Ladies and gentlemen, thank you for standing by, and welcome to NetScout's third quarter fiscal year 2018 results conference call. At this time, all parties are in a listen-only mode until the question and answer portion of the call. As a reminder, this call is being recorded. Andrew Kramer, Vice President, Investor Relations, and his colleagues at NetScout are on the line with us today. If you require operator assistance at any time, please press star zero. I would now like to turn the call over to Andrew Kramer to begin the company's prepared remarks.
Thank you, Keith, and good morning, everyone. Welcome to NetScout's third quarter fiscal year 2018 conference call for the period ended December 31, 2017. Joining me today are Anil Singhal, NetScout's Co-founder, President, and CEO; Michael Szabados, NetScout's Chief Operating Officer; and Jean Bua, NetScout's Executive Vice President and Chief Financial Officer. There is a slide presentation that accompanies our prepared remarks. Both the slides and the prepared remarks can be accessed on the investor relations section of our website at www.netscout.com. The slides can be advanced in the webcast viewer to follow our commentary. We will call out the slide number we are referencing in our remarks. Our agenda today is as follows. Anil Singhal, our President and CEO, will briefly review our performance and then address certain questions that have arisen since we announced our preliminary third-quarter results and revised outlook for fiscal year 2018.
Michael Szabados will review customer adoption trends and major go-to-market activities. Our CFO, Jean Bua, will then review our third-quarter results and detail our updated fiscal year 2018 guidance. Moving on to slide number three, I would like to remind everybody listening that forward-looking statements as part of this communication are made pursuant to the Safe Harbor provisions of Section 21E of the Securities Exchange Act of 1934, as amended, and to other federal securities laws.
Investors are cautioned that statements on this conference call which are not strictly historical statements, including but not limited to the statements related to the fiscal year 2018 financial guidance for NetScout, revenue and profit growth prospects for fiscal year 2019, share repurchase activities, including the ASR, market conditions and customer demand, anticipated revenue from specific customers and specific products, along with all of the other various product developments, sales and marketing expense management, and other initiatives planned for the remainder of this year or into fiscal year 2019, constitute forward-looking statements, which involve risks and uncertainties. Actual results could differ materially from the forward-looking statements due to known and unknown risks, uncertainties, assumptions, and other factors. This slide details these factors, and I strongly encourage you to review each of them.
For a more detailed description of the company's risk factors, please refer to the company's annual report on Form 10-K for the fiscal year ended March 31, 2017, and subsequent quarterly reports on Form 10-Q, which are on file with the Securities and Exchange Commission. NetScout assumes no obligation to update any forward-looking information contained in this communication or with respect to the announcements described herein. Let's turn to slide number four, which involves non-GAAP metrics. While this slide presentation includes both GAAP and non-GAAP results, unless otherwise stated, financial information discussed on today's conference call will be on a non-GAAP basis only. This slide, which we also encourage you to read, provides information about the use of GAAP and non-GAAP measures because non-GAAP measures are not intended to be superior to or a substitute for the equivalent GAAP metric.
Non-GAAP items are described and reconciled to GAAP results in today's press release. Those and other reconciliations and supplemental detail are included at the end of the slide presentation, which again, is available on our website. As we disclosed earlier this month, our third-quarter results were notably below our expectations entering the quarter. We expect further challenges to negatively impact our performance in the fourth quarter. As a result, we've updated our full-year revenue and EPS targets accordingly. I'll now turn the call over to Anil for his perspective on these and other matters. Anil?
Thank you, Andy. Good morning, everyone. Thank you for joining us. Let's begin on slide six with a brief recap of our non-GAAP results. Consistent with our January 10th announcement, we reported third-quarter revenue of $272 million, which was down nearly 13% from last year's third quarter and below our plans entering the quarter by $30 million-$50 million. I'll cover the shortfall against expectations in a moment. We also took steps to recalibrate our cost structure through certain one-time adjustments to variable incentive compensation. Our third-quarter diluted EPS of $0.69 per share also benefited from a lower tax rate. We believe that many of the issues that affected third-quarter revenues are likely to impact our performance in the fourth quarter. Accordingly, as we announced on January 10th, we have lowered our full-year fiscal year 2018 revenue and EPS outlook.
We have spent considerable time over the past several weeks trying to help shareholders understand the issues that are impacting our performance. Rather than follow our conventional format of highlighting key accomplishments in the quarter, I would like to focus my commentary on answering the most common inquiries we have received. The first question asks about the biggest factors impacting NetScout's Q3 fiscal year 2018 and full-year fiscal year 2018 revenue. To answer this, it's important to provide some perspective. As we discussed previously, we entered fiscal year 2018 anticipating that one of our largest tier 1 customers would further moderate their spend with us by up to $100 million. This customer has adjusted his overlap spending after multiple years of elevated spending to build his 4G and LTE footprint.
In recent quarters, the customer has attempted to absorb excess capacity and otherwise redeploy equipment to mitigate growing OTT traffic volumes and deliver high-quality services in a very price-sensitive, highly competitive marketplace. We had previously planned to offset this decline with a strong second half of the year, aided in large part by solid growth in our enterprise customer segment. Our inability to achieve our target reflects three major factors. Ongoing and significant service provider capital spending pressure, primarily in North America. Lengthening enterprise sales cycles as our customer grapples with major digital transformation initiatives and related changes to their technology architectures, and funding delays for multiple large federal government agency projects. More than half of the total shortfall is attributable to the service provider customer segment.
In the service provider market, about three-quarters of the shortfall is associated with overall lower than expected orders for our service assurance products, primarily from the tier 1 providers in North America. The remainder of the shortfall is associated with the delays and reduced orders for Arbor DDoS offerings, in part because the top volumes have moderated from prior years, and that is enabling those customers to defer spending. Our enterprise customer segment revenue was also notably below our plans, due largely to funding delays for multiple large federal government projects and longer than anticipated sales cycle within our enterprise customer base. I'd like to briefly explain each of these factors. Regarding the funding delays in the federal sector, we noted in October that our second quarter government revenue was lower than expected.
At that time, we were disappointed that a significant pipeline of opportunity, which we estimated to be $50 million across a variety of federal agencies, was unrealized because funding for those projects had yet to be secured due to a variety of reasons, including the reprioritization of funds to aid disaster recovery activities. We moved into the third quarter with limited visibility into which unfunded projects, if any, would move forward during the second half of our fiscal year. Unfortunately, we did not see any upside from this pipeline in the third quarter, and we no longer believe it's realistic to expect any material contribution from this pipeline going forward.
In terms of lengthening enterprise sales cycles, our enterprise customers' digital transformation initiatives and the related changes to the technology infrastructure has impacted the timing and complexion of deals involving both our traditional products and some of the newer solutions we have introduced in recent quarters. Our enterprise revenue has also been affected, albeit to a lesser extent, by softer than expected orders for certain ancillary enterprise offerings. For example, the handheld tool product lines associated with the former Fluke Networks unit is likely to end this fiscal year at less than 5% of total revenue. This is a non-core, low-margin product area that lacks synergy with our enterprise sales teams, since these offerings are sold through third-party distributors. Given this dynamic, we are looking at a range of options to resize our resources in this area, including potentially divesting these assets altogether.
Investors have also inquired whether the lower revenue outlook reflects any notable change in the competitive landscape. The short answer to that is no. In the service provider service assurance product area, we have made good progress with the new software-only version of our InfiniStreamNG real-time information platform. We focused initially on driving deployment with international carriers, while the revenue base presented limited risk and greater upside. We have made good progress thus far, and this platform represented approximately 7% of year-to-date product revenue. Our largest carrier customers in North America have been actively qualifying this new platform, and we anticipate purchasing from them to begin in fiscal year 2019. In terms of network function virtualization-related initiatives, service providers are still moving cautiously in terms of commercial traffic and actual spending.
Nevertheless, we believe that we are well-positioned to help carriers in this area, and one of our European customers recently selected our virtualized service assurance solutions to support a multi-year transformation of their infrastructure from physical to virtual. We expect a public endorsement of our virtualization technology by this customer over the coming weeks. In terms of enterprise network and application assurance, we have seen sales cycles lengthen as customers advance their digital transformation projects and related changes to their IT infrastructure. Whereas prior decisions to deploy our core solution quickly followed an upgrade or expansion of their traditional data center infrastructures, enterprises now have a broader range of infrastructure options that include private and public cloud migration, and that is extending our sales cycles. Fortunately, we have already made the necessary investment to expand our portfolio to support customers regardless of whichever path they take.
In security, Arbor continued to win new DDoS deals with both existing and new service providers. However, spending on Arbor solution by North American service providers has been limited due to the combination of excess capacity and more modest attack volumes. In the enterprise, the tailwinds created by headline-grabbing DDoS attacks in the fall of 2016 have largely dissipated. That has resulted in fewer multimillion-dollar enterprise wins and smaller overall deal sizes versus last year. Our initial foray into the Advanced Threat market has yet to deliver meaningful revenue, but we have received valuable feedback from early adopter customers and prospects. We expect to introduce a new release of the solution later this spring that is aimed at taking further advantage of NetScout's technology and footprint across our installed customer base.
We also took certain one-time actions during the third quarter that removed approximately $25 million of costs, primarily through adjusting variable incentive compensation, and we expect there to be some modest benefit from this in the fourth quarter. However, these adjustments are non-recurring, so those costs need to be factored into the fiscal year 2019. Accordingly, we are looking at a variety of actions aimed at increasing operational efficiencies by further streamlining roles across multiple function areas. When combined with potentially restructuring the former Fluke handheld tools business, we believe that this can further improve our profit profile without compromising our long-term growth prospects. With that said, our R&D and sales and marketing costs also reflect our ongoing commitment to support hundreds of customers around the globe who are using legacy products from the former Tektronix and Fluke businesses.
We plan to continue to prudently managing these resources as we continue efforts to migrate these customers to our next-generation products while those legacy products move closer to the end of their respective life cycles. The last two questions about order delays and our views for fiscal year 2019 can be taken together. To be clear, a majority of our annual revenue shortfall is tied to opportunities that lack sufficient visibility to assume that they will materialize next year. This includes certain prospective service provider projects, the unfunded federal government pipeline, and certain other enterprise opportunities associated with Arbor, DDoS, and Advanced Threat offerings. Although we continue to advance our planning process for next year, it would be premature for us to offer any specific revenue guidance for fiscal year 2019 right now.
Nevertheless, we are making the necessary investment that we believe can support top-line growth in fiscal year 2019. Our optimism for top-line growth is based on a few factors. First, we believe that spending from our largest tier 1 service provider customers has reached a bottom that should be stable to mildly improved next year. Second, most of the product integration challenges are behind us. Third, we expect better traction from our new products, especially those that can be sold into new areas of IT and security. Additionally, we believe that our ability to deliver software-centric solutions will help us further fortify our incumbency with key service providers, move us down market to support a broader range of enterprise customers, and to support further gross margin improvements. The guidance we will provide for fiscal year 2019 during our next conference call in May will reflect these dynamics and assumptions.
We move forward with high conviction that we are well-positioned to capitalize on a range of exciting opportunities to drive future growth. We anticipate that the combination of continued gross margin improvement and efforts to closely monitor and manage our cost structure should produce further operating leverage. Together with a lower tax rate and lower share count, this should translate into very compelling EPS growth next year. This optimism is underpinned by our plans to execute an accelerated share repurchase of up to $300 million in conjunction with our amended and expanded credit facility. That concludes my prepared remarks, and I will turn the call over to Michael at this point.
Thank you, Anil, and good morning, everyone. Slide number nine outlines the areas that I will cover, which I believe help convey the progress we are making with our new product cycle, as well as our efforts to advance our go-to-market initiatives. As we discussed on prior calls, during the past two-plus years, we have focused our considerable software resources and expertise on driving innovation across the product portfolio. We have extended the deployment options of our real-time information platform, the InfiniStreamNG, from a traditional appliance to a software-only for commercial off-the-shelf hardware, and to an array of virtual alternatives. We have discussed on recent calls that the software-only version of ISNG has been well-received by major international carriers, and we expect a relatively sizable deployment of our software-only platform in support of a major 4G and VoLTE rollout later this spring.
As Anil described, we are seeing longer sales cycles as our enterprise customers consider a broader range of options for deploying their existing and new applications. When our enterprise customers elect to upgrade their traditional data centers, they can extend their deployments with our appliance and software alternatives. As they move their workloads to private or public clouds, our new vSCOUT and vSTREAM products seamlessly expand their application monitoring coverage into the resulting hybrid cloud. Our customers consider our ability to provide them with continuity of visibility through disruptive architecture changes and workload migrations to be invaluable.
As a result, the number of evaluations and proof of concepts for these products has grown steadily since their launch last summer. We are engaging with new buying centers within the IT organization, such as with DevOps and cloud architecture teams, in addition to working with our traditional user base in identifying new use cases. The enthusiasm and interest we generated at last quarter's AWS re:Invent show was further validation that these solutions are highly differentiated and relevant to multiple departments within enterprise IT organizations as they plan and implement various phases of their data center transformation projects. We are also expanding our pipeline for our nGeniusPULSE software offering that is used to diagnose the root cause of issues impacting infrastructure performance and actively test software-as-a-service applications.
This capability complements and amplifies the service assurance solution delivered by our core technology and eliminates any need for third-party tools in the monitoring and troubleshooting workflows for our customers' mission-critical applications. Our software innovation also extends to our packet broker offering. Earlier this year, we decoupled our packet broker software from hardware with our new PFS 5000 model. By creating an open compute platform option for network packet brokers, we have taken a unique, disruptive approach that is starting to resonate in the marketplace. Orders for the PFS 5000 are now outpacing those for our traditional hardware-based packet brokers. We expect that to further accelerate next month when we introduce support of inline security tools.
To go to market, during the quarter, we consolidated the previously separate NETSCOUT and Arbor marketing teams to help us deliver a more unified presence to the marketplace and better align with our strategy to deliver combined service assurance and security solutions based on our Smart Data platform. We also anticipate higher levels of collaboration and coordination between those sales teams as we move into next year. We expect to roll out the results and further details of this unification, along with our success stories with our new software and virtualized portfolio at our annual Engage customer and sales meeting, making it the most exciting and impactful NETSCOUT user forum yet. That concludes my prepared remarks. At this point, I will turn the call over to Jean.
Thank you, Michael, and good morning, everyone. This morning, I will review our third quarter results, key revenue trends through the first nine months, and our revised fiscal year 2018 guidance. As a reminder, this review focuses on our non-GAAP results unless otherwise stated, and all reconciliations with our GAAP results appear in the presentation appendix. Slide number 11 shows our results for the third quarter and first nine months of fiscal year 2018. Focusing on our third quarter results, total revenue decreased by $39.3 million, or 13%, to $272 million. Our overall gross margin changed by approximately 250 basis points to 80.2%, primarily due to the one-time adjustment related to variable incentive compensation. Our total operating expenses decreased by $23.6 million from the prior year, largely due to the aforementioned changes in variable incentive compensation and lower sales commissions. The operating profit margin for the quarter was 30.9%.
Recently enacted tax legislation reduced our third quarter effective tax rate to 25%, which contributed to diluted earnings per share of $0.69. Turning to slide 12, I'd like to review key revenue trends through the first nine months. Revenue in our service provider customer segment declined 16% through the first three quarters. The decline reflects two primary factors. First, as we have discussed on prior calls, one of our large tier 1 service provider customers has been moderating its purchasing over the past two years. Second, revenue for Arbor's DDoS solutions decreased by the mid-teens. These dynamics were partially offset by low single-digit growth in all other service assurance service provider accounts. As Anil noted earlier, we expect that service provider capital spending pressures will continue to result in softer order volumes for our service assurance and DDoS products in the fourth quarter.
Our enterprise vertical declined by approximately 8.5% through the first three quarters. Enterprise revenue for legacy NetScout offerings declined by mid-single digits through the first nine months. This was compounded by weakness across all other product areas. Most notably, we've continued to see erosion across the former Fluke product lines. Following a very soft third quarter, Arbor's enterprise security revenue only grew low single digits through the first nine months. For the first nine months, the mix of revenue was 53% coming from service provider and 47% from enterprise. In terms of revenue by geography, which is calculated on a GAAP basis, our revenue in the U.S. declined sharply during the first three quarters, largely due to the decrease in revenue from that large tier 1 carrier, while international revenue declined by 5%.
International customers represented 40% of total revenue through the first nine months of this year versus 38% in last year's comparable period. We did not have a 10% revenue customer for either the third quarter or the first nine months. Slide 13 details our balance sheet highlights and free cash flow. We ended the quarter with cash equivalents, short-term marketable securities, and long-term marketable securities of $383 million, an increase of $69.6 million from the end of September. Our free cash flow for the third quarter of fiscal year 2018 was $72 million, and it was $135.9 million for the first three quarters of the year. Our third quarter free cash flow reflects favorable changes in working capital due to the collection of receivables and the lower sales volume. We currently anticipate that free cash flow for the full year will be more than 100% of non-GAAP net income.
As detailed earlier this month and again in today's press release, we are planning to execute an accelerated share repurchase of $300 million later this week. As illustration, using yesterday's closing stock price of $26.10 per share, the ASR would enable us to repurchase approximately 11.5 million shares. While we anticipate that it will take the banks working on our behalf between two to three quarters to execute the buyback, the accounting treatment allows us to reduce our share count by approximately eight million shares, or approximately 70% of the ASR, immediately upon entering into the agreement. Once the repurchase is technically completed, the share count would be further reduced by the number of shares actually repurchased, excluding the number of shares reflected in the initial reduction.
Although there will be minimal benefit to our diluted earnings per share in fiscal year 2018, it will enable us to significantly reduce the number of shares outstanding next year. For the third quarter, we did not repurchase any shares. We plan to fund the ASR primarily through additional debt of $300 million. We anticipate that our net leverage for the fiscal year ending 2018 will be about one time EBITDA. To support this activity, we entered into an amended and expanded credit agreement earlier this month that upsized our existing credit facility. The agreement provides for a five-year, $1 billion senior secured revolving credit facility, which is 25% larger than the original agreement, and it has better pricing and more favorable terms and conditions compared with the original agreement. To briefly recap other balance sheet highlights, accounts receivable net were $249.9 million, down from the end of last year.
DSOs were 82 days versus 83 days at the same time last year, and 80 days at the end of fiscal year 2017. Moving to slide 14 for guidance, I will focus on the non-GAAP guidance and remind you that the reconciliation of our GAAP guidance to our non-GAAP guidance is in the appendix. Anil spent considerable time detailing the drivers associated with lowering our revenue outlook for fiscal year 2018 from the original target of approximately $1.2 billion, to between $1 billion and $1.025 billion. In our service provider customer segment, we now anticipate a full-year revenue decline in the range of 22%-25% from fiscal year 2017 levels due to continued capital spending pressure from Tier 1 carriers, primarily in North America.
This is impacting the timing and magnitude of order levels for the company's service assurance solutions and, to a lesser extent, capacity-related orders for Arbor's DDoS offerings. In the Enterprise, we anticipate that fiscal year 2018 revenue will decrease between 5%-7% from fiscal year 2017 levels, primarily due to weaker than expected federal spending, longer than anticipated sales cycles, and softer than expected orders for certain other enterprise offerings. Given our revised outlook for the full year, we anticipate fourth quarter revenue in the range of roughly $240 million-$265 million. The wider revenue range for the fourth quarter reflects some uncertainty about the timing of certain orders across both customer segments. We currently anticipate that gross margin should increase to around 77% for this year, which implies that fourth quarter gross margins are likely to decline modestly from third quarter levels.
This is primarily due to the one-time variable incentive compensation adjustment from the third quarter, as well as lower sales volume. We anticipate fourth quarter operating expenses that will increase modestly from the third quarter levels by about $13 million-$15 million, again, due to the third quarter's adjustment related to variable compensation. Other full year modeling assumptions include changes to the tax rate, interest expense, and the diluted share count, each of which are outlined on this slide. Based on the quarterly impact of these assumptions, this would translate into fourth quarter diluted earnings per share in the range of $0.25-$0.40. That concludes my formal review of our financial results.
Before we transition to Q&A, I will note that we plan to participate at the Morgan Stanley Investor Conference in San Francisco at the end of February, and we plan to augment that with meetings in other major money centers over the coming weeks. I'll now turn the call back to Andrew Kramer before we start Q&A.
Thanks, Jean. As we've outlined, we're moving forward focused on the opportunities that we believe can lead to improved performance in fiscal year 2019 through both top-line growth and actions to adjust our cost structure. We will plan to share more details about the coming fiscal year in early May when we report our fourth quarter and full year fiscal year 2018 results. Keith, let's begin the Q&A session at this point.
At this time, if you'd like to ask a question, please press star and one on your touch-tone phone. If you wish to remove yourself from the queue, press the pound key. We do ask, in the interest of time, that you limit yourself to one question and one follow-up. Thank you. We'll take our first question from Chad Bennett with Craig-Hallum. Please go ahead. Your line's open.
Great. Thanks for taking my questions. Anil, I mean, you've always had a very good software-only solution. I think the market is probably moved at least on the service provider side, maybe quicker than maybe you guys anticipated, or maybe even us. I guess, if we look out over the next couple of years, and I know it's not easy from a visibility standpoint, when you look at the tier 1s in North America, do you have an idea of mix of how much of network monitoring goes to software-only solutions versus a traditional appliances over the next couple of years? Can you take a stab at that?
Yeah, I think virtually the speed of which this may vary from provider to provider, but we think virtually all the sales into all service provider, including U.S. service provider within the next two years, will move to software.
Okay. I know that the logic or theory previously was, when the carrier's deploying software-only probes, they can put it in more of the network than they could appliances previously, and that net net, from a spend standpoint, there shouldn't be that much of a change to you guys or to the software provider.
That's true. That's true, Chad. I just want to mention that that was not true in the last two years.
Yeah
the spend has moderated to a level where that's going to be true now.
Okay. Got it. Maybe last one for me, and then I'll hop off. How aggressively are you seeing the enterprise customers that you deal with move to software-only solution? I don't think you cited it as kind of an issue on the enterprise side, but are you seeing any traction there? I'll jump off. Thanks.
No, right now, the people have not shown a lot of interest on the enterprise side to move to software. Just to summarize for everyone, that we have three kinds of solution. One is appliance, which you mentioned, which was most of our business or all of our business in the past. Software version of appliance, which is what service providers are using, and that one, enterprises have not shown a lot of interest. The third area, which was always software, but nobody was buying, which is the NFV and virtualization and service function virtualization. There, we are seeing some traction on both sides. We see most of the software on enterprise moving to virtual rather than to COTS.
Got it. Thank you.
We'll take our next question from James Fish with Piper Jaffray. Please go ahead.
Hey, guys. Thanks for the questions. I guess we'll start on sort of gross margins. They were impressive at 80%, and I know, Jean, you alluded to that this was a lot of one time in nature. Is there anything that makes you confident over the next 12 to 24 months that we could actually continue to hit that 80% range, whether it's mix of Arbor Networks or just better execution and less of the Fluke Networks? Just curious as to the sustainability of this 80% level.
The Arbor Networks is already in that range. NetScout originally 3 years ago, was also in that range. Our mix change with Fluke Networks and Tektronix. Most of them either have bottom out or moving to software. Yeah, there is a good chance of getting to that level in the next couple of years.
Okay. Probably, I know, Anil, you talked about this a little bit on the call here, one of the biggest questions that I've gotten is sort of what makes these changes by the large tier 1 carriers, not sort of the canary in the coal mine for the rest of the service providers. What gives you confidence that this is essentially the bottom for service provider being kind of down double digits?
I think one thing is we did hit the bottom in other area. Other than tier 1 provider, we are on the upswing, in generally in international, as I talked about our business and margins have increased in many of the areas when you talk about seven or eight top tier 1 providers outside of U.S. In U.S., it was slow adoption, there were a lot of other dynamic which prevented us from doing that. Don't forget that they have a very big install base, there is a renewal stream of revenue also. That install capacity requires upgrades, software upgrades, and they have to buy renewals. Because of all those reasons, we think that we have reached a level this year in tier 1 provider that we should go up despite their move to software over the next two years.
Got you.
Just to add to that, Jim, since I know you're newer to the story. The top 2 North American service providers had spent disproportionately larger on their 4G networks than even the other 2 top 2 U.S. providers. That spend has moderated by $200 million each over the last few years. Their range is more in line with the next 2 tier 1 U.S. service providers, but it's still probably a little elevated from that. It's really a story over the last few years of the moderation in spend on 4G.
Yeah. No, understood that, and not to beat a dead horse here, but
It seems as though if that's sort of moderating, that we should be talking about decent growth for next year. At the same time, it sounds as if some of the weakness this quarter was the other service providers not spending as much. That's more or less the sense that we're getting here.
Yeah. The other service provider, some of it was the timing issue. The big effect was what Jean was talking about. We feel that we have gotten them to a level where I think we have largely hit the bottom there in the short term because they will buy more software. It will not grow, in a couple of years, we should start seeing moderate growth in those areas.
Got it. Thanks, guys.
We'll take the next question from Mark Kelleher with D.A. Davidson. Please go ahead.
Thanks for taking the questions. Maybe just to follow up on the 4G. What are your expectations for 5G? Do you expect another surge? Are we still a year and a half, two years away from that? What's the expectation there?
When we went from 3G to 4G, there was a complete upgrade of the network. Whereas 5G is more, I would say, for lack of better word, maybe last mile. It does increase the traffic on the core, but it's not a whole scale upgrade of the infrastructure. The surge will not be anywhere close to what we saw in 3G to 4G. It will be a very small fraction of that. Yes, there'll be some traffic requirement, but it will not, I mean, traffic increase, which will cause more capacity requirement or more capacity from a solution, hence more units of monitoring. It'll be largely as a result of traffic growth rather than a network upgrade.
Okay. Just as a follow-up, could you just give a little more detail on your commentary on divestitures? I know you mentioned something about divesting some product lines. Can you talk about that?
Yeah. We have a Fluke Networks tools business, which is basically handheld tools, which we have been trying various things for the last couple of years, and we came to the conclusion that it's not very synergistic with the way we sell other product lines, and that's the area. It's less than 5% of our business right now.
Okay, great. Thanks.
Our next question's from Matt Hedberg with RBC Capital Markets.
Hi. Guys, good morning. Thanks for taking my questions.
Sure.
In terms of lengthening enterprise sales cycles, it sounded like a part of this is due to customers thinking through their cloud and digital footprint. I think last quarter, Anil, you talked about putting some additional sales incentives in place. I guess I'm curious, going into Q4, what else can you guys do specifically to help shorten those sales cycles?
I don't know whether incentive is directly, and we are doing incentive for other reasons, but cloud migration activities will follow at their own pace. It's not really necessarily cloud migration because customers have multiple choices in upgrading their infrastructure. Whenever you have more choices, it slows down the infrastructure upgrade process, which slows down our sales cycle. I don't think incentive can do that, but we have some new things coming up in our story about how we can accelerate their migration activities through our visibility provided by our product. I think that could actually shorten the sales cycle. We have been testing that story and positioning and some features with our large customer. I think we think that that'll be a big hit when we have the user group meeting in May.
That's great. Jean, I guess under the new tax code, can you help us with any potential benefit that you'll see to your GAAP or non-GAAP tax rates in fiscal 2019?
In fiscal 2019, our effective tax rate for this year had one quarter of the lower corporate income tax at 21%, so that's why you have the blended 28. Right now, what we would probably see is that 28 would go down by a few more points next year. That is very provisional given the fact that there's still pieces within the legislation that is still being worked out. On FY 2019, we mostly benefited from the tax rate going forward. On FY 2019, as we finish our planning processes, I confirm up whether 25%, 26%, 24% is a better rate for us to use for all fiscal year 2019.
Great. Directionally, that's helpful, though. Thanks a lot, guys.
You're welcome.
The next question is from Alex Kurtz with KeyBanc Capital Markets.
Thanks, guys, for taking a couple questions here. Jean, just to continue on the model. Anil talked about optimism for growth in fiscal 2019. How does that relate to free cash flow margin and EBIT margin?
Well, given that our cost structure has always been fairly stable, I would assume that increases in revenue would increase our EBITDA margin. Cash flow should also continue to be probably above 100%. Right now, in FY 2019, we're going to be moving from one of our buildings in Texas into another building, a smaller footprint, and that will have some CapEx that is reimbursed, and we're looking at the accounting treatment for that.
You expect to grow. If you grow revenue, is it an operating assumption that you'll be flat even margins now for fiscal 2019? Is that sort of how you're thinking about it? If there's growth, that's how you'll expand on op margins?
I think as Anil said, we were looking at some of the options in our cost structure, we could see an impact in operating costs. As the product continues to morph towards legacy NetScout and legacy Arbor, gross margins will have a higher overall gross margin. Usually, any flow-through in revenue growth, would flow to the bottom line that way.
Okay. Just a couple of last follow-up questions. The 8 million share reduction, we should pencil that in by the September quarter, that you will have 8 million fewer shares in the float, or what are your thoughts?
No. What will happen is, I think on Friday, we will start executing, and we will settle the accelerated share repurchase with the banks. Probably about 70% of the expected shares that we'll take out. I think in the script, I had used 11.5 million shares based on the closing price of the stock yesterday. 70% of 11.5 is like 8 million shares. That should come out of the earnings per share calculation on the beginning of February. It will have the full year effect in FY 2019, and then the remainder will be picked up probably two to three quarters after that.
Okay. Just last question. You made some mention about the Arbor growth rate being down mid-teens. Was that just service provider? If it wasn't, what was the Arbor growth rate expectations in total for fiscal year 2018?
I think we had an uptick because of some software pricing models and the DDoS attack going up, being very high last year. Very unusual year. We didn't have a high expectation of growth this year, but it was down. That was somewhat of a disappointment. Largely, it was down, as we mentioned, that the tier 1 capacity requirements and CapEx spending issues was bleeding into the Arbor DDoS business also.
Yeah. Just so to summarize the year to date for Arbor in the enterprise, they've grown in the low single digits. As Anil said, it is a decline in service provider, and Arbor attributes it to digestion of capacity that some of the large tier 1 U.S. providers bought last year after the Dyn attack.
Okay. Thank you, guys.
Yeah. Thank you.
As a reminder, star and one for your questions. We'll go next to Eric Martinuzzi with Lake Street Capital Markets.
Hey. My question is for Anil. I wanted to jump in the time machine and go back to October 2014 on the announcement of the NetScout acquiring the Danaher Communications business. Just strategically, obviously, that transaction was based on some assumptions. I want to use this as kind of a lead into a question on go-to-market strategy, what has been the biggest market shift if you go back to the logic that made so much sense in October of 2014 versus where we are now, January 2018, what's been the big shift?
Let me see how I can. Let me say, big question, overall, there have been a lot of ups and downs. There have been move to software model. There has been a lot of new product introduction. There has been things about overlap with the Fluke Networks and the tools business. Things have been up and down, and I think some of it was a surprise, overall on balance, I think one thing worth balancing something else overall, I think the biggest surprise at that time was the top tier 2 providers going from hundreds of millions of dollars in revenue, to coming down to much lower level. As Jean mentioned, at different times, they had spent over $150 million each or even $200.
That dramatically coming down as a result of their own CapEx spending pressure, OTT traffic driving even more growth, which puts even more pressure on them. I think if that factor was not there will still be lots of puts and takes, but outside overall number would be on the growth side, maybe single-digit growth. I think that's the highest level summary that was a big surprise and unanticipated at that time that all these things will happen over a course of three-year period. Other places where these changes are happening, we have been able to mitigate them, either through the software model or go-to-market strategy changes or substitution of the product. In fact, we are more competitive right now than we were at that time. Clearly, we were competing with each other before that. I don't know whether that covers your broad question.
That's where it leads me to my part 2 of the question, which is, that business, that $2 billion assumption back in October 2014, which closed in July of 2015, it was kind of a $2 billion run rate, and you had a picture or an idea of what the market was. Now, as we're over 3 years removed from the announcement of the deal and 2 and a half years from the closing of the deal
What's the right go-to-market strategy here? Do you have the right go-to-market strategy? Do we need to do something material in how we're Obviously, there's a reduced appetite on the part of those 2 large players.
I think we have a good, better product line, better vision, technology, new go-to-market strategy. It has been largely ineffective in the face of these other challenges. Those challenges have been mitigated in terms of resizing the business. I think we'll start seeing the positive effect of those strategies. I think those things internally are seen in the international service provider, where we have gotten, you heard about some of the Vodafone deal announcement. You heard about a $75 million deal we announced based on software alone, which was bigger than anything which has happened with that customer 5 years in the past. There are several good things going on in the company on multiple fronts, but they were masked by this effect, which we talked about, the Tier 1 provider.
Now because of we have resized the business, which also requires us to resize the expectation and some of the cost structure, as we talked about, which we'll share with you people in May. I think we'll be able to show a lot of that progress, moving forward over the next couple of years.
Okay. I look forward to that update.
Yeah.
We'll take our final question from Kevin Liu with B. Riley. Please go ahead.
Hi, good morning. Just with respect to your software only strategy, can you just talk in a little bit more detail? You mentioned that you wanted to go international first. Do you see a lot of opportunity even with your top 2 North American service provider customers today? When would you expect to start to close some software only deals with those folks?
I think any business we do with them or a large portion of the business we do with even these Tier 1 providers, the top 2, and rest of them also, where we have not used software for one reason or another, will be mostly software next year. In fiscal year 2019, we think bulk of the sales to date, we have introduced that, they are certifying that. One of the beauty of our software solution is feature compatibility with the appliance solution. You can use it in Tektronix mode, you can use in NetScout mode, you can use it in old mode or combined mode. That allows the people to basically say, "Yes, I can move to software," which affects the top line.
We think that we have bottomed out, and given that, we will see a better margin expansion as a result of any sales to this. It will take a couple of years to get back to some growth here, but I think we have gone to a new start here with them, even with the Tier 1 provider in U.S. We see, we have validation of this trend from international service providers, which were sort of one year ahead in terms of a deployment time horizon.
Got it. Just with respect to the other service provider deals where you kind of cited timing issues, can you talk a little bit about what sort of issues those are? Is it decisions over how to deploy or some other factor that would delay those deals?
Yeah. What I meant was that in service provider, the business is lumpy. For a couple of service providers, one in big international, one in U.S., there are some things we could have gotten in Q4 of this year. Now we may get in Q1, and that's what I was referring to. This is not one of those top 2 or 3 providers we were talking about.
All right. Thanks for taking the questions.
Thank you.
It appears we have no further questions. I'll return the floor to Andrew Kramer for closing remarks.
Great. I'd like to thank everybody for joining us this morning. Look forward to seeing folks when we're at the investor conference on the West Coast, as well as other money centers in the U.S. Certainly, if you have any questions, feel free to reach out to investor relations here at NetScout. We look forward to communicating with you on our next quarterly call in early May. Thank you very much.
This will conclude today's program. Thanks for your participation. You may now disconnect. Have a great day.
Bye.