Good day, and thank you for standing by. Welcome to the Fortinet Second Quarter Earnings Call. At this time all participants are in a listen-only mode, after the speakers’ presentation, there’ll be a question-and-answer session. [Operator Instructions] Please be advised that today’s conference call is being recorded. I would now like to hand the conference over to your speaker today, Peter Salkowski. Please go ahead.

Thank you, Hope. Good afternoon, everyone. This is Peter Salkowski, Vice President of Investor Relations at Fortinet. I’m pleased to welcome everyone to our call to discuss Fortinet’s financial results for the second quarter of 2022. Speakers on today’s call are Ken Xie, Fortinet’s Founder, Chairman and CEO; and Keith Jensen, our Chief Financial Officer. This is a live call that will be available for replay via webcast on our Investor Relations website. Ken will begin our call today by providing a high-level perspective on our business. Keith will then review our financial operating results for the second quarter before providing guidance for the third quarter, updating the full year. We’ll then open the call for questions. During the Q&A, we ask that you please limit yourself to one question and one follow-up question to allow for others to participate. Before we begin, I’d like to remind everyone that on today’s call, we will be making forward-looking statements, and these forward-looking statements are subject to risks and uncertainties, which could cause actual results to differ materially from those projected. Please refer to our SEC filings, in particular, the risk factors in our most recent Form 10-K and Form 10-Q for more information. All forward-looking statements reflect our opinions only as of the date of this presentation, and we undertake no obligation and specifically disclaim any obligation to update forward-looking statements. Also, all references to financial metrics that we mine in today’s call are non-GAAP, unless stated otherwise. Our GAAP results and GAAP to non-GAAP reconciliations are located in the earnings press release and in the presentation that accompany today’s remarks, both of which are posted on the Investor Relations website. Ken and Keith’s prepared remarks today for the earnings call will be posted on the quarterly earnings section of our Investor Relations website immediately following today’s call. Lastly, all references are on a year-over-year basis, unless noted otherwise. I’ll now turn the call over to Ken.

Thanks Peter. Thank you to everyone for joining today’s call to review our outstanding second quarter 2022 results. Total billings increased 36%; the fifth consecutive quarter of at least 35% year-over-year billings growth. Revenue grew 29% driven by 34% product revenue growth. SD-WAN and OT bookings grew over 60% and 75%, which together accounted for 25% of total second quarter bookings. Our better-than-expected performance demonstrates the strong demand for our cybersecurity innovation. Fortinet is at the forefront of networking and security convergence, enabling our customers to reduce complexity, while securing and connecting hybrid and remote users to advanced security with superior performance. Today we announced the FortiGate 4800F, our latest innovation in converged Network Security. The 4800F is the world’s fastest compact firewall for hyperscale data centers and 5G networks. Powered by Fortinet’s NP7 SPU, the 4800F delivers Security Compute Ratings of on average 5-10 times better performance than competitive solutions, across the six most common and important functions. A leader in the Gartner Magic Quadrant for WAN Edge Infrastructure, Fortinet continues to take market share for Secure SD-WAN. Our integrated secure SDWAN solution, powered by Fortinet’s SPU SOC4, delivers huge performance, security and efficiencies over traditional offerings. In addition to convergence, consolidation of vendors and product functionality is another major trend, particularly in Network Security. In a recent CISO survey, Gartner found the percentage of companies surveyed who want fewer security providers increased to 75% from only 29% in 2020. With over 30 products lines built mostly by our in-house strong engineering and development innovation, Fortinet is benefiting from this consolidation with our Security Fabric MESH offering. The Fortinet Security MESH platform delivers unparalleled protection with broad, integrated and automated protection across multiple edges, from endpoint, to data center, and hybrid cloud environments. These two major trends, convergence and consolidation, position Fortinet well for long-term growth. Before turning the call over to Keith, I’d like to thank our employees, customers, partners and suppliers worldwide for their continued support and hard work, that are contributing to Fortinet’s strong growth.

Thank you, Ken, and good afternoon, everyone. Let’s start the more detailed quarterly discussion. Second quarter results were solid and broad-based across geographies, customer sizes, industries, and use cases, driving market share gains and demonstrating the strong support from our three key growth drivers: first, an elevated threat environment; second, the convergence of security and networking; and third, the consolidation of security products across our platform offerings. Total revenue of $1.03 billion was up 29%, passing the $1 billion milestone in quarterly revenue for the first time in our history. Total product revenue growth was up 34%. Core Platform and Platform Extension product revenue growth was up 35% and 33%, respectively. We continued to see robust product revenue growth from a wide range of security use cases, including Secure SD-WAN and Operational Technology, or OT. Total service revenue growth increased sequentially to 25%, resulting in service revenue of $629 million. Support and related service revenue was up 26% to $289 million, while security subscriptions service revenue was up 25%, or 2 points sequentially, to $340 million. Service billings, defined as total billings minus product revenue, were up 36%. The year-over-year growth rate for short-term deferred revenue has increased for six quarters in a row from just under 21% in Q4 of 2020 to just over 31% in Q2 of 2022, the highest short-term deferred revenue growth rate in over six years. The accelerating growth rates for service billings and short-term deferred revenue reflect the earlier pricing actions that quickly appeared in product revenue and that are now beginning to appear in service revenue. The pricing benefit more than offset various service revenue headwinds, including suspending services in Russia, an increase in the average number of days between when a customer purchases and subsequently activates a security service contract, and the…

Thank you, Keith. As a reminder, during the Q&A session, we ask that you please limit yourself to one question and one follow-up question to allow others to participate. With that, Hope, please open the call for questions.

Thank you. [Operator Instructions] Our first question comes from the line of Brian Essex with GS.

Great. Thank you. Good afternoon and thank you for taking the questions. Congrats to team on a nice set of results for the quarter. I was wondering if maybe -- and Keith, I certainly appreciate the commentary and the granularity with the full year revenue guide. Can we maybe unpack some of the commentary on the services side and the lower, I guess, services revenue guide for the year? Maybe help us understand what’s going on there. And maybe pair that with your comments on delays in activations and how much insight you might have there that gets folks comfortable that there isn’t pull-forward? Thank you.

Yes. I don’t think -- Brian, I don’t think pull forward really applies to the service revenue line, but maybe -- it’s actually a couple of questions at once. I think the answer to your question about service revenue, I think the biggest change from where we started the year is really about Russia. If you think about Russia, we talked at the very beginning, it’s about 1.5% of our total revenue. And it applies to the service revenue line as well. Earlier in the year, we stopped recognizing revenue on existing contracts for services that provide in Russia in conjunction with our suspending of services. At that time, we did not anticipate it would be a full year event, but we are now. And if you kind of think through that 1.5% of service revenues in Russia, so we really backed out now for the full year about $25 million of revenue related to Russia. That’s the largest change there. I think the delay in registrations from when contracts are sold, when they’re actually registered by customers, I think we pretty much got out of the quarter what we expected on that. That seems to be something in the current environment with inventory constraints that we’re going to continue to see. I think the linearity part of it is a little bit new in that because the shipments occurred later in the quarter, there really wasn’t the opportunity to get the service revenue from those Q2 shipments that we would normally get. And so, I think those are really the parts to put together there. On the other side, I would probably point to, again, the short-term deferred revenue billings and the number that we’re putting up here and the growth that you’re seeing with that number as well as service billings itself. And I think the last comment on this, and we haven’t talked about it before, is that the service contracts are really a use it or lose it contract, meaning it’s not that they have the ability to cancel, they just have the ability to postpone the registration for a period of time, whether -- depending upon the geography, whether that’s 90 days or one year or what have you. So, eventually, it comes to revenue, but the timing has been pushed out to that aspect.

Our next question comes from Fatima Boolani with Citibank.

Hey. Keith, this one’s for you. Just with respect to some of the backlog detail and commentary that you shared, I want to hone in on the cancellation rates that you quantified for us. I believe you said it was about 4%. Can you give us a sense of what are some of the reasons behind the cancellation? And what gives you confidence that that 4% isn’t going to stretch or escalate into maybe mid to high single digits? Thank you.

Yes. I think I would point to some of the factors that we’ve talked about before. I think that last quarter, we talked about the cancellation rate being 5%, I think we’re seeing it now 4%. I think it’s unlikely that existing customers, particularly those that have received partial shipments are going to cancel. Also, I don’t want to -- that it’s naïve that as backlog gets older, and it’s also why we provide that 60% of shipments from prior backlog number for The Street to try and understand it. But if that number starts to tick up, obviously, there’s more risk in it. I think it’s important to understand the guidance that we provide really isn’t reflecting in any sort of shift in the backlog that we’re going to ship a lot of things from backlog. So, I think we’re fairly prudent in that regard, and I think we’re also comfortable with described as the stickiness of the backlog number.

Our next question comes from Adam Borg with Stifel.

Maybe just on the macro. You talked a little bit about the demand environment and highlighted some delayed deals and elongated linearity. Can you maybe go a little bit deeper here and talk about what these customer conversations look like? Are these tied to any particular verticals or geographies? You talked a little bit about large enterprise or larger deals. I’d love to hear about the midsize to smaller deals, too. Thanks.

It’s Ken. So, we do see a lot of customers, especially enterprise, they’re starting to design some new infrastructure to support and work from anywhere and also expand security beyond the traditional network security and to whether like internal segmentation to prevent nowadays ransomware attack or go to work from home and at the same time, have multiple security products, need to be automated together. So we call the consolidation, both on the product side, on the vendor side. I see this kind of trend will be last for the next few years, will be pretty long-term change. And at the same time, Keith also mentioned elevated security environment also other drive. So, that’s where we see the trend we’re keeping going for the next few years. If you look at the billing number compared to two years ago, I mentioned in my script is -- we have over 35% of billing increase in the last five quarters compared to two years ago, Q1 2020 is only about 14%. So, we do see the change in acceleration and also the convergence consolidation going on in the whole space right now will be pretty much the amount of growth.

Our next question comes from Saket Kalia with Barclays.

Keith, maybe for you. I’d like to talk a little bit about bookings. Can you just talk about how bookings did versus your expectation this quarter? I think the guide coming into Q2 was for about 40% growth. We came in at 42%, clearly better, but a bigger delta on the billings versus the guide. And so, can you just talk about how to read into that, if there’s anything to consider there with just those two kind of in relation to each other?

Yes. I don’t -- I think, obviously, 42% bookings. I think we’ve been over 40% now for 3 quarters, maybe 4 quarters in the bookings line, we feel really, really good about it. I think the one thing that we’re looking at internally, it’s just that the interplay between bookings, backlog and billings and trying to really get a sense of the direction of the business. And the example I kind of gave is we had a very good quarter on the midrange of the product. But some of that was due to availability. Demand was very strong, but it was also because we had the midrange product available. We didn’t have as much product available in the low end. Now, as we shifted to this third quarter, I think we’re probably going to see the low end availability improved fairly dramatically. And so, when you’re looking at billings information that we have historically disclosed and trying to gauge the direction of the business, it gets a little bit distorted now just in terms of what’s available. And I think of the total when you look at bookings and backlog, there’s some of that as well in terms of the characterization of what the booking is versus the characterization of what’s available to ship and what comes in the billings line.

Also compared to one year ago, maybe a tough comparison because we’re started to see the acceleration about five quarters ago.

Our next question comes from Adam Tindle with Raymond James.

Keith, you talked in the prepared remarks about still expecting supply challenges for the rest of the year and to offset you’re thinking increased purchase commitments, qualifying additional suppliers and pricing actions. I wanted to zoom in on that last point on pricing actions to see if you would maybe put a finer point on timing and magnitude for expected pricing actions? And secondly, you sounded positive on elasticity and confident on the elasticity moving forward. But just curious what underpins that confidence, especially in international markets with dollar strengthening in local currencies and fixed budgets, et cetera? Thank you.

Yes. Great question. And I’ll take the last one first because I think it’s probably very important, and I probably forgot the first one already. So one of the things that we do is we track very religiously in our CRM tool when a customer -- if we lose a deal, we want to know why, right? Is it because we could not overcome the incumbency, is it because the channel partner may have had a bias, if you will, to one of our competitors, a feature issue, a functionality issue or something like that, but also very specifically, do we lose on price. And we’ve been tracking that now for over a year. And that percentage, which is low -- lower than the other ones that I just gave in to you, has been extremely consistent. And so with that very consistent loss percentage, if you will, I translate that into price elasticity, which tells me that the question is always how far can you push the envelope. We know we come into the conversation with a significant price and performance advantage, The Street sometimes -- or channels just may say 30% or 40%. We have known that from the beginning of this phase of the economic cycle. And the question has become how far can we push that. But again, keep in mind, our goal is really to try over a longer period of time just match the cost increases and maintain a consistent margin. It’s not that we’re really trying to take down more margin. Now you will get volatility quarter-to-quarter because of the mix and things like that. So, a long-winded way to say, I think I really hang my head on what I’m seeing and we’re tracking on the CRM data about reasons that we lose deals and reasons that we win deals.

Yes. And also a few other comments about pricing. Our policy tends to be just by the price by small step, but also kind of more open, like we do have a new price book basically released every quarter. Also, the other thing is really like the product we released today, on average for the same function, for the same price range, our product has a performance factor 10 times better than competitors. Like Keith mentioned, on the CRM on the tracking, we don’t see any big loss on kind of changing or even -- if you do not -- actually improved because we still have a huge price advantage compared to competitors. The other thing also may be mentioned to the service. I think one we may try to improve a little bit going forward is really even in the last few quarters, we increased some of the price more on the new product release, but we are not changing the price for the expired product. Basically, the product is still added to the service like 5, 6 -- right? So, that’s one thing we made because the labor cost on the service and supporting, so we may have to increase the price of the outdated product no longer shipping because all the service or the renew still tied to the old product, which is no longer shipping but is the customer still buying the renewal, buying the service based on the old product. So, that’s probably -- we can also help in the improving the service and also helping the margin and compensate our additional costs, especially on the labor.

Our next question comes from Michael Turits with KeyBanc.

So, Keith, two questions. First, maybe it seems obvious in some ways, but do you feel like -- and this is the first time that you’ve talked about this linearity issue as well as the extension of the negotiation cycle. So, do you simply tie it to macro being worse right now, or do you have any other insights to it? And then, I just wanted to make sure that in your mind, second question, you really think that it’s really primarily services as a result of those things, this shortfall in the year and you’re happy with product?

Yes. So, I think the -- well, putting aside the -- Mike, you broke the rule and you asked two questions and Peter said only one and a follow-up.

Sorry about that...

I think what we’re seeing in linearity, unlike a “normal world” where you can kind of look at linearity and DSO and you can get a sense for whether or not a company is pushing to close deals at the end of the quarter and maybe it’s a more challenging quarter. Because of the timing of when the finish -- when the inventory is delivered from the contract manufacturers, we see a shift in that linearity of when we receive inventory from our contract manufacturers. That shift then it translates into when we can turn around QA, et cetera, and ship it out to our customers and sell it. So, there’s a different aspect of linear that’s come into play here now. So, service contracts that maybe would have been sold in the first month of the quarter, actually got sold in the third month of the quarter. So, we lose service revenue from that. And you see that appearing in the DSO and you see it appearing in the free cash flow. On the negotiating side of it, I think what we saw, and I don’t know if this is common to others -- what we saw was probably the first two weeks of June and maybe there was more conversation around recession and concerns there, if you will, a bit of a pause in terms of deal closure rates for those first 10 days of June and then a reacceleration as we got through the end of June. We did notice or I did notice during that time frame maybe additional parties were being introduced to either as an approver or a negotiator, if you will, on some of the larger deals just to make sure on the customer side that they were making the right decision. And I think that’s why I went on to say in the prepared remarks, not only did we notice this shift, but the close rates, which were important actually were up just a tad in the quarter. So, I think there was just for whatever reason, there was a slight pause there for a couple of weeks in June and everybody came back and got the deals done by -- on the customer side and our side at that last week in June.

The other reason for a little bit longer closing time really the bigger deal grow faster. So, like I mentioned, the deal over $1 million grew over 50% year-over-year. So, that’s the bigger deals also tend to be -- take a little bit longer time to close. And also, we see more like a deal involving multiple products, not just the FortiGate, but also we call the non-FortiGate, call the platform retention, which also take a little bit long time to test -- evaluate to close. On the supply chain, since really compared to before the pandemic, we probably ship majority -- most of the products by sea. Now we’re pretty much shipping every product by air. That’s where the timing of the supply shipping the product to us is pretty critical automatically. And so last quarter, we did experience a lot of product being shipped in end of the quarter from suppliers to us, that drives the linearity. Even we have the booking -- but we are the way -- like a few weeks or even a couple of months before the product was shipped to the customer more towards the quarter end because the supply shipping goes pretty much in the end of the quarter. So, we do see long term, this will be changing, improving. We’ll keep increasing some of the product inventory and improving the product turn and also balance among not just shipping everything by air, but some by air, some by ocean.

Our next question is from Keith Bachman with BMO.

Good segue from Michael’s question. Keith, I want to try to understand, you talked about a few different things impacting the year guide. To put context around it, your revenue guide for ‘22 isn’t changing, which I think is viewed as a disappointment to investors. Now underneath that, services revenues is getting compressed a little bit. And so, as you think about why the revenue estimates are going higher for the year, is there a change in, a, the demand level, whether it’s the elongation because you said, in fact, there was two weeks, sort of weak at the end of June, but it sounds like during the last -- through the 3rd of August, things have normalized, or is it b, supply chain issues that are causing you to not raise your revenue guidance even as you’re raising billings modestly? I’m just trying to understand what are the forces that are impacting the lack of raise, if you will? Is it the demand side and/or is it the supply side?

Yes. I don’t really think it’s necessarily either demand or supply. I would start the conversation off by saying I think the pipeline growth is extremely strong. We feel very, very good about that. I do think there’s a fair amount of uncertainty as we look out beyond the third quarter to the fourth quarter in terms of directions the economies may go, what inflation may do and a little bit of supply chain. I don’t think -- we did, as you point out, cover the shortfall, if you will, in the service revenue -- economies may go, what inflation may do and a little bit of supply chain. I don’t think -- we did, as you point out, cover the shortfall, if you will, in the service revenue, in the product revenue. So I think that’s a fairly good size of us being bullish and feeling very, very good about our competitive advantage. And I think that the other aspect we talked about is just the large deals and how we’re seeing the success in the enterprise and getting a little more dependent on large deals than we have in prior years and some of the close rates around those. I think that while we’re bullish, we think we have competitive advantages. I don’t know as we get to the fourth quarter, if this is really a good time to think about that in a very, very aggressive fashion.

Okay. So, as I just clarify, so it sounds like you want to be a little bit conservative or you don’t want to get ahead of yourself on particularly the Q4 guide, so lead numbers where they are on revenues, in particular?

Yes, I think that’s a fair description.

Our next question comes from Shaul Eyal with Cowen.

Maybe segueing from the prior question from revenue maybe to OpEx. So, your hiring plans appear to remain largely on track. What’s the current thinking on second half? Is it becoming a little easier in recent months, given some layoffs at some private competitors?

We want to maintain healthy margins and then also keeping growth and gaining market, agreed hiring is relatively a little bit easier compared to like a few quarters ago, especially in the cybersecurity space. So for us, we feel we have a good pace on hiring, especially we still -- we’re keeping gaining market share. And the margin -- and it’s a healthy margin, basically, both on the gross margin and also on the operating margin side. So, we feel we have pretty solid plan and balance among the growth and margin.

Yes. I think Ken’s spot on with that. I would probably offer a couple of things to support. One is you continue to hear us talk about our inventory commitments looking out now 6 quarters or more. I think the read through that is that we still feel fairly bullish about it. And the other aspect of it. And the other aspect of it and Ken made reference to it is we talk about 25% operating margins in different ways over the years as being an average or target, what have you. And obviously, to -- in this environment to -- with high inflation, to come in successfully and still be providing guidance for the full year of 25% to 26%, while growing the top line aggressively while taking market share, I think we feel very good about how the sales team, the engineering team, the operation teams and support teams, et cetera, are all working together and driving the growth of the Company and the execution.

The next question comes from Hamza Fodderwala with Morgan Stanley.

Maybe a question for both, Ken and Keith. Ken, just given the general pressure on budgets in the macro environment, are you seeing a little bit more impetus to -- from customers to want to consolidate to a converged security networking platform like Fortinet? And then for Keith, if you’re seeing any weakness, let’s say, elongating, negotiating cycles and whatnot? Is it more weighted towards the core platform FortiGate side or the platform extension side, which is in access points?

It’s a very good question. Definitely, we see the convergence among the network and security. Also the pandemic accelerated this kind of change, especially inside the company, campus network and also work from remote anywhere. So that’s where we see that pretty strong growth. And also a lot of connected devices like in the OT space, also, we see very strong growth. Like we mentioned, SD-WAN grew 60% and the OT growth 75%, and we’re still keeping growth and gaining market share there. And at the same time, have to secure the whole infrastructure not only expanding our network security to the networking side, but also like beyond the network security, with the end point, with the cloud with all the other like application level from email, web, working together. So we do see all these -- we keep saying the convergence and consolidation will benefit Fortinet multiple years going forward, it’s more long-term growth driver for us. And we prepared this in the last like 22 years since start company with investment like from ASIC technology for the R&D with most of the product internally developed to integrate, automate together. And so, we do see the time is starting coming and for all this investment starting to see some good return. And also, we feel we have a very healthy business model since IPO now is about 13 years now, want to maintain the balanced growth and also healthy margin. And that is what makes the company to last longer. And at the same time, we’re also kind of keeping invest in the long term to follow the change and also keep up the innovation and quickly customer benefit from our innovation and also the long-term investment.

Yes. And Hamza, I think your speculation about where the larger or the timing comes in is accurate and that it’s going to be in that 1/3 of our business that is large enterprise. One, the dollars are larger, obviously. And so, they’re going to -- customers spend a little more time with the ROI. But I think more importantly, and to your kind of second point, your question on, by adding in more of the platform products into a deal, you’re perhaps a little more likely to run into additional competitors or people internally that are champions of those competitors. And so, there’s a little more than it takes to get across the finish line because they are more complex in that way. I’ll fill the void here. As a reminder, we did 122 deals over $1 million in the quarter, which -- that’s a pretty fantastic number for us.

Our next question is from Gray Powell with BTIG.

So, Keith, I know you hit on this once or twice already, but I just want to make sure I understand a dynamic on the services billings. So, if I back out product revenue from short-term billings, it looks like the annual recurring component of billings actually accelerated pretty nicely. I’m calculating like 29% in Q1, improving to 40% in Q2. I don’t need you to bless the numbers, but directionally, does that seem right to you? And then, if so, how much of that was driven by pricing dynamics that you talked about versus just the natural cadence of the business?

I do think your math is directionally correct, but it’s lot more time to get anything more about how actually it may or may not be because kind of looking at a different way. And I would say, again, if you think about the timing of where -- when a price increase is effective, right? It’s got to go through the process of being preannounced to the channel partners, they get, I think, 60 days of advanced notice and then whether it’s actually start to have an impact on it. But keeping that in mind, you do see the impact on pricing actions fairly quickly on product revenue and on billings, whether it’s a product or whether it is a service item, right? You will see it there. But on service revenue, you won’t see that benefit for an extended period of time. And I think one thing that may help people, if you think back of our shift from 8 by 5 support to 24 by 7 support. We talked about that for several years because when we turned off the 8 by 5 support, with that came a price lift. And the question that we were addressing, seen for 8, 12 quarters probably in a row, was how was that mix shifting and how was that coming into it. And we were providing information back then about all the buildings, so to speak, are in the 24 by 7, but you’re not seeing the revenue mix that way because it’s kind of -- that mix has to evolve over time as you go through the installed base. Price increases for service revenue, this is just another flavor of the same thing that way. You’re going to see the benefit over a much longer period of time on the service revenue line. You will see the benefit in billings much sooner, and that’s why we gave that information earlier, and you kind of -- Gray, you’re looking at it, that’s a very good leading indicator of where service revenue growth is going to go in the future.

Okay. That’s really helpful. And then just a real quick follow-up. You mentioned $25 million headwinds on service revenue from Russia, which is a new headwind. Does that apply to billings as well, or was that purely a revenue dynamic?

That was a revenue dynamic, and I would probably say 40% of that probably, 30% to 50% of that would have been a billings dynamic, in terms of where we were from the beginning of the year where we’ll end up now.

Our next question is from John Weidemoyer with William Blair.

On your platform extension, cloud security capabilities. I’m curious of the type of customer profile that’s interested there. I suspect they’re probably an existing Fortinet customer that’s transitioning to the cloud. And I’m curious if there are also maybe a fabric mesh. And so there might be an all-in customer -- an all-in Fortinet type customer. Can you talk about the characteristics of the people that are going down?

We do see the cloud security growing well, pretty much on a similar pace as other networking appliance growth. And also, we do see a lot of cloud security come from the service providers, especially carrier service provider, someone also combined with an offer SDWAN and some also with OT, all this together. So that’s what we keep saying, for better security, the new secured whole infrastructure, not just the cloud, but also appliances and some other product infrastructure. So, we do see more and more customers want to consider overall together. So basically, cost security also drive a lot of other part of cybersecurity, other part of infrastructure for cyber security. And also, we do believe long term, the service provider, both in the telecom space, also in the securities integrate and also like even a cloud provider, will play a very, very important role on this security, especially in the service part. And that’s where we also want to keep supporting them. So, that’s where we see that is kind of a still more hybrid environment going forward and especially with more and more device connected with a lot of other, we see kind of the whole infrastructure security more and more important connect, consider all parts of security together is quite important.

Our next question is from Gregg Moskowitz with Mizuho.

I’d like to ask about your backlog, which has significantly and consistently increased for each of the last three quarters. It’s dramatically above year ago levels. So, Keith, you made it clear that the backlog should further rise by year-end, which is great. But at the same time, it’s not going to grow forever. And it’s common to see dips in the company’s backlog due to seasonality, significant order shipments, cancellations, et cetera. And so, it would just be helpful to get your sense of perhaps when we begin to see ebbs and flows in the backlog metric. If you could offer anything there, that would be helpful.

When supply chain is going to get better Ken. So, I’ll let you handle that one.

If you compare to end of Q1, we increased backlog $120-some million. And then end of Q2, we increased about $72 million. It’s a little bit better increase, less than end of Q1 and also less than end of Q4. So, there is some kind of improvement on increased backlog. But also, we put a lot of effort to sourcing different parts or different vendor designed products. So, you see the product line we got quite broad right now. And also can help leverage some kind of -- I mean, some alternative of kind of a more broad supply chain for us. So, that’s why I do believe because the demand is still very, very strong, and so we do see probably keeping -- get a little bit better and better. But like Keith said, we probably not expect backlog will reduce in Q3, Q4. But the increase probably will be less each quarter, I’ll put it this way. And then maybe next year, we’ll see -- starting to see some kind of improvement within the backlog. But overall, with our engineering effort, with the kind of the investment we’ve made in the operation in the manufacture, we do see since that get improved a little bit better now.

I think it was a logical place to have questions and Ken’s comments are great. I would just add to that, keep in mind, the -- this is why we provide some of the metrics there. We’re not -- there’s not airplane orders, right? These are relatively small dollar items compared to what you may see in other industries. And that’s why we gave some of the metrics on the size, if you will, and the fact also that their existing customers and many of those have been partially fulfilled. And we all have the same concern and the question is, how do you get comfortable that that backlog is sticky and it’s going to be here when the product and the supply chain loses up. And that’s why we’re giving those metrics for people to get some sort of context. But keep in mind these are comparatively to other industries, construction industry, airline, what have you, these are very small dollar amounts.

Also, the age of our backlog probably much better than our competitors. Like Keith mentioned in the last 2, 3 quarters, every quarter, even if we increase the backlog and we fulfill probably like 60%, even over 60% of previous quarter backlog. So, that’s making our age in the backlog also pretty short, few months compared to most other competitors sometimes may take 1 to 2 years to deliver. So, that gives us a pretty good position and also the customer trust continue working with us during this supply chain issue. And at the same time, we offer quite a broad product. There’s always some kind of alternative product because suggest out customer to use if 1 or 2 provide some shortage. I think we deliver over 90% of any booking in every quarter, I believe.

Yes, the bookings...

Between 90% to 95% of the booking, we deliver the product.

Our next question comes from Andrew Nowinski with Wells Fargo.

Congrats on another great quarter. I had a question on free cash flow. So I think you said you expect the low-end supply appliances to dramatically improve in Q3. So is there a margin or free cash flow impact that mix shift in Q3? And then related to that, given the shortfall on services that we saw in Q2 and the negative impact it had on your free cash flow, should we expect free cash flow to rebound in Q3 and Q4, or are you assuming the linearity remains unchanged in those quarters? Thanks.

Yes. I’m assuming the linearity is -- I don’t really have any reason to think that’s going to be any different. I’m looking for a reason to find, but I certainly have not found one yet. So, when we look at what our expectations are internally, we don’t guide to free cash flow, we’re trying to give information is helpful to others. I would assume that -- I have no reason to assume anything other than we’ll still see more of the same, if you will. And then, I think the first part of your question was -- you asked about low end and about margin. And maybe I can offer a little bit of commentary there. When you look at our FortiGate firewall product families, the entry level, the low end, you call it, mid-range and high end. In general, the margins increase -- the gross margin increases as you move up from the entry level to the higher end of it. So from that aspect of it, and that’s why the comment in the script that there can be gross margin volatility both from the pricing actions that we’ve taken and the discounting as well, but also the mix of our product. So, in a quarter that we see a higher mix of higher-end firewall shipments, margins will be higher by definition. But there’s many puts and takes in there that we -- when we go through the gross margin guidance that we give, hopefully, we’re considering all the different puts and takes that are in there, not just the mix of the inventory and the pricing actions.

Our last question comes from Roger Boyd with UBS.

Keith, I was curious just to go back to the backlog for a second. You had mentioned the split being about 50-50 between FortiGate and networking portfolio. Just wondering if you could talk about how you expect that mix to trend through the end of the year? And I guess, the follow-up to that is what you’re seeing around the supply constraints between those two product portfolios? Thanks.

Yes, I might double check the numbers. I think it was 50%, 50-40 between FortiGate and networking equipment. I have it backwards. Networking equipment is 50% and firewalls 40% and then cash and dollars the remainder of it. I think everything that we’ve seen to read in here can probably know more, the pressure certainly seems to be for the term, more intense on switches and access points than they do on firewalls. And for a lot of the reasons I think we’re more successful with firewalls.

Yes, I agree. Probably on the direction wise, we do see the FortiGate inventory will keep improving. So probably the percentage, maybe a little bit more on the backlog, probably a little bit more towards the switching and AP side, which is probably the whole industry is suffering some play issue, because we are more able to redesign and also use our own ASIC, which is also helping kind of reduce the backlog and supply on time for the customer. But it’s -- from the beginning of this backlog issue almost one year ago, definitely, we see the shifting a little bit more towards the networking side that have longer backlog.

I would now like to turn it back to Peter Salkowski.

Thank you, Hope. I’d like to thank everyone for joining the call today. Fortinet will be attending investor conferences hosted by KeyBanc, Citibank, Evercore, Stifel and Goldman Sachs during the third quarter. Fireside chats will be available through our IR website. Please let me know if you have any follow-up questions, feel free to contact me and have a great rest of your day. Thank you very much.

Thank you for your participation in today’s conference. This does conclude the program. You may now disconnect.