Radware Ltd (RDWR) 2012 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Radware Limited fourth quarter results conference call. At this time, all participants are in a listen-only mode. Later we will conduct a question-and-answer session and instructions will be given at that time. (Operator Instructions). As a reminder, today's conference call is being recorded. I'd now like to turn the conference over to your host, Mr. Roy Zisapel, President and CEO of Radware. Please go ahead.

Thank you. Good morning, everyone, and welcome to Radware's fourth-quarter 2012 earnings conference call. Joining me today is Meir Moshe, Chief Financial Officer. Meir will start the call by reviewing the financial results and afterwards I will discuss the business highlights of the fourth-quarter results. After my comments we'll open the discussion for Q&A. Meir.

Thank you, Roy, and welcome, everyone, to our fourth-quarter conference call. First I would like to read you the Safe Harbor language.

During the course of this conference call we will make projections or other forward-looking statements regarding future events or the future financial performance of the Company. We wish to caution you that such statements are just predictions and that actual events or results may differ materially including, but are not limited to, general business conditions and our ability to address changes in our industry, changes in demand for products, the timing and amount of orders and other risks detailed from time to time in Radware's filings. We refer you to documents the Company files from time to time with the Securities and Exchange Commission, specifically the Company's last Form 20-F, filed in March 2012.

And now, ladies and gentlemen, for the financials. We are very pleased to report an additional record quarter and additional record year of revenues and profitability. Revenues for the fourth quarter totaled to a record of $49.8m, representing 5% sequential growth and 11% year over year. Revenues for 2012 totaled to a record of $189m, an increase of 13% compared to revenues of $167m in 2011.

Non-GAAP gross margin remained at 82% in the fourth quarter as well as for the whole 2012, compared to 81% in 2011. The non-GAAP net income this quarter has increased to $11.2m, or $0.48 per share, compared to income of $9.5m, or $0.42 per share, in the fourth quarter of 2011. The non-GAAP net income for 2012 has increased to $40.5m, or $1.7 per share, an increase of 32% compared to a net income of $30.7m, or $1.3 per share.

Stock-based compensation expenses in the amount of $1m, amortization of intangible assets in the amount of $800,000 and exchange rate expenses in the amount of $300,000 bring the GAAP net income this quarter to $9.1m, or $0.39 per share, compared to net profit of $6.6m, or $0.29 per share, in the fourth quarter last year.

Non-GAAP operating expenses reached $30m in the fourth quarter, bringing our non-GAAP operating margin to a record of 22%.

The headcount for the end of the quarter was 800 employees.

During the fourth quarter we generated cash in an amount of $10m, bringing our total cash generation in 2012 to an amount of $56m. Thus our cash position, including short-term and long-term deposits and marketable securities amounted to $275m and we have no debt. Shareholders' equity amounted to $271m.

Guidance for the first quarter. We expect revenues to range between $48.5m to $49.5m, 82% of gross margins. Operating expenses will range between $30.1m to $30.3m. Financial income at $1.4m and non-GAAP EPS to range between $0.43 to $0.46.

As you can see, ladies and gentlemen, revenues are up, profitability continues to improve, cash is up, we expect higher and better results in 2013. And now I'd like to turn the call over to Roy.

Thank you, Meir. Our fourth quarter results reflect another quarter of solid performance. The North American market continued to grow significantly for us, closing the year with over 30% of annual growth. We are seeing a lot of activity in cloud data centers, data center consolidation projects and Cisco ACE replacements, yet the most significant increase in demand in pipeline is coming recently from our Attack Mitigation Solutions, given the ongoing waves of attacks on financial institutes, government agencies and online businesses.

We believe these growth solutions are expanding Radware's addressable markets on the traditional load balancing and application delivery use cases to becoming the control plane and critical component in enterprise data centers and carrier infrastructures.

On the ADC front our value solutions run from physical application delivery controllers into virtual instances that can reside either in a dedicated hardware platform or in a software-based virtual appliance on top of VMware and KVM, or as an instance on top of a multi-instance consolidation platform that we call ADC-VX.

We are leading the industry in the density of instances per platform we can provide, which directly impacts the cost of a carrier or cloud provider to serve the customers. We also lead the market in offering application delivery instances across all platforms and capacity ranges, enabling our enterprise customers to cost effectively consolidate their infrastructure while providing better availability and full turnarounds for their IT operations.

During the last several quarters we considerably enhanced our value solutions with Radware ADC fabric, which is the first solution to pull virtual ADC instances across multiple platforms and data centers. The result is improved application-side agility, mobility and bursting within a single data center, across data centers or to the cloud.

We also introduced AppShape that provides an application perspective for shaping the data center infrastructure to specific application needs. AppShape dramatically accelerates the rollout of new business applications and services, which can be integrated into virtual data centers and cloud ecosystems. We have released AppShape models for leading applications such as SAP, SharePoint, Exchange, Oracle and others to drive agility and simplicity of the application delivery fabric.

An important proof point for the superiority of our ADC fabric and automation capabilities comes from our joint integration of VADI with VMware vFabric Application Director. Through this integrated solution, whenever a customer uses VMware Application provisioning solution, automatically an [AppShape] instance can be initiated with the corresponding AppShape template already deployed and preconfigured.

This enables our customers to fully automate and simplify the management of software-defined data centers and provision a complete virtual application suite silo with one click.

Switching to the application security space, our Attack Mitigation Solution continues to prove its unique capabilities in blocking major cyber attacks against our customers' data centers. We believe we offer, if not the only solution, then one of the very few that can address the volume of threats that we've seen recently.

Our November industry survey revealed that 65% of respondents experienced at least three denial-of-service attacks per year, with downtime amounting up to $100,000 per minute. Our most recent research report, issued last week by our Emergency Response Team, demonstrates that attacks lasting more than one week doubled in frequency during 2012 with 70% of those attacks categorized as advanced persistent threats, [versus] 30% in 2011.

Our solution is unique in the fact that it combines multiple security technologies, including denial-of-service protection, intrusion prevention, web application firewall, network behavioral analysis and reputation services, all together to protect against these complex data center attacks.

In the fourth quarter, we announced our new high-end DefensePro product, with the ability to handle 25m packets per second of attack traffic, regardless of packet type, as well as up to 40 gig of legitimate traffic. The DefensePro x420 offers the world's highest mitigation capacity and is designed to protect organizations from the industry's highest volume of denial-of-service attacks. This new product is between three to four times faster than our current top-of-the-line product and creates larger opportunities in the carrier market for us.

During the fourth quarter we were awarded a contract to upgrade an existing network protection system for a leading US national cellular network provider. This current contract award is part of a multi-million dollar enhancement to its capabilities in order to provide secure internet access and mitigate denial-of-service attacks on internal customer-facing networks throughout multiple locations in the United States.

Another win we had this quarter is Brinkster, a leading cloud and hosting provider. Brinkster offers hosting and cloud computing services to over 50,000 customers ranging from small start-ups to large corporations. As Brinkster business grew, it had to protect its infrastructure against incoming hacking attempts on its servers, as they risked not only the customer under attack but all of the customers that operate in the cloud data center.

In addition, Brinkster had the need to protect its infrastructure against attempts from hackers to use its cloud servers to launch outbound attacks on other entities. We believe that in the coming years every cloud and hosting provider will be in need for a similar solution. And we are very pleased to see more and more business opportunities in the segment as we have a unique ability to identify and block abnormal traffic behavior.

To evidence the way these widely reported attacks against financial institutions were launched using servers at hosting data centers. These new server-based and not client-based or PC-based botnet attacks represent a new risk and liability for hosting providers and we expect that to increase in 2013.

The Brinkster win joins the win at [Siros] that we've announced in the third quarter and additional wins we have in this segment that are not public.

To summarize, while the macroeconomic conditions remain challenging, we do believe that the virtual in-cloud application delivery market and the application security market can provide ample growth opportunities for us in 2013.

Today we have a leadership position in the market that is strengthening every quarter; it has been backed by a strong product pipeline. We have consistently grown our revenues over the past years and quarters, we have steadily developed and introduced market-leading solutions [founded] in the key market trends and we have demonstrated, as Meir mentioned, increased efficiency in our business and continuously improved our operational results.

Before concluding, I would like to thank our customers and partners for their continued support and trust. I would like to thank the Radware team for all their effort, commitment and success in growing our business. With that I would like to open the discussion for Q&A.

(Operator Instructions). Our first question comes from Alex Henderson of Needham & Company. Please go ahead.

Hi guys, could you give us a little bit of a breakout between the geographies? You know, delineate how the US did versus Europe and what you're seeing in terms of those geographies as you're looking out into your guidance? Particularly, when you look at the US, can you also break out between the rate of growth of product sales versus the rate of growth of service?

Okay, I'll give the break for the fourth quarter. The US was 31% of our total revenues, the international 69% split between EMEA 33% and Asia-Pac 36%. Overall the product sales were growing faster than the overall Company; that means faster than the services and Roy will give more color on that.

Yes, so in terms of the geographies, we believe the conditions in North America continue to be very good and we continue to see strong growth in that market. The most challenging geography we believe is EMEA and currently we're seeing flattish performance there. We believe there are opportunities for growth, but from the macro perspective that's the most challenging, and Asia Pacific is in between, it depends on the exact countries.

As Meir mentioned, we are seeing, and it's consistent over the last year, stronger product growth, product growth versus maintenance renewals in services. And obviously we think there is also room for improvement on the services side given what we are seeing in the industry.

One more question, if I could. So is it reasonable to think that their services lag considerably to the product sales growth here? It looks like your product sales growth was up 31% in the -- the overall revenue was up 31% in the US, down 9% or so in Europe. So if I look at that mix, is it accurate to say that your product sales growth was closer to 40%?

In the US you can assume that, yes.

In the US.

Yes.

Okay, thank you.

Our next question comes from Michael Saloio of Oppenheimer. Please go ahead.

Hi, thanks for taking my question. Could you give us a sense of the size of the Check Point revenue at this point, or at least how we should be thinking about the growth trajectory through this quarter and looking at next quarter?

And then I was wondering if you could also give us some similar color on the Juniper relationship. Thanks.

Okay, so regarding Check Point, we recognized minimum revenue this quarter and we continue to be on track for that in Q1 and beyond. So we believe there is a lot of joint opportunities together with them and we are optimistic on the actual revenue potential in 2013 from this partnership. Obviously, we continuously need to train in the various Check Point offices in China and it's an ongoing task, but we constantly see an increased pipeline and very interesting deals in that regard.

Concerning Juniper, there is some booking maybe in Q1 and Q2, we think there may be an upside given some of the wins that they've communicated to us, but overall we need more work to run this alliance to where we think it can get to.

Okay and then secondly could you just give us a breakdown between enterprise and service provider in the quarter?

Yes, the enterprise was 72% while the carrier 28%.

Okay, thanks.

Thank you.

Our next question comes from Joseph Wolf of Barclays. Please go ahead.

Thanks, a quick follow-up on that last question is are you able or willing to give any sort of revenue target for the OEM agreements for 2013?

And then, as a follow-up question, could you go into some of the end markets, or could we get the end markets split for 2012, data center cloud security within the businesses?

Unfortunately, Joseph, we don't break out by the specific application or target environment our revenue.

Regarding OEM, we would like to see more actual Check Point ramp-up revenue and then we can be more specific with the exact target number for 2013. So it's a bit premature, hopefully in a quarter we will be in a better position to assess that.

Right now you're more focused on where you see more near-term opportunity for Check Point and Juniper from what I'm understanding here. Is that fair?

that's correct, but there's a difference in the opportunities. With Juniper we're working on the carrier market, so there may be fewer opportunities but they are generally larger in size. The Check Point solution targets all segments but, more specifically, the enterprise market and there we are seeing many, many opportunities, but obviously of smaller size. So I think it's different in nature and in target markets, but short-term we believe that the Check Point revenues would surpass the Juniper revenue.

Okay great, thank you.

Our next question comes from Jeff Lubert of Wells Fargo. Please go ahead.

Hi guys, thanks for taking my question -- couple of questions. First, can you comment on overall visibility levels? To what extent you're seeing activity and visibility improve or deteriorate, to what extent you're factoring an added level of conservatism in your outlook here?

So I don't think there's a major shift or change in the visibility that we have, I think maybe second half we did see a decline in visibility in EMEA but I think we are not expecting any worsening environment there from what we can tell. So that level of visibility or performance is in our guidance.

Regarding the other markets, I think there was no change in the past years to the level of visibility that we have into them, so I think we can be pretty accurate.

In Europe are things better or worse on the enterprise or carrier side? Any incremental detail you can offer there?

I'm not sure whether it's our own view or in general, but I think enterprise, depends on the segment, are doing well. In some countries the carrier was a bit weak for us in the second half, but I'm not sure it's something to take as a conclusion. It can be also from timing of projects and sometimes execution that we can improve, etc., but that's what we've seen.

And then finally it seems like hiring slowed a little bit in Q4 relative to Q3. Can you discuss what drove the sequential decrease? Talk about how you're thinking about hiring going into 2013.

And previously you've discussed getting to a 23% operating margin exiting 2012, is there any granularity or detail you can offer as to where you think you might end up exiting 2013.

So, first regarding hiring in the numbers you see that. But I think especially in hiring of sales and support personnel, you are going to see some changes. Some of it is timing, so I believe we will go back to a higher hiring rate, and I think you see that already in Q1 headcount numbers.

And in the total headcount there was some shift internally between our departments. So I would say our hiring on the field and especially North America has not slowed down, it's actually accelerated in Q4, and we foresee that also for the coming quarter or two quarters.

Now regarding the model, I'll let Meir answer that.

Okay, regarding the model we prefer not to (technical difficulty).

Ladies and gentlemen, please stand by your conference will resume momentarily. You may resume your conference.

Hello?

Hello?

Yes, we apologize for that. So I'll let Meir refer to the model.

Okay. Again about the model we prefer to give right now long term model three years target of 30% operating margin rather than short term model. Later on in the year, as Roy discussed, we will have more visibility so we can share with you also short term targets. Right now we'll leave it as it is the long term model of 30%.

Well, Meir, would you expect to see leverage in 2013? It does sound like you're going to be ramping hiring here in the upcoming year, so is that longer term target -- is that leverage happening in '14 and '15 and not necessarily '13? How are you thinking about that?

It's a little bit complex what you ask. It makes sense what you mention, but on top of that you have to bear in mind that we enjoy from over 80% of gross margins.

So the marginal $1m can impact our operating margin over one full point. That means that we should again review this one later on in the year, and to see if we expect to go higher than any number that we see right now. So because of that let's leave it for the next quarter.

We are committed to increasing top line, bottom line and margins. But of course cannot be committed for every given quarter, such because that the revenues always are lagging behind expenses. We hire people right now; when they are fully on board and contributing to the sales it takes six to nine months this is the time frame.

All right, thanks guys.

Our next question comes from Rohit Chopra of Wedbush. Please go ahead.

Thanks very much. Three questions, one I just wanted to get a sense of the security business. Can you just tell us qualitatively at least is the security business growing faster than the core business?

My other question is on Juniper. Roy, you sounded a little bit more optimistic for Q1 and Q2. Can you tell us if you've made any changes on the go-to-market strategy there, and if you've had any discussions with Juniper?

And lastly, Meir, maybe you can just tell us about linearity.

Okay. So regarding the Juniper relationship, I'll start with that. We are constantly in discussions with them and working together. The go to market is obviously for Juniper sales organization to take it to market. And we are helping them when they feel it is necessary to bring us front of customers, etc. We do sound better because we see more -- they communicated more wins. For us that resulted -- we have this visibility.

Again, I want to remind you that Juniper is booking with us and we are recognizing only in the very final stages of their project delivery. So they are winning the customers, they are installing and they are actually booking with us in the final stages.

So on one hand we have pretty good visibility (inaudible) win and on the other hand there cannot be also good surprises beyond what we know, because that would translate into revenues only into future quarters.

As for the linearity the first month of the quarter in Q4 was 28% of revenues, the second month 25% and December was 47%.

Okay. And then just remind us on security is that as a business or as a segment growing faster than the overall?

So that depends on the quarter, it depends on the quarter and the geography. Recently in Q4 I think it grew faster, but overall if I look at this, a couple of years back it was relatively the same growth as for example the ADC market or ADC virtualization and so on. So security is more -- is growing faster when there is attacks in that geography or in that market segment. For example, I would say that North America currently is a very hot market for security solutions, especially financial markets, online and government agencies.

Thank you.

Our next question comes from Mark Sue of RBC Capital Markets. Please go ahead.

Thank you. Gentlemen, if we could get a sense of how you're thinking about new product introductions for 2013. I understand that there is new DefensePro, but maybe if you could talk about approach to new products; where you might be focusing on and how that you might be proactively upgrading your install base that would be helpful.

So -- now, I would speak in general terms, because as you know we are not discussing future product introductions until they are available and we would rather keep this discreet today. So, in general we have we believe a very strong pipeline of introductions in application delivery and also in application security in 2013.

In application delivery, we believe that the Radical set can be extended. The virtual application delivery infrastructure can be extended with additional services and capabilities that are very well adapted to a cloud data center that is more remote from the users than the enterprise data centers. And with additional services that provide the availability, performance and security of this infrastructure and more robust for this type of cloud, hybrid cloud, private or public environment. I believe that we are coming with new product introductions already in first half.

In terms of security, we just released in Q4 our next gen high-end platform, the x420. The reaction from the market is very good. And a lot of interest from the large online companies, carrier companies and so on. We will be announcing additional products in the security. There are some big shows coming up relatively soon. And we believe we will have major announcements in expanding that business as well.

Okay, it sounds good we'll see you at RSA. Maybe if we could switch gears and talk about the Cisco opportunity; the pace of upgrades for you. Many of the competitors are doing trader programs. Anything else that you can do to gain more than your share of that business as a replacement opportunity.

So we believe the Alteon ADC is very well adapted to replace the Cisco ACE platform. The Cisco claim to fame was the concept of virtual context. It is similar in principal to our instances but this is an inferior solution because there is no physical separation between the instances. It's just the virtual context of just separated from a management point of view.

And we believe customers will see our ADC solution as the next gen ACE architecture where the instances are separate not only from a management point of view but from operation, from a software, firmware point of view, from actual capacity where (inaudible) is allocated and so on.

We are involved in several replacements. As a matter of fact a large North American carrier is going live next week, beginning in February with our Alteon 10000 multi-instance, merging, converging their Cisco ACE infrastructure. And we have similar opportunities around the world.

I would say that the deals that we are executing right now of replacement are deals that were started before the Cisco announcement. Maybe the Cisco announcement helped accelerate them, but it's really deals that the customer's already decided to migrate or to look for an alternative solution before the announcement.

Following the announcement, and you know since [start of the year] we are seeing more activity, the Cisco ACE customers are starting to look at their options. And obviously we are focused on that. So we think we have a good product, better than the competitors to match with the Cisco ACE and especially the blade customers will use these.

And at the same time we are focused with our sales force which we believe is covering well the large enterprise and carrier markets to execute on this. So we are well positioned and that's why we think we can get more than our fair share there.

All right that's helpful. Good luck, gentlemen, and thank you.

Thank you.

Our next question is a follow up from Alex Henderson of Needham & Company. Please go ahead.

Thanks. So we had a lot of product announcements out of F5 the last quarter and half, particularly several announcements this morning. I am sure you guys have been quite cognizant of what they are announcing. Can you talk a little bit about how you see your competitive position against, particularly the rackable line extensions that they've come out with? I think they are out with the 2000 and 10000 today along with the 4000 which was out earlier.

Yes. So frankly when F5 say that they are the first ADC vendor to do something I'm not sure anymore what the first word means. So let's start with the 2000. The claim to fame there is a 10 gig interface on a low-end ADC, as far as I understood from the press release. Our Alteon 5224 that was announced more than a year ago, starts at 1 gig all the way 16 gig, with built-in gig ports and 24 gigabyte port and has all our VADI capability, the multi instances and so.

So I don't think the 2000 or the 4000 brings any new dimension to the competitive landscape as far as I see today. And as far as we can tell they don't include the instances, the virtualization capability. So our advantages as we see them and our focus stays the same following these announcements.

So just to be clear you see both the 2000 and the 4000 as conventional ADC with no hypervisor?

Correct, correct.

Okay, thanks.

One word regarding the 10000, again the announcement is focusing on the first ADC with a 40 gig port, we've announced the Alteon [6420] I think around six months ago with four 30-gig port and 20 10-gigabyte ports and a complete set of our virtual capabilities. Again we believe it's dramatically more than what the 10000 offers. So we think competitively versus F5 we do have a superior product and especially towards virtual and cloud data center, our leadership stays intact.

So if you look at the 10000 that does come with a hypervisor, and I believe it allows six virtual instances. How does that compare with your products in a comparable price point? What do you see the price point at?

Yes, so we don't know yet the exact price point of F5's release, we would -- it was just -- the press release was just this morning. But regarding the density, this platform is probably targeting the high end of the market at 10000, 80-gig market or around and they have [there] six instances.

So each application if you divide it is roughly 14, 13 gig. We don't believe that is a typical application in an enterprise customer. As a matter of fact, in our 1-gig solution we already have 24 instances, so four times of this at this high end.

If you want to have many instances, our new high capacity, our Alteon 10000 offers at 80 gig, which is a similar bandwidth capacity to the 10000, it offers 256 instances versus six.

So we don't understand how a cloud provider or carrier can use only six instances. We believe it's going to be at least 50, 100 and probably multiple hundreds. So we don't clearly understand why these six instances will be the total amount of instances a platform would give.

If I could just ask one more question along that line. Is there environments at the very high end of the market, say, take a JP Morgan or Citigroup with a huge data center that would justify a much larger device with a lot of -- or with very high throughput characteristics that would have fewer instances. Or do you see the need even in the high end application environments to have a large number of virtual instances on each of the boxes?

We believe the right architecture is [driven] instance per application and that guarantees for (inaudible) isolation between applications. So one does not impact the other and does not interfere with the other. So, for example if you need to move that application from one data center to the other it's very easy, because it's already like a silo; it has its own ADC instance. If you don't have a dedicated instance, how are you going to break out this one application out of the other 100 applications that are sharing the same instance.

So architecturally and operationally and from a compliance point of view we clearly believe that instance per application is the right way to go. And then if you think about a Citibank or Bank of America you are speaking about hundreds of applications.

So there isn't enough volume in any one application to absorb the capacity of a high end product and therefore you'd still need the number of -- the higher number of virtuals?

Yes.

That's good, thank you.

Definitely not in the enterprise market, you might think about a Facebook or something like that, but not in the enterprise market.

Okay, thank you.

(Operator Instructions). Our next question comes from [Lewis Skull] of Open Field Capital. Please go ahead.

Good morning, gentlemen. I was wondering, given that your competitors are disclosing their products by product line and breaking out security and products I was wondering what the reasoning is for you not to disclose that.

I am not sure to which competitor you refer to when you say they are breaking it by product line. As far as I know for example F5 is (multiple speakers).

The product category (multiple speakers) -- yes, I see F5 break out security and ADC revenues. I am just wondering given that you participate in those two markets as well, maybe in different areas in the security side, why you are not breaking out security and product -- security and ADC.

I am not aware that F5 breaks security and ADC. I am aware that they are breaking out the storage, virtualization and all the rest. Regardless, our own policy, not to focus too much on F5, we think that a lot of our modules are shared between application delivery and application security.

For example, if it's an encrypted attack we are using our ADC module. The Alteon product is part of the complete AMS solution. So if we are building a system for a bank or for an online company that uses HTTPS, the encrypted web protocol, in every project we have a combination. Now is that in ADC or is that a security? It's very, very hard to set.

So we as a company we believe application delivery and security for -- this in the end is going to be one market which will include all the services and application needs in order to be delivered in an optimum manner from the data center. So we think this break between the two segments is maybe interesting from a product and an R&D point of view but not from our customers and market approach.

Thank you.

(Operator Instructions). I am showing no further questions at this time. I would like to turn the conference back over to management for any closing remarks.

I would like to thank everybody for joining us today. Have a great day. Thank you.

Ladies and gentlemen, this does conclude today's conference. You may all disconnect and have a wonderful day.