瞻博網絡 (JNPR) 2020 Q2 法說會逐字稿

Greetings, and welcome to Juniper Networks' Second Quarter 2020 Earnings Results Conference Call. (Operator Instructions) Please note, this conference is being recorded. I would now like to turn the conference over to your host, Jess Lubert, Vice President and Head of Investor Relations. Thank you. You may begin.

Thank you, operator. Good afternoon, and welcome to our second quarter 2020 conference call. Joining me today are Rami Rahim, Chief Executive Officer; and Ken Miller, Chief Financial Officer.

Today's call contains certain forward-looking statements based on our current expectations. These statements are subject to risks and uncertainties, and actual results might differ materially. These risks are discussed in our most recent 10-Q, the press release and CFO commentary furnished with our 8-K filed today and in our other SEC filings. Our forward-looking statements speak only as of today, and Juniper undertakes no obligation to update any forward-looking statements.

Our discussion today will include non-GAAP financial results. Reconciliation information can be found on the Investor Relations section of our website under financial reports. Commentary on why we consider non-GAAP information a useful view of the company's financial results is included in today's press release. Following our prepared remarks, we will take questions. (Operator Instructions)

With that, I will now hand the call over to Rami.

Good afternoon, everyone, and thank you for joining us during these difficult times. Like many of you on this call, we are continuing to navigate the COVID-19 pandemic and taking actions to both meet the needs of our customers and ensure the safety of our workforce. Most of our employees are continuing to work from home and successfully leveraging the various technologies enabled by the network to maintain a high level of productivity despite the current environment. To this last point, the strategic importance of the global network has never been clearer. And I believe the long-term outlook for the markets we serve remain positive. We are investing to not only survive the current environment but to capitalize on the opportunities our markets present and come out stronger on the other side.

Now on to our results. We delivered better-than-expected results during the June quarter with revenue and non-GAAP earnings per share of $1.086 billion and $0.35 respectively, both exceeding the midpoint of our guidance. Overall orders experienced mid-single-digit year-over-year growth with double-digit improvements in our Cloud and Service Provider segment, more than offsetting a mid-single-digit decline in our enterprise business. We are entering Q3 with strong backlog and remain optimistic regarding our ability to navigate ongoing supply chain disruption.

We are executing well in the current environment. While the COVID-19 pandemic continues to present challenges, we believe we are successfully meeting the needs of our customers and helping many of them deliver the critical bandwidth required to support the global economy as millions of people around the world work from home and increasingly leverage cloud-based services. We remain optimistic regarding our competitive positioning and our ability to capitalize on some of the large industry transitions that are likely to play out over the next few years.

On this last point, I'd like to highlight that we secured our first 400-gig win during the June quarter, with opportunities that span across each of the verticals and geographies that we serve. While our initial wins are for wide area use cases, these opportunities represent net new footprint, and increased confidence in our ability to deliver the system density, programmability, power footprint and software needed to gain share in both wide area and data center use cases. While our pipeline of 400-gig opportunities is healthy, and we are encouraged by recent wins we've secured, many of the bigger opportunities we are targeting have yet to be decided.

Based on our latest customer conversations, we continue to expect our 400-gig revenue opportunity to begin in earnest during early 2021 and become a more material driver through the course of the year. Despite healthy momentum entering the second half of the year, the macro environment remains very uncertain and our longer-term visibility remains limited, particularly with respect to the trajectory of our enterprise business. As a result, we are continuing to offer limited full year guidance and would encourage you to build your models conservatively.

Now I'd like to provide some additional insight into the quarter and address some of the key developments we are seeing within each of our core verticals. Starting with Cloud, we experienced healthy results during the June quarter, as the business was up slightly and grew for a fifth consecutive quarter despite a more difficult year-over-year comp. We continue to see momentum within our customers' wide area networks, particularly for some of our routing products. Order trends remained healthy, with good momentum at multiple hyperscale accounts as well as with our Tier 2 cloud customers.

Highlighting the increased diversity of our hyperscale business, our largest cloud customer in the June quarter was different as compared to the March period. Based on our Q2 results and recent quarters, we still expect to see low to mid-single-digit cloud growth in 2020, although we would expect to see some seasonality during the September quarter. We maintained strong, durable franchises at each of our hyperscale customers and should be well positioned to benefit from continued capacity growth in the use cases we own.

While our Service Provider revenue modestly declined during Q2, this business continued to be most impacted by our supply chain challenges with orders seeing a second consecutive quarter of year-over-year growth. Although we are seeing some COVID-19 related capacity benefit, we believe much of the Service Provider order strength we experienced is attributable to our diversification effort across customers and products over the last few years. To this point, we are continuing to see improved momentum with several of our U.S. cable customers as well as Tier 2 and Tier 3 carriers in international markets. We are also seeing increased carrier adoption of our Switching and Security offerings in addition to our traditional routing platforms.

We believe we remain well positioned with our Service Provider customers and that our continuing efforts to diversify our customer base and increase the breadth of our offerings should benefit this business through the remainder of the year. While we acknowledge that some of our Service Provider customers are continuing to face business challenges that may impact their ability to spend in future quarters, based on our recent momentum and customer conversation, we continue to believe our Service Provider business is likely to see a mid-single-digit decline in 2020.

Our Enterprise business slightly declined year-over-year, but exceeded our initial expectations for the period. Strength in the financial services business as well as with some of our larger strategic accounts more than offset weaker-than-anticipated results from our U.S. federal business, which was impacted by COVID-19 timing dynamic that we expect to reverse during the current quarter. While our overall Enterprise business is being impacted by the uncertain macro environment, which has caused some customers to reevaluate plans, we're continuing to see very strong momentum with Mist, which is driving an increasing level of confidence in our ability to gain Enterprise share and return to growth once the pandemic subsides.

To this point, I'd like to highlight that Mist reported another record quarter with orders rising more than 170% on a year-over-year basis and new logos increasing by more than 100% year-over-year. Mist has now secured 4 Fortune 10 accounts, and we saw a material increase in demand generation from the channel, reflecting the true differentiation of the product. In addition, Mist launched a new software subscription premium analytic that has generated strong interest due to its ability to enable use cases such as proximity tracing, journey mapping and hot zone alerting to helps enterprises enable social distancing and keep employees safe as they start returning to work.

Our Enterprise at Home offering also generated strong interest from enterprises due to the increase in work from home. While Mist is continuing to exceed expectations, our strategy to Mistify additional elements of our switching, enterprise routing and security portfolio through the years is helping us take share from competitors and should create incremental pull-through opportunities for our Enterprise offerings in future periods.

Our Software revenue declined in the quarter and accounted for less than 10% of sales due to a lower mix of certain products that drive higher on-box attach rates of perpetual licenses. That said, our software orders grew 7% year-over-year due to a combination of strong Mist and Security subscription as well as our efforts to transition certain perpetual software offerings to term-based subscription. We believe growth in these recurring software offerings is an encouraging dynamic that should improve visibility over time and give us confidence in the long-term outlook for our software revenue.

I'd like to mention that our Services team delivered another solid quarter and continued to grow on a year-over-year basis due to strong renewals and service attach rates. Our Services margins continue to improve year-over-year, and our customer satisfaction rates are currently at record levels. Our Services team continues to execute extremely well and ensure our customers receive an excellent experience. I would like to extend my thanks to our customers, partners and shareholders for their continued support and confidence in Juniper. I especially want to thank our employees for their hard work and dedication, which is essential to creating value for our stakeholders.

I will now turn the call over to Ken, who will discuss our quarterly financial results in more detail.

Thank you, Rami, and good afternoon, everyone. I will start by discussing our second quarter results, then provide some color on our outlook. We ended the second quarter of 2020 at $1.086 billion in revenue and non-GAAP earnings per share of $0.35, both above the midpoint of our guidance range. We experienced strong demand in the quarter with orders growing mid-single digits year-over-year, exceeding our expectations. Revenue was down 1% year-over-year. As expected, supply constraints resulted in extended lead times throughout the quarter.

Looking at our revenue by vertical, on a sequential basis, all verticals grew, with Service Provider growing 16%, Cloud growing 9% and Enterprise growing 1%. On a year-over-year basis, Cloud grew slightly year-over-year, while both Enterprise and Service Provider declined 2%. From a Technology perspective, Routing and Switching decreased 3% year-over-year and Security decreased 1% year-over-year. Our Services business increased 1% year-over-year. As Rami mentioned, Software revenue was below 10% of total revenue for the quarter and declined year-over-year; however, software bookings grew 7% year-over-year.

In reviewing our top 10 customers for the quarter, 6 were Cloud, 3 were Service Provider and 1 was an Enterprise. Non-GAAP gross margins were 58.3%, below our expectations primarily due to higher-than-anticipated COVID-19 related logistics costs. If it weren't for the COVID-19 elevated logistics costs, we would have posted non-GAAP gross margins of approximately 59.5%. Non-GAAP operating expenses were flat year-over-year and declined 3% sequentially, which is in line with our guidance range. Our operating expenses in Q2 benefited from COVID-19 related savings.

Cash flow from operations was $98 million. We paid $66 million in dividends, reflecting a quarterly dividend of $0.20 per share. Total cash, cash equivalents and investments at the end of the second quarter of 2020 was $2.6 billion, slightly up from the first quarter of 2020.

Now I would like to provide some color on our guidance, which you could find detailed in the CFO commentary available on our website. At the midpoint of our Q3 guidance, we expect to see sequential revenue and earnings growth. Confidence in our forecast is driven by strong backlog and strength within our Service Provider and Cloud verticals. We believe these factors should help offset continued uncertainty in parts of our Enterprise market. We expect to see sequential volume-driven improvements in our non-GAAP gross margin and a more favorable customer mix during the September quarter.

We expect logistics and other supply chain-related costs to remain elevated, consistent with Q2 levels due to the effects of the ongoing pandemic. We expect third quarter non-GAAP operating expenses to be essentially flat compared to Q2, as we continue to benefit from lower travel costs due to COVID-19. We will remain focused on prudent cost management while continuing to invest to capture future opportunities.

Turning to our capital return program, our Board of Directors has declared a cash dividend of $0.20 per share to be paid during the third quarter. We remain committed to paying our dividend, and will remain opportunistic with respect to share buybacks. In closing, I would like to thank our team for their continued dedication and commitment to Juniper's success, especially in this challenging environment.

Now I'd like to open the call for questions.

(Operator Instructions) Our first question comes from the line of Rod Hall with Goldman Sachs.

I wanted to start off with the Service Provider trends, those are way better this quarter than they have been for many quarters, actually, that sequential on that up a lot and the year-over-year down not much. So I wondered if you could dig into that a little bit more, talk about how sustainable that trend is and maybe what drove it, a little bit, in terms of color? And then I've got a follow-up to that.

Yes. Thanks, Rod, for the question. I mean certainly, we're pleased with the results that we saw for service provider in not just Q2, but in the first half of the year. Revenue doesn't point to the whole story. As we mentioned, orders in the Service Provider vertical were up double digits and I think there are a number of different factors. There is an element of this that is COVID related. Customers that are trying to deal with an increase in network capacity or customers that are trying to get ahead of any potential for supply chain disruptions. That's an element of it.

But I actually think the bigger factor at play here are a couple of things. It has to do with, first, customer diversity. We have, as a matter of strategy, been diversifying our reach within the telco segment to Tier 2 and Tier 3 telcos, both here as well as internationally. And then there is a technology diversity element to it. We saw a really strong Switching momentum among our telco customers and Security. We have a strong mobile and especially 5G-related Security solution that's seeing some really great demand with our customers. So certainly, I think that to the extent that we're in a new normal relative to telcos spending, dealing with capacity constraints, that should last. And then, there are -- there is a meaningful element that's just not COVID related at all and that's really a matter of execution, and that's where I just think we're executing really well, Rod.

Okay. And then, on the follow-up, I wanted to ask you guys about the supply chain elongation of lead times and whether -- I know you started talking about that last quarter, and it continued this quarter. And now you're saying it will continue the following quarter. Is it a case now where revenues are kind of normalized for that? Or do you -- so we just keep pushing revenue out into the future because of these supply chain disruptions? Or do you think you're actually losing revenue as a result of it? Can you just comment on the effect on the kind of what we should expect in terms of the underlying revenue trajectory? And if you've got any thoughts on when that might ease in terms of impact to you, that would be interesting as well.

Yes. So we are continuing to manage through various supply chain challenges that we talked about last quarter. Lead times did remain extended throughout the quarter. Normal lead times for us on average is about 2 to 4 weeks. We're seeing today's lead times or Q2's lead times, I should say, closer to that 4- to 8-week on average level, so you can see the level of extension there. We did see some stabilization, quite a bit of it towards the end of Q2, and we do expect things to stabilize throughout Q3 and actually get better over the second half. We just don't have perfect line of sight as to when those improvements are going to happen.

So we're not factoring in improvements into our Q3 guide; however, clearly, the team is driving to get improvements each and every week. So we do expect improvements in the second half, just not factoring into Q3 guide, given all the uncertainties involved. From a revenue perspective, I would say it's a similar dynamic where you saw the bookings growth in the quarter, but revenue was modestly down similar to what we saw in Q1. When we start to actually ship more backlog than we book, you will see revenue growth be faster than bookings. But for Q2, we saw kind of a similar dynamic that we saw in Q1, which was more of a backlog build.

Yes. And just to add, Rod, to the question, I don't believe we're losing as a result of this. There's no doubt it's a challenging situation, but I don't think it's unique to Juniper. And I think our customers recognize that this is an industry-wide challenge that we're dealing with here. Couple that with the fact that it's -- these products are not typically interchangeable from one technology provider to another technology provider, depending on the particular use case that we're -- that you're addressing.

Our next question comes from the line of Tim Long with Barclays.

If I could just start off with the 400-gig side, Rami, you talked about a win there and some larger opportunities still to come. Could you just give us a sense as those start to come in or materialize next year, how are you thinking about the breadth of that business and where it'll be playing? And curious if you can just give us some of your thoughts on Nokia's announcement to kind of go after this market a little bit more, does that change the dynamic? And then I had a separate follow-up.

Yes. Thank you for the question, Tim. So 400-gig has been a big bet for us for quite some time. This is an area that we've invested in. And as I have mentioned previously, we've introduced really a brand-new technology stack that includes new technology systems, and we've also revamped significant portions of our operating system to make it essentially cloud-ready for 400-gig deployments. There have been early wins, and we're very pleased with that because of the fact that they have been competitive wins. They have been wins that have taken us into net new networks.

I know at least a couple of those wins went into WAN for Service Providers as well as WAN for Cloud providers, where they are essentially net new use cases that we have now been selected to deploy. So although they are not all that meaningful yet from the standpoint of revenue for our business, to me, the important thing is that it gives us the confidence that that new technology stack that we've brought into the market is working, is working well, and it's very competitive.

So that leads me to the next part of your question around the competition. We're very used to operating in an environment where there are competitors, strong competitors. And I'd just say that we're very confident that the technology stack that we have is competitive. We're seeing the early proof points. And then add to that, I think, the footprint that we have among telcos and the cloud providers that truly gives us the opportunity to leverage, to launch into 400-gig use cases and deployments, I think, is something that we can absolutely use to grow this part of our business.

Okay. Great. And then just a follow-up on the enterprise side, it sounds like Mist is doing well and you guys have obviously had a good -- big push there. Could you just talk, Rami, a little bit about, obviously, the change in dynamic on what the workspace or the enterprise or the campus is going to look like from a number of people. So do you think there's an impact on your growth, medium, longer-term based on the slope of employee repopulating and the scale to which there's enterprise? Or do you think the growth rate there for you is unaffected?

Yes. It's a great question and obviously, one that we've thought long and hard about. The net of the answer to your question is that whereas in the short term, there is going to be some disruption and there's going to be some challenges as a result of the macro situation. I do believe that as we emerge from the pandemic, we will see actually strong momentum and growth, both in the markets that we serve and in terms of the strength of our technology and our ability to take share. And the reason being that we focus not just on your standard legacy based Enterprise wireless LAN deployments, we're really focusing on an end-to-end solution that includes wireless LAN, LAN and WAN. And many enterprises, even today, are thinking about how to transform their businesses to take into account or to take advantage of these cloud-delivered AI-driven solutions.

I think our solution here is very strong. Honestly, the win rate is phenomenal even in this environment. So yes, we saw some weakness in the Q2 time frame. It was still better than what we expected and I believe we're actually taking share on the strength of our portfolio. The last comment I will mention here is, we really are addressing solutions that are very pertinent to what our customers are thinking about today. So our customers want safe return-to-work solutions that will allow them to understand if there are hotspots in their campuses or if there is, in fact, an issue with an employee getting sick, they can understand where that employee has been and who needs to be notified. These are solutions that really go to the heart of the kinds of problems that need to be solved as companies start to return to work.

Our next question comes from the line of Alex Henderson with Needham.

I was hoping you could talk a little bit about your comment on the Enterprise business. Specifically, you stated that the U.S. federal business was well off in the quarter, but you expect it to rebound, and it seemed somewhat countervailing to the trajectory of spending in government and state and local, given the clear physical constraints that are developing as a result of the COVID. What gives you confidence that that's going to recover and improve in the third quarter?

Well, first, the third quarter is traditionally -- it's the fiscal year-end for the federal government, so it's typically been a stronger quarter. There's that dynamic, but I think it's more than that. We've obviously had our business reviews and we're looking at our pipelines, we're looking at projects. And in all of those dimensions, there are opportunities out there. There are projects out there. We don't believe -- although we're monitoring very closely, we don't believe that the projects have been canceled or even moved out in any meaningful way. There has been disruption as a result of COVID to the short-term timing of the project, but still remain pretty optimistic about this part of our business.

The follow-up question is, obviously, Enterprise is a very important piece of the puzzle and a driver going forward. Mist is a key piece of that. It looks like your Mist business is booming, yet your Enterprise business was soft. Is that a function of the timing differential between current revenue recognition and the incremental pull-through? When would the pull-through metastasize post a Mist order? Is that lagged somewhat?

Yes. So we do -- the Mist model is about 60% of the revenue is recognized upfront in the form of hardware and then there was a good amount that's deferred in a SaaS-delivered license -- cloud-delivered license. So there is a fair amount of deferred revenue in a Mist transaction. I would just also highlight our Enterprise business is quite large and has many use cases, data center use cases, campus brand switching, security, routing, et cetera. So while Mist is very fast-growing, it is still relatively small as compared to the rest of our Enterprise portfolio at this point. But it is growing quite nicely, obviously, and it is pulling through other solutions as part of the Mist sale.

The question was the timing of the pull-through relative to Mist orders. When orders come in, how long a lead time before you get a pull-through for other products?

I see. Yes, first, I can tell you that when we sell Mist, we're really selling a portfolio. I know Mist when we acquired the company, it was a wireless LAN solution, but when we've combined with Mist, we really did it because the strategy was to take that cloud-delivered AI-driven engine and to extend it across our entire portfolio. We've started to do that with our wired switching and then now we're starting to incorporate our WAN transformation, our SD-WAN solution.

In at least 3 of the 4 Fortune 10 accounts that I mentioned in my prepared remarks, those are combined wireless LAN and LAN switching under the Mist umbrella. But when you have a break-in, especially in a large account, typically, it's a small break-in to begin with, and then you build on it over time. The glue, the real secret sauce here is the cloud-delivered AI-driven experience to our customers. Once you've sort of captured and sort of impressed your customers with that solution, it becomes much easier to scale over time. So I'd say we're early days in terms of reaping the full benefit or the full potential of these solutions.

Yes. And it's hard to quantify precisely but our models think it's a 3- to 6-month kind of pull-through effort to really start to see the leverage of the Mist pull the rest of the portfolio through.

Our next question comes from the line of Ittai Kidron with Oppenheimer.

Just a couple of questions from me. First on the Enterprise. Rami, can you talk about the linearity in the quarter? Was this kind of an even quarter or started strong, ended soft? I know it sounds like a lot of the issues you've had with Enterprise last quarter feels like a repeat this quarter. I'm just trying to understand from a pattern standpoint, did it end on a positive note or not?

I think it was normal. I'll look to Ken here, but I think it was -- there was nothing out of the normal in terms of the linearity of the quarter. And as I mentioned, the results that we saw in the Enterprise were really as expected or maybe even a little bit better than what we expected considering the dynamics. It's left to be seen, but based on our win rates, and our competitive positioning, I believe we're taking share in the Enterprise.

Got it. And then as a follow-up on the software side, that it was down there and we've missed clearly having a good quarter. Can you help me understand what components of software did not do well? Is that the on-box software that I assume somewhat underperformed in the quarter?

Yes, so let me give you a couple of statistics on our software business. Off-box orders increased by 50% year-over-year, and then off-box subscription orders increased by even more than that, more than 50%. And then SaaS subscription increased by over 85% year-over-year, so the off-box metrics are very healthy. The weakness was entirely a result of on-box Flex licenses and it just so happened that many of the products that we sold in the Q2 time frame were products that did not yet benefit or don't have the Flex on-box licensing component.

Yes, just to clarify, it was actually some of the on-box perpetual licenses that we sold as disaggregated model. The Flex license model is actually ramping within the on-box. Those are the term-based licenses. We are seeing an increase in term-based subscription licenses, but the perpetual disaggregated hardware-software licensing scheme that we had in the past, we saw a decline in that model.

Just so maybe I can add to that, Ken, on that front, given that the component that was soft was the perpetual, is there anything to be learned or to deduct from that with respect to the mix of chassis versus blades or something like that? Is there a correlation there between that and in your perpetual mix?

It's really not a chassis versus blade thing. It's really a transition from legacy perpetual into more term-based subscription, recurring revenue models that we're driving. And we did see good success in those metrics, that's really what's important to us. So we expect to grow software from Q2 as we enter into Q3.

I mean -- and just to add, our strategy has always been to grow the number of platforms that will leverage our Flex licensing model, and that certainly is going to help in the up quarters.

Absolutely.

Our next question comes from the line of George Notter with Jefferies.

I guess I wanted to ask about the Switching business. I know you guys have been working on a transceiver development that was kind of feeding into your 400-gig efforts. And obviously, you're having a bit of success now in 400-gig in terms of customer wins. But can you talk about that transceiver development, where is that? And is it translating at the win level?

The short answer to, is it translating to our wins, is no, because it's not actually shipping yet. This is a pretty ambitious project where we believe we have truly differentiated technology that will achieve a level of integration in optical transceivers that is better than anything else that's out there. But it's a difficult project, a challenging project and one that we are working through. Our objective, our goal has always been to release products in time for the ramp of 400-gig next year, and that's what we're working towards. The success -- the early success that we're seeing in 400-gig, quite frankly, is based on the merits of our software, our silicon and our system strategy. Transceivers, our optical transceivers, once we ship them would really be sort of an icing on top, not mandatory for our success in 400-gig.

Our next question comes from the line of Samik Chatterjee with JPMorgan.

I just wanted to kind of dig into the cloud vertical a bit here and have a couple of questions on that. Firstly, you had a good quarter with the cloud customers, but when I look at the product segments there, it seems like the strength really came from routing and sort of switching. So just wanted to understand kind of what kind of insights can be drawn in terms of what is driving the spend from the cloud customers with your product there? Is it more of the upgrade on the kind of the certain layers that's not really benefiting you on the switching side? Just wanted to understand the dynamics there. And I have a follow-up.

Yes. Thanks for the question, Samik. So just keep in mind that our switching solutions that we sell into the hyperscale Cloud Provider space are actually used in wide area networks. So this is something that we've talked to you about in past quarters as well. It remains true today. So to think of switching in sort of the traditional sense of switching, it's not really the use case that's deployed by hyperscale Cloud Providers. All up routing and wide area use cases within the Cloud Provider segment, especially hyperscale Cloud Providers, was very strong. We saw a meaningful order strength on a year-over-year basis. And that's, I think, a function of the existing deployments or the footprint that we have and the strength of our portfolio.

And then just following up on the 400-gig win here. I understand, it seems like it's one of the smaller cloud providers. But in terms of timing, are you expecting most of the -- based on your visibility, are you expecting most of the Tier 2 to kind of have a similar timing in terms of their upgrade cycles? And any updates on how you're thinking about the timing for the hyperscalers as well on 400-gig?

Yes. I believe that it's going to be very much based on the economics and the availability of 400-gig optics. And that's very much sort of an end of year and really getting into revenue growth next year. So I believe that's true for both hyperscale as well as some of the smaller cloud providers.

Our next question comes from the line of Tal Liani with Bank of America.

I went back to previous quarters, and 2Q is always strong. I mean that was -- 2 years ago, it was up 11% sequentially; last year, 10%; this year, 9%. So I'm trying to understand why you're focusing on the sequential trends and not on year-over-year trends because last year, the comps were not very difficult. The comps -- the year was down 8% -- the quarter was down 8% year-over-year. So the question is what drives you to be positive, given that year-over-year, the growth is minus 1.5% and higher declines of products? Are there any projects that you were waiting for that are starting to materialize? Or can you link some of the trends you see beneath the surface that maybe the numbers don't tell us, can you link it to some bigger kind of agendas of carriers like 5G or anything that you can refer to?

Yes. So why don't I start then -- and pass it to you, Rami, to add more color. But a lot of the commentary is based on -- as compared to the guidance and expectations we had when we entered the quarter, clearly those numbers -- those expectations were lowered as compared to when we entered the year. I mean, we absolutely have a long-term principle to grow top line and to expand bottom line. That was our expectation this year as well, not too long ago, but kind of in a pre-COVID world, we felt we are on track for that.

With COVID, we do believe it could have a negative impact on our business, particularly, in Enterprise. And so we reset expectations, if you will. And we are ahead of those reset expectations. I think that's the main takeaway. We did see year-over-year growth in bookings, which is a great sign. From a revenue perspective, it was modestly down year-over-year, but the opportunity to deliver against the expectations we have internally now, I feel pretty good about. Rami, do you want to add more color there?

Nothing other than just that you're asking about underlying trends. I think the order strength and the order momentum is that trend, a double-digit growth in Cloud Provider, double-digit growth on a year-over-year basis in the Service Provider segment. And like I said, I think we're executing well in a challenging dynamic environment on the Enterprise side based on just the strength of our portfolio.

Got it. I don't know if I have time for another question, but is your outlook for routing better now than before? And let's put COVID aside, routing has been going on under pressure for more than 5 years, 7 years and the question is whether we got to a point where routing finally starts growing because there's just a need for bandwidth processing power?

Well, let me put it this way. We've said Cloud Providers should grow mid-single-digit for the year all up on a full year basis year-over-year and Service Provider would be sort of mid- single-digit decline. We're off to a very strong start in both of those segments. We're not yet ready to change our full year perspective, but let's just say I'm very encouraged by the momentum that we've seen to date.

Our next question comes from the line of Simon Leopold with Raymond James.

I want to see if maybe you could unpack the cloud customers a little bit. I think I heard you mention in the prepared remarks that the top cloud customer was different this quarter than the last quarter. And what I'm really trying to get an understanding of is what kind of concentration you have within this particular group of customers and maybe the mix of what you classify as Tier 2 versus hyperscale? Just trying to really get down to understanding diversity in that. And then I've got a follow-up.

Yes. Let me see how I can answer that. There's always a high degree of concentration when you're talking about hyperscale, just by virtue of the fact there are a very few number of hyperscale customers. I mean we measure them top 5 by their revenue, right? Their own services revenue. So within hyperscale, the point that we just wanted to make is, we enjoy really unique, solid footprint with a number of them. It's not a one hyperscale customer story. And the fact that in the last couple of quarters, we've had 2 different cloud providers, hyperscale cloud providers emerge as the #1 cloud provider customers in a particular quarter is, I think, indicative of that.

And then there's the broader cloud provider opportunity where we've really focused on broadening our reach into Tier 2, Tier 3 as well as into the large cloud providers in Asia, in China, in particular, where we have seen some success. So that's how I'd respond to your question around the concentration of the opportunity and the diversity of our opportunity.

And to just clarify, how material would be Tier 2s in that overall cloud number? Is it sort of roughly 1/3 of that? Is that the way to think about it? I guess I'm looking for how material Tier 2 is?

Yes. We haven't broken down exactly Tier 2, but what we have said previously, the top 10 customers account for approximately 80% of the overall vertical. It's a very concentrated vertical. So that gives you a good -- a pretty good feel for kind of how big our Tier 2 business is.

And if you were to look at that from the standpoint of just CapEx spend by those customers, it would look probably similar.

And then the follow-up I had was regarding the MX. You've announced a series of refresh elements, new line cards. And I have the impression that a lot of the certification evaluations maybe took longer than you once expected. And I want to see if you've got a sense of whether or not the completion of evaluations by your Service Provider customers of MX refresh is contributing to the improved router growth, if this is sort of what we've been looking for, if that's still on the come or whether that occurred and didn't really move the needle?

Yes. It's a good question. In both MX and PTX, we saw very strong sequential growth as we expected from a revenue standpoint, and we saw solid bookings growth on a year-over-year basis. And yes, part of the momentum in the MX is that we're starting to see a ramp in some of these new line cards that are now -- that have emerged from the certification phase and they are going into the early revenue phase.

Our next question comes from the line of Meta Marshall with Morgan Stanley.

Great. Two questions for me. Maybe on kind of the campus business, have conversations started with customers as to what changes of investments are potentially going to take place if a greater portion of their employees work from home, either additional in-home equipment or less kind of on-campus equipment? And then maybe a second question, just on the Tier 2 cloud. Clearly, a lot of those SaaS providers saw a huge increase in traffic due to COVID-related instances and has there been any change in conversation as to whether they'll continue to build their own data centers versus leveraging other public clouds?

Yes, 2 great questions. First, on the campus opportunity. I think there are a couple of things that we should keep in mind. First is, we're focusing on enterprises that are large that typically have businesses that are going to be more resilient to the pandemic crisis. Government, public sector, higher ed, are all -- even retail, large retailers that have businesses that are resilient, especially those that are in e-commerce that are resilient to COVID are areas that we are very much deliberately focusing on. And that's helping our business in the current state and I think it certainly will help us post-pandemic. The second element is that many of our customers today are looking for next-gen cloud-delivered AI-driven solutions, not sort of the legacy on-prem stuff. They're really looking at transforming their wireless LAN, LAN and WAN solutions. And that's true now and it's certainly true for post-pandemic, and those are the opportunities that we're sort of laser targeting right now and we're seeing some very good success there.

On Tier 2 and Tier 3 cloud and sort of shift to public cloud, it's a mixed bag. I think there are puts and takes. There are definitely some Tier 2s and 3s that will, in this time, in particular, pivot to using public cloud and honestly, we'll benefit from that through serving the public cloud, the hyperscale cloud providers. But there are a lot of Tier 2, Tier 3 cloud providers that we have benefited from over the last couple of quarters with momentum in this part of our business that are sticking to their own solutions because they believe that's just going to be the most economical way forward for them.

Our next question comes from the line of Ryan Koontz with Rosenblatt Securities.

I wonder if you could comment on your strength in APAC, both year-over-year and Q-over-Q, it looks like some fairly easy compares there, but how much of that do you attribute to like 5G programs in Japan, Korea? And then the second question, kind of, more broadly, how do you see 5G driving metro upgrades globally in some of your top SPs?

Yes. Great question. So I'm happy with our results in APAC. In fact, sequentially, all of our geos were up, so we saw good momentum sequentially across Americas, EMEA and APAC. APAC, you're right. I mean, it's working off of an easier compare after several quarters of decline. I think we're executing well in APAC. I think Asia Pacific is also a little bit ahead of the U.S. when it comes to dealing with the pandemic and emerging from the economic turmoil that we're benefiting from. It is -- a lot of the strength is telco-related, and I do believe that some of it is 5G-related. And in particular to 5G, there are a number of ways that we benefit. One is in transport, whether it be metro build-outs or edge and core opportunities. The other one would be in Security, where we have seen, now for a number of quarters, real strength in our Security business, specific to telcos and especially telcos that are starting to build out their security infrastructure in preparation for 5G. And then, the last dimension of success that we've seen, again, not just in APAC, really worldwide, but includes APAC, would be in our telco cloud solutions. 5G will be a cloud-native solution. Many of the services that will be offered in 5G will essentially be offered in software in a virtualized form, and we have a very strong solution for that today that we're benefiting from.

Helpful. A little follow-up there on the telco cloud. Are you seeing more of a trend kind of swinging back towards telcos looking to outsource that cloud? We've seen announcements from Microsoft and Amazon there? And how do you see that playing out in the telco NFV cloud?

Yes. It's also a very good question. Again, there, I think it's very much a mixed bag. There are going to be some telcos that are comfortable and want to work with the hyperscale cloud providers. And again, there, I think we benefit in indirect ways. But there is definitely an opportunity that we see where telcos really want more of a -- I wouldn't say build it themselves, but to partner with technology providers like Juniper and others, to put together telco solutions to offer differentiated edge services where they have the best locations already in their networks available to deliver those services. And that turns out to be an area where we have a lot of experience, some great technology and good momentum with the telco customers.

Our next question comes from the line of Amit Daryanani with Evercore ISI.

This is Irvin Liu dialing in for Amit. I had a question and a follow-up as well. My first question is on your off-box software subscription momentum driven by Mist. Can you talk about the originations of some of these sales? Are you seeing these wins from mid-cycle add-ons? Or is this more an upgrade cycle driven story? And longer term, should we still expect software revenue to trend above 10% of revenue?

Yes. So on the first question, a lot of the Mist momentum we're seeing are net new wins and net new logos, so there is essentially the introduction of SaaS-based subscription services into net new accounts. There are certainly some renewals that are out there. I don't know the exact numbers off the top of my head, but a lot of the Mist momentum has been as a result of the 100% year-over-year growth in net new logos, new opportunities that we are winning. And the second question, I'll pass to Ken.

Yes, on the 10% of the revenue, that's -- clearly, our goal is to grow software over time. We have goals beyond 10% over the next couple of years. As far as going forward, I do expect us to grow in Q3 off of Q2 levels, but really no commitment yet on the 10% going forward other than, yes, we do expect to get back there and surpass it over time as our subscription business starts to become more and more relevant in overall software story.

Got it. And I actually wanted to ask about the other technology transition, which is Wi-Fi 6 cycle. Can you share your thoughts on when you expect wireless LAN to become a more material portion of your revenue looking forward? Is there a potential for there to be a stand-alone revenue breakout for wireless LAN on your P&L in the near future?

Well, that's a homework assignment for Ken and I. We're not breaking it out now. But I'll say this, Wi-Fi 6 is a great opportunity for us. We already have our Wi-Fi 6 access points in the market today that are selling. But Wi-Fi 6 on its own, I think, is -- there are going to be many vendors that offer it. The AI-powered cloud-delivered Wi-Fi 6 because with Wi-Fi 6 comes a certain level of complexity that operators need to overcome, and you want to offer them a solution that essentially it runs itself. And quite frankly, there's just nothing out there that comes even close to what we have on that front.

So I think we have the right combination of the high-performance Wi-Fi 6 access points as well as the AI-driven solutions that simplify the deployment and the ongoing operations of these solutions that's helping us win large accounts around the world.

Operator, we're going to take 3 more questions.

No problem. Our next question comes from the line of Jeff Kvaal with Wolfe Research.

I was hoping for a little bit more color on the cloud business and the timing of that. At one hand, it sounds as though you've gotten some pretty good orders teed up, which is nice. On the other hand, we should be a little bit careful about seasonality in September. I'm wondering if you could sort of ferret that out for us a little bit.

Yes. Let me start, and then maybe, Ken, you want to add some color. I mean, for cloud all up, what we have been saying is single-digit growth for the full year is sort of a good assumption to make. Obviously, we're encouraged by the momentum that we've seen in the first half of the year. That gives us a lot of confidence in our ability to meet, if not exceed, that outlook that we provided. And again, I think there is a COVID-related element to this, either -- whether it be supply chain-related or capacity related where cloud providers are actually seeing meaningful growth in their networks. But I also think that there is another component, which is not small, a large element of this around execution, Tier 2, Tier 3 clouds, international clouds, strength of our solutions and the unique footprint that we enjoy with our Cloud Provider customers.

Yes. As Rami mentioned, we do expect it to be up on a full year basis, clearly. That said, we do have a seasonality. We've seen that in the last couple of years, where we've seen a little bit of drop in Q3 in our cloud business. We actually expect, obviously, revenue to be up in Q3 at the midpoint of our guidance. If you break it down by vertical, we expect to see growth in Service Provider and modest growth in Enterprise and Cloud to be slightly down. Now those numbers could obviously differ from those expectations, but that's our current expectation at this point.

Our next question comes from the line of Jim Suva with Citigroup.

Rami, in your prepared comments, you mentioned an example of some trends or product selling that didn't quite have the typical, I believe, you said software attach rate that we're used to. Can you maybe clarify a little bit. Is this a trend that you should expect going forward? Why or why not? Or kind of what happened there? And maybe I just didn't catch the full example that you were giving.

It's entirely based on mix of products. There are some products that we sell that just have a larger term-based software component attached to them than others. And it turned out that we sold more of the products that didn't have that software component, and that might change over time. Ken, do you want to add anything?

Yes, I was just going to say, yes, so we've had a software business for a long, long time. That's been largely an on-box perpetual license software business, and it really is additional features and function that we monetized via Junos. What we're doing now is transitioning that business to more of a sustained subscription business with term licenses or even cloud-delivered SaaS licenses over time. And what we're seeing is, depending on what products you buy and how you buy it, you could have differences in the overall revenue for software. And we saw that the on-box perpetual licenses were down, but the rest of the business was actually up in Q2.

And just to add, again, our strategy is to make Flex the one licensing scheme for software on-box and off-box across all of our portfolio. That can't happen overnight. We've been working this methodically throughout our portfolio. And over time, as Flex starts to touch all of the portfolio, it will sort of alleviate some of these ups and downs that we see from a quarter-to-quarter standpoint.

Our final question comes from the line of Paul Silverstein with Cowen.

I am hoping before my cell phone runs out of battery to return to some questions asked by Tal, Simon and some others. First off, with respect to routing, Service Provider routing in particular, I too, once upon a time thought that routing was in secular decline, but looking at the numbers, it's not. I think contrary to what many in the investment community appeared to think, that market actually all in is growing, albeit modestly. And also, for that matter, the same is true of edge and core. You've got 5G impact ahead of you, I suspect in the next 12 to 18 months.

Rami, you mentioned tight capacity leading to increased demand among your Service Provider customers. You've got your new MX line cards that have been in the market now for, I think, a little over a year, this should be ramping. You've got the Ericsson relationship, I assume that, that's starting to pay some dividends. Or whether or not you've got expansion, as you said, a footprint into non-U.S. Tier 2 Cloud, excuse me, non-U.S. Tier 2 Service Providers. But at the same time, you've got Cisco, with some pretty impressive, at least on paper, performance in the 8,000 family ongoing Nokia strength.

So now for the question, which is, I appreciate that you all characterize your Service Provider business, which I assume is mostly routing, not all but mostly routing. I appreciate that you all have been referencing a low single-digit decline for some time now. Why isn't there the possibility -- or isn't there the possibility, given the growth in the market, the coming 5G impact, your new MX line cards, et cetera, for that business to return to some semblance of growth, albeit perhaps not dramatic growth? Or is that just not practical?

Yes. Paul, look, it's a great question. I think there are a lot of trends coming our way that actually could be tailwinds for us, okay? 400-gig is going to be -- there is a Service Provider element to that and what I mentioned earlier around early wins, I mean, there are telco early wins that are net new footprint for us in accounts -- international accounts, in particular, where we have relatively little footprint, little deployment. 5G is going to be another area. And that's something that we capture, not just with routing, but with other elements of our portfolio as well.

I think the reason why we just are sometimes cautious when we talk about telco business is that telcos, in general, the long-term trends around telcos is that they're challenged, their business models are somewhat challenged. They're seeing revenue stagnate. At the same time, they're dealing with growth in their networks, and that puts pressure on their margins, and they put -- they translate that pressure into pressure on their technology vendors like Juniper.

So we have to be a little bit cautious and careful in the long-term modeling and outlook. But certainly, we are doing everything we can from our metro portfolio to our 400-gig portfolio, to our automation solutions for the wide area network, to solutions that span into telco cloud and security to capture that telco opportunity. And telco, of course, still represents 45% of our business, so it remains a very important vertical for us. I'm very encouraged by the strength of the first half of this year. And certainly, I'd say it's exceeded our own expectations.

Rami, before asking a question about margins, my final question. Can you just remind us what percentage of Service Provider is routing and what percentage of routing is Service Provider?

We haven't updated the numbers, but we did in late 2018, and it was about 80% routing at that time.

Service provider is 80% -- 80% is...

Service provider is routing. We haven't gone the other direction, but you can -- that will help you, that would give you the bit -- lion's share of the routing number.

Got it. All right. Ken, a question for you on gross margin. It sounds like your third quarter guidance would be something close to the 60.5% to 61% book for the expedite costs. Is that a correct assumption before I ask the question?

Yes. That is a correct assumption. We saw a 120 basis point impact -- 1.2 impact in Q2, and I expect similar elevated levels in Q3, so it would be approximately the same level of impact in Q3.

All right. The very basic question is, over the next 12 to 24 months, where can you get your gross margin to in a reasonable best case? And can you remind us of what the key levers are in getting there?

Yes. So we put out -- from a model perspective, 58% to 62%. We've been hovering at that kind of midpoint 60% on an annualized basis for the last couple of years. The path to growth is many -- multifaceted, I should say. Clearly, Software is a big part of that strategy. And as we shift more of our business to Software, especially recurring software over time, that will carry a margin lift for us. We're doing a lot from an engineering perspective and the design for value and value engineering front. We've talked about this in the past where, particularly, in our Switching lineup, we didn't have as competitive a cost of goods sold as we thought we needed. We think we fixed that, and we're excited about the opportunity ahead, particularly in 400-gig, on that front, making sure we get the cost per bit right.

A lot going on in just inventory levels and management and supply chain optimization, et cetera, as well as pricing. So it's really not one answer, but a lot of moving parts there. I would say the biggest headwind that these moving parts are meant to more than offset is going to be just shift in business. Our margin profile for many of our Enterprise products are less than our Service Provider products, as an example, and we do expect Enterprise to outpace Service Providers. So we do need to improve mix software predominantly as we move forward.

Ladies and gentlemen, this does conclude our question-and-answer session as well as today's conference call. You may now disconnect your lines. Have a wonderful day.