2 for that action, I think if you're talking about going forward -- your question was on the install base. What I think you're basically asking, what so they renew at? They do get the increase.

Your ability to get that, though, is somewhat challenged. A lot of people lock that piece of the budget in, so it really takes a while to cycle through that. You are doing it more on new customers than renew, even though that is what the quote will say where that ends up maybe slightly different place, though, quite frankly.

That is a little bit of why you end up with a minimal impact near term. It is good to think of that longer term.

And then yes, I think to the extent that if we have 235,000 customers, they don't all have bundles. That's something we clearly look at as a growth vector going forward in a way to ensure that when we see a renewal or we see a customer, we open that door to sell them something else.

A very last one for me, if I could. Just wanted to clarify again that you are expecting to do another round of increases on prices come back half of this year. I am just curious as to how widespread this will be within the portfolio.

No, were not planning any price increase in the novice plan. We feel the current price is reasonable.

What we're going to offer is additional service, additional distributing service, like a mobile security service. So a customer, whatever they have the current on the price you can buy it, we're not increasing and just additional new service. That is also could be a different bundle, whatever the bundle they have today we're keeping there and also same price.

That's it for me. Thank you very much for the color.

Melissa Gorham, Morgan Stanley.

Just one quick one for you, for Drew. I'm just wondering in the extent that you are selling more virtualized appliances, is there any implication from a gross margin perspective that we should think about as you ship more virtualized form factors versus hardware?

Obviously, software has a higher margin. It's not a material part of our business, though, Melissa. That is the problem. Again, to remind everybody, we've guided for the year, so the [14] is the right number.

Theoretically, you are correct. The problem is just on a major part of the top line yet.

Okay, great. I wanted to quickly drill down on the public sector, given that we're getting into the September quarter. Today you announced another certification with the public sector. Just kind of broadly wondering if you can give us an update on how your government business is performing and what the opportunity is there?

Sure. I think the OPM breach highlights a bit of the opportunity.

I think we hire a vertical team there, a smaller investment than in the healthcare and financial services, because we already had a government team, or fed team, but we augmented that with a new leader back in, I think, October or November of last year. That team is doing better. We've given guidance for Q3, obviously, and obviously September being the year end, that is reflected in there.

We do feel like we are getting more looks anyway from the government. We ask this question as a team, obviously. It is an obvious question and they felt they were getting a better presence and better attention in the government than they had in the past. And clearly OPM piece being a part of it.

I think part of the comment they made was that there's a lot of the longer-term legacy providers installed there, and they seem to be at risk, or more at risk than they have been. And there was a high resistance to change in the past, but clearly the breach, and I think this is why the breach matters. I think it stimulates some rethinking of the architectures.

Okay. Thank you. That's it for me.

Catharine Trebnick, Doherty.

Can you give us a service provider split out from the earlier call?

Catherine, you want the vertical -- the percentage of --

Yes.

It's 22% for Q2. I think 23% Q1. I think it's -- it's probably, in the last few quarters, around 22%, 23%.

Q2 was 22% of total billings.

Okay, perfect. My question is, you did a lot of discussion on the internal firewall and what I'm trying to understand in that aspect, as you invest in your R&D, are you looking at identity and management as another product extension? Complementary to where the internal firewalls are?

Some of that, but I feel the internal more try to start the intrusion of malware as not quite access. Access a lot of time still goes to the perimeter firewall but it is internal is really more high-speed, is how to prevent some of the malware, some of the -- it's more like a more leakage prevention kind of sense. It is not quite in the -- the management of them is very important so we have good partner with some other vendor. They do a lot of data analysis and the management and that is where we're working more closely with them, but is not quite in the authentication or access control yet.

Okay. That's helpful.

Earlier in the year and last year, there was a big issue with the port contracts in the LA area. At any point did you switch some of your shipment directly to Europe and not to the US?

Yes, that's probably -- we planned to pull in some inventory I had over time cost to Q1 inventory to be a little higher but over the year, the issue has calmed down now. I think it's very normal now.

Okay. Thank you very much. Good (multiple speakers), guys.

(Operator Instructions)

Sterling Auty, JPMorgan.

A quick follow up. When you look at the guidance for the September quarter, I think we're looking at some acceleration in terms of the revenue lines. Just curious which of the revenue lines should see the biggest acceleration, the subscription area or the product area?

We don't guide that way, Sterling.

All right. Thank you.

Ladies and gentlemen, that concludes our Q&A session. I'd like to turn the call back to your hosts for any concluding remarks.

Thanks everybody for your attention and, again, your support. We really appreciate it and look forward to talking to all of you soon.

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