Good afternoon. Welcome to the F5 Networks third quarter fiscal 2019 financial results conference call. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there'll be a question and answer session. If you would like to ask a question during this time, simply press star, then the number 1 on your telephone keypad. If you would like to withdraw your question, press the pound key. Today's conference is being recorded. If anyone has any objections, please disconnect at this time. I would now like to turn the call over to Ms. Suzanne DuLong. Ma'am, you may begin.
Hello and welcome. I'm Suzanne DuLong, F5's Vice President of Investor Relations. François Locoh-Donou, F5's President and CEO, and Frank Pelzer, F5's Executive Vice President and CFO, will be making prepared remarks on today's call. Other members of the F5 executive team are also on hand to answer questions during the Q&A portion of the call. A copy of today's press release is available on our website at f5.com, where an archived version of today's call also will be available through October 24th, 2019. The replay of today's discussion will be available through midnight Pacific Time tomorrow, July 25th, by dialing 800-585-8367 or 416-621-4642. For additional information or follow-up questions, please reach out to me directly at s.dulong@f5.com. Our discussion today will contain forward-looking statements, which includes words such as believe, anticipate, expect, and target.
These forward-looking statements involve uncertainties and risks that may cause our actual results to differ materially from those expressed or implied by these statements. Factors that may affect our results are summarized in the press release announcing our financial results and described in detail in our SEC filings. Please note that F5 has no duty to update any information presented on this call. With that, I'll turn the call over to François.
Thank you, Suzanne. Good afternoon, everyone. Thank you for joining us today. I'll talk briefly to our business drivers before handing over to Frank to review the quarter's results in detail. We continue to aggressively execute our strategy of transitioning F5 to a software-driven model. From our efforts to reprioritize our development resources to the introduction of new flexible consumption models and, most recently, the acquisition and integration of NGINX, customers are seeing a new F5. We believe the steps we have taken, including very deliberate go-to-market changes, are accelerating our transition to a software-driven business while driving overall product revenue growth. As a result, it is no longer accurate to view F5 through the narrow lens of a traditional ADC player. In Q3, we delivered 4% total revenue growth and 3% product revenue growth with 79% organic software growth.
If we include the partial quarter of NGINX, our software growth was 91%. Our exceptional software growth in the quarter is being driven by security use cases, including web application firewall and bot defense and mitigation. The quarter also included a few multimillion-dollar ELA deals. Software growth was partially offset by our systems business, which was down 11%. Our services business delivered 4% revenue growth in the quarter and continues to produce robust gross margins with consistently strong attach rates. I will speak more to our business dynamics and the NGINX integration later in my remarks. Overall, the team is executing very well against our long-term strategy. We believe with the combination of NGINX, we are exceedingly well-positioned to capitalize on continued application growth in a rapidly evolving application services landscape. Frank?
Thank you, François. Good afternoon, everyone. As François noted, we delivered another quarter of strong revenue and EPS growth. As you know, we closed the acquisition of NGINX on May 8, and our Q3 results include NGINX from that point through quarter end. As a reminder, our guidance for Q3 revenue and earnings was set prior to the close of the NGINX acquisition and accordingly did not include any impact related to NGINX's results. For this quarter only, during my remarks, I will break out NGINX's contribution to several Q3 performance metrics. Going forward, NGINX's contribution will be fully integrated into our results and guidance. Third quarter revenue of $563 million was up approximately 4% year-over-year. NGINX contributed $5.1 million in the period. Excluding NGINX's contribution, we delivered revenue of $558 million, near the top end of our guided range of $550 million-$560 million.
GAAP EPS was $1.43 per share, and non-GAAP EPS was $2.52 per share. Excluding the impact of NGINX, non-GAAP EPS would have been $2.57 per share at the top end of our guidance range. Q3 product revenue of $249 million was up 4% year-over-year and accounted for approximately 44% of total revenue. NGINX contributed $4 million to product revenue in the quarter, all of which was subscription software revenue. As François mentioned, software grew 91% year-over-year and represented approximately 27% of product revenue, up from Q2, when it was approximately 19% of product revenue. Excluding NGINX, software revenue grew 79% year-over-year and represented 26% of product revenue. In the quarter, we had very strong uptake on our software solutions sold as ELAs and annual subscriptions.
Under the modified retrospective approach to ASC 606, we are required to compare our results under 606 to what they would have been under 605 during the first year of adoption. In the quarter, the implementation of 606 resulted in $29 million more in recognized revenue compared to what it would have been under 605, excluding any impact of NGINX. I would like to remind everyone of one point about our business. In the year-ago quarter, almost all of our revenue was being driven by perpetual licenses or other consumption models that are not impacted by the adoption of 606. Had ASC 606 been applied to our Q3 2018 quarter, you would have seen a de minimis difference in revenue. Therefore, the adoption of 606 had little impact on our revenue growth in the quarter.
Systems revenue of $181 million made up approximately 73% of product revenue and was down 11% year-over-year. Services revenue of $314 million grew 4% year-over-year and represented approximately 56% of total revenue. NGINX contributed $1.1 million in services revenue in the quarter. On a regional basis in Q3, Americas revenue declined 1% year-over-year and represented 53% of total revenue. EMEA was up 2% year-over-year and accounted for 24% of overall revenue. APAC was very strong in the quarter with revenue growth of 22% year-over-year, representing 23% of total revenue. Looking at our bookings by vertical, enterprise customers represented 60% of product bookings and service providers accounted for 20%. Our government business was very strong, representing 19% of product bookings, including 8% from U.S. federal.
In Q3, we had three greater than 10% distributors, Ingram Micro, which accounted for 18% of total revenue, Tech Data, which accounted for 11%, and Westcon, which accounted for 10%. Let's now turn to operating results. GAAP gross margin in Q3 was 83.9%. Non-GAAP gross margin was 85.4%, in line with our expectations. GAAP operating expenses were $370 million. Non-GAAP operating expenses were $295 million. Non-GAAP operating expenses excluding NGINX were on the higher end of our expectations as a result of higher sales commissions related to software sales. Our GAAP operating margin in Q3 was 18.2%, and our non-GAAP operating margin was 33.1%. Excluding NGINX, non-GAAP operating margin was 34.2%, which was in line with our expectations. Our GAAP effective tax rate for the quarter was 20.1%. Our non-GAAP effective tax rate was 20.7%. Turning to the balance sheet.
In Q3, we generated $150 million in cash flow from operations, down from recent levels, mainly due to the strength in subscription sales, including ELAs, where revenue precedes the collection of cash. Cash and investments totaled $1.15 billion at quarter end. DSOs at the end of the quarter was 51 days. Capital expenditures for the quarter were $33 million, up sequentially as we approach the final phases of building out our new facility in downtown Seattle and our new facility in Hyderabad. Deferred revenue increased 14% year-over-year to $1.17 billion. A little less than half of the increase over the prior year quarter relates to the adoption of 606. We ended the quarter with approximately 5,195 employees, up 400 people from Q2, reflecting the addition of NGINX and our continued hiring in growth areas, including sales and research and development.
In Q3, we did not repurchase any of our common stock, opting instead to rebuild our cash balance following the NGINX acquisition. We continue to view cash as a strategic asset for our future growth. Our primary focus with cash generation is augmenting our strong balance sheet, we may opt to repurchase shares during any open trading window. Let me share our guidance for fiscal Q4 of 2019. Unless otherwise stated, please note that my guidance comments reference non-GAAP operating metrics and include our NGINX business. We continue to make strong progress transitioning our business to a software-driven model. We remain confident in our position in the market and expect our growth will be driven by the growth of applications and increasing demand for our multi-cloud application services. We also expect continued strong demand for our software solutions, including subscription and ELA offerings.
With this in mind, we are targeting Q4 2019 revenue in the range of $577 million-$587 million. We note for Q4, we expect NGINX to contribute less than $8 million in revenue, which includes the impact of purchase accounting write-down of deferred revenue. We do not expect to provide NGINX specific guidance after this quarter. We expect gross margins of approximately 85.5%-86%. We estimate operating expenses of $301 million-$313 million. We anticipate our effective tax rate for Q4 will remain in the 21%-22% range as previously provided for the full fiscal year. Our Q4 earnings target is $2.53-$2.56 per share. In the quarter, we expect share-based compensation expense of approximately $43 million and $4.6 million in amortization of purchased intangible assets. We expect Q4 CapEx of $25 million-$35 million as we complete the build-out of our new corporate headquarters.
With that, I will turn the call back over to François. François?
Thank you, Frank. I'll spend just a few minutes on the trends we're seeing in the business as well as the NGINX integration before moving to Q&A. F5 today is very different than the F5 of even 24 months ago. A year ago in March, we laid out our strategy for transitioning F5 to a software-driven model. Since then, the team has been executing on that strategy and making deliberate changes that have substantially reshaped F5 and our growth opportunities. What do I mean by that? In the last 24 months, we have, number one, executed a wholesale reprioritization of our resources, aligning the business with our growth priorities. Number two, introduced new consumption models like subscriptions and ELAs that make it easier for customers to purchase and deploy F5 application services as software. Number three, we have doubled down on security, including investments in WAF and bot mitigation.
Number four, we have also brought new solutions to market, solutions like Cloud Edition and Central Management, orchestration, and APIs, all of which have opened new opportunities for us. As a result, it is no longer accurate to view F5 through the narrow lens of a traditional ADC player. Our efforts to expand our reach and broaden our role have accelerated our software transition, and it is becoming evident in our performance. This is true even before we factor in future software growth catalysts, including NGINX and F5 Cloud Services. Today, F5 is a leader in an emerging and rapidly expanding multi-cloud application services space, a space that has arguably been underserved, which is why we are generating the kind of software growth we are. The multi-cloud application services opportunity differs from the traditional ADC opportunity in three fundamental ways.
First, multi-cloud application services offer customers a much broader range of application services beyond load balancing and traffic management to include security, analytics, API management, application performance management, and service mesh. Second, multi-cloud application services reach beyond the data center to public cloud into containers. They are more versatile and agile, and as a result, they can support a much larger universe of applications. Third, multi-cloud application services are easier for customers to deploy, consume, and manage. It is clear that customers view F5 as a multi-cloud application services player. For instance, we continue to see security use cases driving software growth and accounting for a higher share of our overall product business. In particular, we continue to drive strong traction with our web application firewall, anti-bot, and machine-generated traffic monitoring and blocking capabilities.
During Q3, for instance, we secured our largest global web application firewall deployment yet with an international financial institution. This customer had been using multiple WAF platforms to protect its applications and selected F5 as their enterprise-wide WAF solution after thorough evaluation of a number of competitors. Customers are also increasingly choosing F5 for our advanced capabilities in automation, orchestration, and central management. We help them simplify an increasingly complex combination of environments and sprawling deployments. During Q3, we secured a win with a government customer that selected BIG-IQ to manage a large and growing number of BIG-IP deployments. The customer also expects to deploy our Application Security Manager to manage web application firewall policies. We also continue to see customers who are migrating to cloud environments demanding the same level of application security and agility as they have in their data centers.
This was the case with a U.S. enterprise customer that selected our Cloud Edition during the quarter. The customer expects to transition individual applications to cloud environments over time, which made cloud editions per app consumption and deployment model an ideal solution. Finally, our new subscription consumption models continue to facilitate software growth across our customer base, with both enterprise and service provider customers leveraging friction-free F5 services procurement and deployment. As an example, one of the ELAs closed in the quarter was with a large next generation mobile carrier in APAC. F5's 5G ready NFV solutions will address the needs of this customer's growing 4G mobile broadband consumer services. We will enable connectivity and security for voice over LTE IMS services and enterprise IoT services. Our solution reduces both capital and operating expenses while increasing service agility with faster build, test, and deployment cycles.
We also simplified the network with software-based network functions and the ability to dynamically manage and orchestrate services. This enables the customer to tailor innovative services to subscriber preference and usage. A word on our systems business and how it fits into our multi-cloud application services opportunity. We believe our systems business is seeing the effects of two factors. The first relates to the actions that we have taken to make it easier for our customers to consume F5 as software, reducing friction with new consumption models, ensuring easier provisioning of software, simpler licensing management and models, and generally reducing operating complexity. The second is that our customers are better able to operationalize and manage a virtualized infrastructure environment, and as a result, a number of them are implementing software first policies.
As a result of these two factors, we are seeing an accelerated shift in our product mix towards software. Before we move to Q&A, let me talk to our combination with NGINX. With the NGINX acquisition completed on May 8th, the teams have come together well, and we are executing our integration and value creation plan at a rapid pace. I will highlight progress on two key work streams in particular for this audience. One, go-to-market, and two, product integration. First on go-to-market. NGINX's sales team has worked hard to maintain the momentum of NGINX current offerings and since close, has been operating as an overlay to the F5 sales team, all of whom have been trained on NGINX. Our ability to bridge the gap between NetOps and DevOps is resonating with customers, and we are already seeing the power of our combined sales efforts.
We estimate that the joint F5 NGINX and F5 initiated sales efforts have increased NGINX's net new Q4 pipeline by roughly 20%. As an example, during Q3, we secured a joint F5 and NGINX win with an APAC-based international telecommunications provider. The customer needed a security solution for its private cloud, distributed over 10 major city points of presence. They selected F5 software-based WAF to protect traffic ingress points. They selected NGINX for microservice security, including protection for API gateways, cloud governance, and reverse proxies. The combined F5 NGINX solution resolved scaling issues with central controls and provided a consistent approach to security postures across the business while underpinning faster application delivery. Briefly on F5 NGINX product integration. As planned, the F5 cloud native product development team has been combined with the NGINX team, reporting to Gus Robertson.
The teams have come together well and have made very strong progress on engineering and product integration. As a first priority, the teams are moving quickly to converge NGINX's controller and F5's cloud native application services platform. In fact, we expect the first release of a converged F5 NGINX offering within the next six months. We expect the converged F5 NGINX controller will be an accelerator for our NGINX business, expanding both the addressable market and potential deal size by spanning a broader set of use cases across DevOps and super NetOps customer personas. Overall, we are more enthusiastic than ever about the opportunities we are pursuing as a combined F5 NGINX team. Near term, we are addressing a critical challenge for customers by bridging the NetOps DevOps divide.
The combined F5 NGINX provides the management and ease of use features that traditional infrastructure buyers, including network buyers expect, enabling F5 to cover the full spectrum of application and modernization needs. Going forward, we believe applications will be increasingly disaggregated into smaller components, containerized, and distributed across multi-cloud environment. NGINX is true software ideal for these container-based DevOps environments, which means unparalleled agility and lower cost for our customers. We are confident that F5 and NGINX, with our combined solutions and application expertise, bring significant advantages to these cloud-native and containerized environments. In summary, we are very pleased with the progress we are making overall, and we are confident we can continue to drive software momentum.
One of the things we find most encouraging is that the software growth we have delivered thus far is largely from solutions that have been in the market for over a year, including our BIG-IP Virtual and Cloud Edition. As we look ahead, we see even more growth coming from new offerings, including our recently introduced SaaS platform, F5 Cloud Services, and our combined F5 NGINX offerings. In closing today, my thanks to our partners, our customers, and our shareholders. My thanks, too, to the entire F5 team, and especially to those working to ensure the combined F5 NGINX is as successful as we believe it can be. With that, operator, we will now open the call to Q&A.
Thank you. At this time, I would like to remind everyone, in order to ask a question, please press star and the number 1 on your telephone keypad. We'll pause for just a moment to compile the Q&A roster. Your first question comes from the line of Alex Kurtz from KeyBanc Capital Markets. Your line is open.
Yeah. Thanks, guys, for taking the question. Just on the impact from software, which is really great to see, François, in the quarter. Any dynamics around how customers are consuming it? These comments are excluding NGINX for a second, but just what it means for deal size, what does it mean for how the revenue's being recognized in the quarter, and how we should think about that relative to systems business?
Hey, Alex. I'll take the first part, and perhaps Frank can comment on the revenue recognition portion. The way that our customers are consuming our software is both via our perpetual licenses, and also via ELAs, which are typically three-year subscription agreements, and also one-year subscriptions. The part of the business in software that has the fastest growth is, in fact, the subscription portion of the business, because it gives our customers a lot more flexibility in terms of their own cost models, and also the ability to consume the features as they go. We're seeing a lot of traction with these consumption models, and we're seeing the range of deals, to your point around deal size, ranges from small one-year subscription that are typically multiple tens of K's, all the way to multimillion-dollar three-year enterprise license agreements.
Alex, with the adoption of 606 this year, the revenue recognition associated with a lot of subscription deals look exactly like what we would have had as perpetual deal in previous years. Instead of ratably recognizing, you split out the value components and recognize that proportion of the revenue, which is frankly modeled after our perpetual. On a year-over-year comparison, it's very close to the same.
Okay. Thank you.
Your next-
Thanks, Alex.
Your next question comes from the line of Paul Silverstein from Cowen. Your line is open.
I appreciate it. A couple, if I may. First off, my math must be wrong, but on the software revenue, I'm coming out to $67 million in the current fiscal Q3, $65 million if we exclude NGINX based on what you said, versus $45 million a year ago. That will work out to 42% year-over-year growth, not the 79% organic and 91% in total. Can you just help me out on the numbers? Did I mishear y'all that it was 19% of product revenue a year ago and 27% in the current quarter, 26% ex NGINX?
Yeah. I think your number, Paul, for this year is correct, though the number for last year is not. It's much lower. I think it's something close to $34 million or $35 million.
Did y'all not do $239 million of product revenue last year in the year ago quarter?
Yeah, about 239.
19% of 239 is $45 million.
No, 19% that I referenced.
Unless my 19% is off.
19% that I referenced was last quarter, not the year ago quarter.
My apologies. What was it in the year ago quarter?
I'm doing this off memory, Bud, but I think it was about 14%.
All right. I'll come back to you after the call on that.
Sure.
Related to software, NGINX in particular. I think there's a view out there among most investors that NGINX displaces existing F5 platforms, and you all bought Silverline back when, previous acquisition. You're working hard on it, if I recall, for the past two plus years. Can you just clarify whether NGINX totally displaces what you've done with respect to cloud platforms or whether it augments? I've got a quick question on VMware Avi.
Paul, thanks for the question. On NGINX, no, it does not displace either our perpetual software sort of virtual edition offerings or Silverline. Silverline is essentially a managed security services platform for customers that want to consume WAF, and DDoS, and other security services, with associated 24 by seven support and leveraging our infrastructure to scrub the traffic and provide managed services to them. NGINX is a great platform for DevOps environment, essentially injecting application services in the code of application logic, and it plays to a very different market than the market we have played in to date, either with Silverline or with our virtual editions.
All right. Frank, before I ask the Avi question, what % of your total revenue is now recurring?
I would say it's north of 70%, including the services piece.
Do you know what it was a year ago?
I'd have to get back to you on that one, Paul. I don't have that off the top of my head.
If you could, that'd be appreciated. For the Avi question. Avi was talking a really good game before it got acquired, and it just got acquired by VMware. Obviously, that lends resources, incumbency, customer base, all things are needless to say. I know it hasn't been a long time, what are you seeing in terms of their success or lack thereof currently versus whatever degree of success they were having previously in competition against you?
Hi, Paul. I know this one has gathered quite a bit of interest after the VMware acquisition, I would take the time to answer your question thoroughly and put a few facts on the table. Avi is essentially a software load balancer with the value proposition around nice analytics and a nice GUI. Where that value prop plays is in use cases where a customer would value the GUI and analytics over having a truly lightweight, truly scalable solution. I'll tell you that this is actually a very small fraction of the market, let me be specific about that. Historically, we have seen them in less than 2% of our deals. Even when we've seen them, we win the large majority of these deals.
Even in the accounts that they had claimed to be lighthouse accounts for them, we've looked at them, and F5 still, for the most part, has 95% or more of the wallet share of ADC in these accounts. Then again, going back to some of these lighthouse accounts, we've seen a number of customers that are coming back to F5 from Avi because of a lack of scale, lack of features, or difficulty in getting this stuff operationalized. That's where historically things have been, and I just wanted to make sure the facts were clear around that. Now, VMware is a great company. We have a lot of respect for them. We do partner with them. Of course, they're going to provide larger distribution for Avi.
Where I think this is going to play out is that the VMware Avi proposition, I think will appeal to customers who have gone all in on NSX and value deep integration between NSX and a software load balancer more than they value a true multi-cloud architecture and a broad range of application services. Again, I think that's going to be a fairly limited portion of the market. Let me get to how we're going to compete in this space. The first piece around this, the GUI and analytics, frankly, we're neutralizing that with our controller, which, as you heard in my prepared remarks, is going to be released within the next six months.
This is the combined NGINX and F5 controller, which is a combination of the NGINX controller that's in the market and the work that F5 had done on our cloud application services platform. There's a lot of excitement about this combined platform. Beyond just that, when we looked at the market and ended up buying NGINX, we've always said this was an offensive play on the architectures of the future. What we did is we looked at all the players in the virtual ADC space, and we moved very quickly on what we believe was the most attractive assets in that space. We chose NGINX because number 1, they have the smallest footprint, which enables them to be inserted in microservices environments. They're a natural fit for these cloud-native microservices environments. Number 2, they have proven scalability.
They are the platform that everybody builds on, and the largest public clouds around the world are built on NGINX. Number three, they are a true platform that enables a broad range of application services, including things like API gateway, app security, app server. They're not a virtual appliance with a set of constraints and limitations. For all these reasons, NGINX is a natural choice for DevOps and application development teams. That's why we chose NGINX, and we're very excited with what the combination of F5 and NGINX is going to do in the market.
I appreciate the response. Thank you.
Your next question comes from the line of Sami Badri from Credit Suisse. Your line is open.
Thank you. I just had a quick question first on some of your cash, and where your cash balance is going to be by about mid-year 2020, or at least the next fiscal year. Basically, two to three quarters from where we are today. I just wanted to understand maybe what the corporate plan was, what you're going to do with this cash balance. Are you going to consider more M&A? Are you going to potentially reinstate your cash buyback? Would you consider some other factors? Just to give us a bit of an idea on what you're going to do with this quickly escalating cash balance.
Hi, Sami. Look, I think you can look at it this way. For now, we're placing a significant emphasis and focus on augmenting our balance sheet and rebuilding our cash balance. We're going to continue to look at opportunities in the marketplace, whether it's potential M&A or, if the opportunity presents itself, potentially more buybacks. All of these options are on the table, but in the very short term, there is more emphasis toward rebuilding our cash balance.
Got it. Thank you. My next question has a lot to do with, given the formation of some of the new offerings F5 and NGINX are going to come together and go to market with, would it be safe to assume that systems revenues specifically, or at least the growth rate, or I guess you'd say the decline, could accelerate in probably six months out from today, given that the fair majority of the new offerings coming to market are more software-based versus hardware? This is also probably a bit more correlated with the IT backdrop that you're trying to see some metrics start to decelerate in a broader macro IT world. Could you potentially see further deceleration of growth in the system side, and then naturally see some acceleration further in the software side?
I don't necessarily think that the factors you mentioned, Sami, would lead to further deceleration on the hardware side. There are also other sort of compensating factors. When you look in 2020, I think we should probably see a bit more from service provider around sort of capacity upgrades that are more naturally hardware driven. We also have some evolution of our own platforms, hardware platforms, that would contribute to that. I think, there are factors on both sides of that equation. We've taken all of that into account when we have given our guidance for Horizon 1 around mid-single digit growth. I think we're still comfortable with that guidance. In that guidance, Sami, our thought was that hardware would decline at sort of low single digits or flat, and software would grow in the 35%-40% range.
I'd say given what we've seen over the last six months, I think it's fair to say that perhaps software is going to grow a little faster and hardware is going to decline also a little faster than what we thought. At this point, it's hard to give you a hard answer on the exact numbers for each. Overall, we still think the balance of revenue growth is going to be what we thought.
Got it. Thank you. My last question has to do with the APAC deal regarding the carrier. Just to give us an idea, could you give us an idea of mix on software versus hardware in terms of what this carrier bought from you guys, just so we have maybe some kind of illustration on what future carrier deals could look like?
Sami, it's a good example of a carrier that is moving to NFV in anticipation of the 5G rollout. In this case, they bought software, essentially. It was 100% software ELA with a number of virtual functions included in the deal.
Your next question comes from the line of Rod Hall from Goldman Sachs. Your line is open.
Hi, guys. Thanks for the question. I guess I wanted to go back to this question of systems and systems decline and ask you what you think your market share position looks like for the quarter. Do you think you maintained share? Do you think you lost share? What would you expect to happen in the future, like the next few quarters? Do you think that within the systems business or for ADCs, you'll be able to maintain share and just kind of track market performance, or would you expect to kind of outperform the market still?
Well, Rod, first of all, I think looking at just the ADC market is a fairly narrow lens to look at when it comes to F5 now. As I said in my prepared remark, I think we're really becoming more of a multi-cloud application services player, and I think that's a broader opportunity. That being said, if you're going to look at the lens of the ADC market, overall, hardware and software, I definitely think we're gaining share. Specifically in the hardware space, have we gained share this quarter? I can't answer that, and we won't know until market share reports come out in a couple of quarters. What I can tell you is there are things that we have done that are deliberate actions from F5 that are transitioning some of our hardware business to software.
These include things we've done in our software offers to make them much easier to consume and remove all of the friction that exists in our customers adopting these solutions. It also revolves around the commercial offerings and the way we've structured and changed our contracting structure. It also revolves around the things we've done on go to market and the incentives we've put in our field teams to proactively help customers who want to transition to software, transition to software faster. Those actions are driven by us. Of course, there are effects that have to do with the market around customers themselves wanting to implement software first policies, and also wanting to have this multi-cloud portability, which is made easier with software.
When you look at these two factors, that's what I think is driving the decline in systems, but some of it is driven by us.
And then maybe-
Comment on this, Rod.
Oh, go ahead, François.
If that's kind of where what you're getting to, we have not been losing hardware deals, if this is your question. We actually continue to be very happy with the win rate on our hardware deals.
Yeah, no. The angle of my question was more you guys had been gaining share in hardware, and I wondered if you thought you would sustain that or whether you thought maybe you would match the hardware market more, but that was kind of where I was coming from. I also wanted to ask, François, given the high software growth in the quarter, can you give us any idea what you think software growth, either for the market or even better yet, you, looks like over the next year? Are you thinking that you can maintain this kind of ballpark in terms of growth, or is this a one-off, or what sort of growth do you think is likely as we look out?
I think, Rod, if you look at the analysts' reports on ADC software growth, by the way, my comments here exclude ADC as a service, which is largely the native cloud offerings. I'll come back to that in a moment. If you exclude them, the analysts' reports so far have pointed to, I'm going to say, 20%-ish growth rate in the ADC software market. If you look at that, we definitely believe that we are gaining a significant share in the ADC software market, even though our software growth isn't just driven by ADC. It also has a strong component of security, which has very strong software growth. That's kind of when you look at things today. To your question around, can we maintain the software growth in the coming quarters? Look, I don't expect 90% software growth every quarter.
As I said before, my expectation is that we are likely to do better than the 35%-40% software growth that we've put in our Horizon 1 guidance, in part because we have growth catalysts that haven't played out yet, specifically NGINX, and also F5 Cloud Services, which is more of a sort of back half of 2020 contributor and certainly a contributor in Horizon 2. I think those are growth catalysts that are yet to materialize for us. Overall, when you combine that and what's happening with hardware, I still think what we've shared with you for Horizon 1, for overall aggregate revenue growth for Horizon 1, is correct.
Okay, great. Thank you, François. Appreciate it.
Yep. Thank you, Rod.
Your next question comes from the line of Jeff Kvaal from Nomura. Your line is open.
Thank you. Yes, I have two questions. I think first, François, for you. The tremendous success in software this quarter does put you on a much higher trajectory when it comes to that piece of the business. It doesn't seem as though your overall corporate growth rates have picked up despite the software explosion. I guess I would have expected there to be a little bit of uptick in the overall corporate growth rate if you are expanding into standalone security, et cetera. I was wondering if you could offer some thoughts about that. Frank, on your side, the OpEx number that you are offering us for the September quarter is a little higher than we had modeled, and there's an integration of an acquisition, so I get that.
I'm just wondering if you could help us understand where the OpEx is and where it might go over the course of the fiscal year, to the extent you can. Thank you.
Jeff, thank you. I'll take the first part. Look, I think if you look at our overall revenue growth, this is playing out so far pretty much in line with what we have shared at AIM and what we have shared when we made the NGINX acquisition in terms of overall aggregate revenue growth. I think we shared that in our Horizon 2, we expected this to pick up. That's still our view. For Horizon 1, I think when you look at the dynamics combined of the transition we have to go through as a company, starting from a large base of hardware and transitioning to a majority software business, that's how it's playing out, and I think the growth rate reflects that for the time being.
Jeff, in terms of non-GAAP operating margin for the rest of our fiscal year, the coming quarter that we have is the rest of our fiscal year. We're not providing any guidance at this point for FY20. As we talked about in the acquisition of NGINX, we expected that our non-GAAP operating margin in Horizon 1 was going to be in the range of 33%-35%. This is absolutely in that range. We did that with the anticipation of the expenses that were going to come on through the NGINX acquisition. That's exactly what we're seeing.
Okay. All right. Thank you both.
Your next question comes from the line of Samik Chatterjee from J.P. Morgan. Your line is open.
Hi. Thanks for taking the question. François, I just wanted to start off with a more broader question on what are you seeing in terms of spending trends from enterprise customers? I think last quarter particularly, there were a few companies that had mentioned that given the uncertain macro, there was some kind of softness in signing large deals. What are you seeing on the ground? Are you seeing kind of any hesitation in signing those large deals at this point?
Hi, Samik. I would maybe give you a couple of pointers. Overall, I think the macro environment is less healthy than it was a year ago. I wouldn't characterize it as being kind of difficult, but it is not as healthy as it was a year ago. Where we're seeing specific areas of softness, I would say the U.K. and Germany for us in Europe, and specifically the U.K., where uncertainty persists, and we have seen a number of deals there being delayed or canceled altogether. I would also say specifically in hardware, we continue to see more scrutiny applied to hardware deals, of course, especially in companies that have software-first policies. That probably elongates the deal cycle specifically for hardware deals.
Got it. If I can just quickly follow up with a question on the software side. You had very strong sequential growth in the software revenues, and you mentioned a lot of that was driven by the security products. You also launched new SaaS offerings, I think, I believe in March. Can you just kind of help us understand if the sequential growth, how much contribution was there from the new SaaS offerings, or if it's immaterial now, how much time do you think it takes to ramp those up so that they become material to your kind of outlook?
Yeah, Samik. Yes, you're right. We launched F5 Cloud Services, our SaaS platform, in March, and introduced our first service on the platform, which is DNS as a service. We're now in the process of launching our next set of service. Global Server Load Balancing actually is next and is going to go GA pretty soon, followed by WAF as a service, et cetera. Where we are is we're very early days, and we basically just have the first service, but even on the first service, we actually already have the first paying customers on the platform. And we have actually a large number of customers in trial, and we expect that the pickup will accelerate as we introduce the new service like GSLB and WAF. This obviously did not contribute to the quarter.
We don't expect it to be really material, certainly not in the next three quarters. SaaS businesses take a bit of time to ramp. Certainly going into FY21 and '22, when we get into our Horizon 2, we would expect our F5 Cloud Services platform to be a meaningful contributor to our overall business and certainly to our software growth.
Got it. Thank you.
Your next question comes from the line of James Faucette from Morgan Stanley. Your line is open.
Great. This is Meta Marshall for James. Maybe a first question. Understanding that the 33%-35% OpEx is kind of within the Horizon 1 that you had noted. Just also, you kind of noted a change to the go-to market. I just want to get a sense of how much of this is just layering in NGINX versus how much of this is there a cost to kind of the change in go-to-market approach? Then maybe the second question, just if we could get a sense of what is the impact of the purchase accounting on NGINX for kind of your forecast of you mentioned it's an $8 million contribution in the quarter. What would that have been without kind of the purchase accounting adjustment? Thanks.
Hi, Meta. I'll take the first part and Frank will take the second part. On the 33%-35% operating profit guidance for Horizon 1, Meta, this is mainly due to investments in our software platforms, largely NGINX, and the investments we have to do there. Some of F5 Cloud Services, but it's mainly NGINX. It's not really about the change in the go-to-market models. Even though we have increased the size of our go-to-market teams with the addition of NGINX, this is more about the platforms than a fundamental change in our go-to-market.
In relation to the $8 million, that was a discussion about the contribution to NGINX next quarter without a purchase accounting. My assumption is that would be just approximately $10 million, maybe slightly more than $10 million next quarter.
Got it. Great. Thanks, guys.
Your next question comes from the line of Simon Leopold from Raymond James. Your line is open.
Great. Thanks for taking the question. I wanted to maybe come at the OpEx question a little bit differently. I want to reflect on the June quarter. Excluding the acquisition you had forecast, I believe $275 million to $287 million, which would imply something like $60-ish million coming from the acquisition. If you could maybe unpack what occurred in the June quarter, and then help us understand what's maybe one-time-ish, what's sort of ongoing from the acquisition, because obviously you're forecasting significantly lower operating expenses in September. I just want to make sure I understand the moving parts, the components to operating expenses.
Sure, Simon. All that $60 million, I think where you're getting that figure is when we talk about the not just contribution from NGINX, but also some of our non-GAAP things that we had to split out. That reconciliation can be found in the press release, and so I'd point you to that. Some of the bigger items were there was a portion of the NGINX acquisition where it was part of the consideration, but accounting makes us take that as an expense in the quarter, and it was effectively a holdback or a revest for key employees that was part of the deal negotiation. That was a large portion of that. There were some facility exit charges. The rest of what we would discuss would be the inflow of the new NGINX employees.
All of that was anticipated as part of our guidance and our guidance going forward.
The incremental expenses from NGINX by itself, in terms of what it's doing to operating expenses, is in kind of a $25 million-$30 million range per quarter?
I think it's actually, Simon, a little bit less than that, but it's not a number that we're splitting out for all practical purposes. We've already combined some of the teams together, and it's not the way we're tracking the business.
Okay. Maybe just to pivot, in terms of the cash position, I understand this is, I think the first quarter in more than 10 years that F5 has not bought back stock, so sort of a significant break in the pattern. What is the cash level that you feel is necessary to run the business? Because it seems as if you've certainly got adequate cash to do everything you'd like and could still buy back stock. So I guess I'm trying to really, even if you're not ready to give us a number, if you can maybe explain your philosophy.
Sure, Simon. I mean, for a long time we've had plenty of cash to run the business and have chosen to do share repurchase as opposed to other things that you can do with cash, including inorganic expansion. We are adding back up that flexibility to continue to use cash for multiple purposes, and we will be strategic with it. I can't say that there is ultimately one level of where the cash balance has to be before we go back in the market and repurchase shares. That could be this quarter.
Could you maybe just follow that with your thought on instituting a dividend?
Yeah, we don't have any anticipation at this point of instituting a dividend. We think that that's actually not a great tax-efficient way of redistributing cash to our shareholder base. We'd be much more opt to do share repurchases as that vehicle to redistribute cash.
Great. Thanks for taking the questions.
Thank you, ladies and gentlemen. This concludes today's conference call. You may now disconnect.