NetApp Inc (NTAP) 2006 Q2 法說會逐字稿

Good day, ladies and gentlemen.

Thank you for standing by.

And welcome to the Network Appliance second quarter fiscal year 2006 Q&A conference call. (OPERATOR INSTRUCTIONS) I would now like to turn the presentation over to your host for today's conference, Tara Calhoun, Senior Director of Investor Relations.

Good afternoon and welcome everyone, and thank you for joining us today.

With me on today's call are Dan Warmenhoven, our CEO, Tom Mendoza, our President, and Steve Gomo our CFO.

In the course of today's conference call we will make forward-looking statements and projections that involve risk and uncertainty, including statements regarding our forecasted operating results and metrics, anticipated sales trends for our current and future products, our anticipated hiring activities, statements regarding anticipated benefits to be derived from our acquisition of Decru, and regarding anticipated benefits from our relationships with IBM and other industry partners.

Actual results may differ materially from our statements or projections.

Important factors that could cause actual results to differ from those in the forward-looking statements include, but are not limited to, customer demand for products and services, increased competition, any decline in general economic conditions, and foreign currency exchange fluctuations.

Other equally important factors that could cause actual results to differ from those in the forward-looking statements are detailed the Company's 10-K and 10-Q report under the section captioned, Risk Factors, on file with the SEC and accessible through our website.

All of which factors are incorporated by reference in today's discussion.

We disclaim any obligation to update information contained in these forward-looking statements whether as a result of new information, future events, or otherwise.

Today's call is being webcast live over the Internet, and will also be available for replay on our website at www.netapp.com, along with earnings release, the reconciliation between GAAP and non-GAAP numbers, and other financial and statistical information presented during the call.

At this point I will turn the call over to Steve to review this quarter's financials and our financial outlook for Q3.

And then Dan will share his thoughts on the quarter before we wind up with Q&A.

Steve.

Thanks,Tara.

Good afternoon everyone.

Network Appliance delivered strong second quarter results, driven largely by the strength of our new FAS 3000 midrange product line.

Overall our business bounced back solidly from the business transition issues we experienced in the first quarter.

Let's walk through some additional details of our results to illustrate.

Please note that all numbers comply with GAAP, unless stated otherwise.

For the second quarter total revenue came in at $43.1 million, up 29% over quarter two of last year, and an increase of almost 8% sequentially.

The foreign currency effect subtracted approximately .5 percentage point on a sequential basis, and added .5 percentage point year-over-year.

Product revenues of $424.8 million grew 26% year-over-year and 8% sequentially.

Add-on software and software subscriptions accounted for over 35% of total revenue again this quarter.

This figure is a combination of software subscriptions that were 12% of total revenue, and add-on software products that were almost 24% of revenue.

Revenue from IBM and Decru were both less than 1 percentage point each this quarter.

Revenue from services, which includes hardware support, professional services, and educational services was 12% of total revenue, up 8% sequentially and up 52% over Q2 of last year.

Both service maintenance contracts and total professional services grew 8% sequentially.

Non-GAAP gross margins were 62.5% this quarter; their highest level in over three years.

Although there was a slight uptick in the mix of add-on software, the impact of FAS 3000 was the most compelling influence on this performance, and the performance of our product gross margins, which finished at 67.5% on a non-GAAP basis.

The volume of FAS 3000s exceeded our expectations, especially the increase in the number of system heads shipped without disk.

This stems from the combination of an increase in the number of clustered systems shipped and some upgrades from our current midrange products.

We expect the head to disk ratios to normalize in Q3.

Non-GAAP service margins also experienced a healthy increase to 26.5%.

Service margins benefited from two credits this quarter, one for parts and one for European VAT recovery.

These two items contributed over 3 full percentage points to service gross margins.

We experienced non-GAAP -- excuse me, we expect non-GAAP service margins to be in the mid-20s for the balance of the year, as we continue to scale our service programs and offerings, particularly professional services.

Turning to expenses, our non-GAAP operating expenses totaled $215 million, or 44.5% of revenue.

Expenses increased approximately 4% from Q1, including the prorated effect of Decru.

Employee headcount increased by 331 people, which includes 83 people from Decru, ending the quarter with 4,332 employees.

We also ended the quarter with over 50 pending starts.

So while we remain somewhat behind our target to hire 1,200 people this year, we expect to get close to that goal by the end of April.

GAAP operating expenses include the prorated effects of intangibles amortization, stock compensation charges, and the $5 million write-off of in-process R&D associated with the Decru acquisition.

At $87.1 million, non-GAAP operating profit finished at 18% of revenue, the highest level in just over five years.

Non-GAAP other income finished the quarter at $9.4 million, primarily as a result of slightly higher interest rates.

Non-GAAP pretax operating income for the quarter was $96 million, or 20% of revenue.

Our effective non-GAAP tax rate was 18%.

Non-GAAP income -- net income for the quarter was $79 million, or 16.4% of revenue.

Non-GAAP earnings rounded $0.21 per share, including the dilution from Decru, and are based on approximately 385.4 million shares outstanding.

GAAP net income was $71 million, or $0.18 per share.

As mentioned earlier, this quarter's GAAP earnings statement includes several charges specifically associated with the acquisition of Decru.

In addition, we incurred a new restructuring charge of approximately $350,000 related to the move of the bulk of our global services center's operation from Sunnyvale to our new flagship support center at our Research Triangle Park facility in North Carolina.

We expect to incur additional restructuring charges of about $500,000 associated with this transaction during the second half of the year.

The full reconciliation between non-GAAP and GAAP financial results is provided in a separate reconciliation table in our press release, which is posted on the Investor Section of our website at www.netapp.com.

Moving to the balance sheet, cash and investments totaled $1.1 billion, down about $105 million from Q1.

We increased our stock repurchase rate this quarter buying back over 6.3 million shares of outstanding common stock at an average price of $23.56 per share, for a total cash outlay of roughly $149 million.

In addition, a net of $42 million of cash was spent on the acquisition of Decru.

Cash generated from operations totaled $102 million.

Capital purchases were almost 30 million, and depreciation and amortization totaled $19.7 million.

The deferred revenue balance increased by $49 million this quarter to $535 million, a 10% sequential increase and up 57% year-over-year.

Increasing deferred revenue is a strong leading indicator of our continued penetration into enterprise accounts.

Accounts Receivable days sales outstanding were 61 days compared to a low 49 days reported last quarter.

The increase was primarily due to the tremendous demand for the FAS 3000 products which strained our supply chain during the first two months of the quarter.

Orders were not as back end loaded as our DSO might imply, but shipments were back end loaded, with FAS 3000 on allocations for the first two months of the quarter.

We now have our contract manufacturer ramped appropriately, and have reached a more balanced relationship between supply and demand.

DSOs should return to the mid-50s next quarter.

Inventory turns were 16.4 times compared with 17.2 times in the first quarter.

Now before I turn the call over to Dan for comments, I will talk about our operating model for the third quarter.

Our outlook is based on current business expectations and market environment, and reflects our non-GAAP presentation.

I will remind you that we're making forward-looking statements and projections that involve risk and uncertainty.

Actual results may differ materially from the statements or projections.

We expect FY '06 third quarter revenue growth to be in the range of 25 to 28% year-over-year, which reflects about a 7 to 9% sequential increase over Q2.

We expect Company gross margins to return to a more typical level of around 61% in Q3.

Our target operating margins during the year are expected to remain between 15.8% and 16.4% of revenue.

Third quarter non-GAAP earnings are expected to be in the range of $0.20 to $0.21 per share.

GAAP earnings are expected to be $0.17 to $0.18 per share.

For the full year, revenue growth is expected to be in the range of 26 to 28% year-over-year.

Non-GAAP earnings per share are expected to be in the range of $0.77 to $0.80 cents, with GAAP earnings of $0.70 to $0.73 per share.

In addition, the NetApp Board of Directors has just approved a total of $650 million to be used to continue our stock repurchase program, including the $76 million remaining from previous authorizations.

We will be in the market periodically buying stock, depending on the stock price.

As a result, we are not expecting that our total shares outstanding will change significantly over the remainder of the year.

And with that, I will turn the call over to Dan for his update.

Dan.

Welcome everyone to my 40th consecutive quarterly earnings conference call.

I should point out that Network Appliance went public 10 years ago next Monday.

And if you want to hit the NASDAQ site next Monday, you can see most of the senior executive team opened the NASDAQ.

Before I begin the business update I would like to take this opportunity to introduce two new executives who have joined the NetApp management team.

Tom Georgens, the former CEO of Engenio, has joined NetApp as Executive Vice President and General Manager of Enterprise Storage Systems.

This is a new position that has been created to bring together our core storage systems businesses, which includes all of our FAS and NearStore systems, as well as software and connectivity.

The other new addition to the management team is Jay Kidd, former CTO and a VP and Product Manager at Brocade, who has joined NetApp as Senior Vice President and General Manager of our newly created Emerging Products Group.

As we continue to branch out into adjacent markets that are complementary to our core business, we want those newer areas to have the appropriate focus and resources to achieve our long-term growth objectives.

The products in this group include our V-Series heterogeneous data management solutions, the NetCash product line, the Alacritus VTL solution, and our most recent addition, Decru.

We are pleased to have those industry veterans as part of the NetApp executive team.

Now let's focus on Q2.

As Steve described, NetApp delivered a very solid quarter.

And results are strong across the board, particularly in our core business.

Total core systems shipped were up almost 20% sequentially.

And the highlight of the quarter was the rapid takeoff of the FAS 3000 product line.

It had a whopping 250% increase in units shipped, and accounted for almost 30% of all systems sold.

The FAS 3000 also had a very strong software attach rates.

But most importantly, the FAS 3000 demonstrated a trend that will likely be one of the key factors fueling our increasing competitiveness.

Our total petabytes shipped slowed to single digit sequential growth, from 35 petabytes last quarter to 37 petabytes this quarter.

Yet our revenues increased by 8%.

Why?

We believe it is because Network Appliance is delivering dramatic, demonstrable increases in storage utilization rates with our FlexVol and FlexClone technologies.

Customers are seeing disk utilization rates going from industry norms of 30% utilization up to over 50 or even 60% with NetApp.

So customers don't have to buy as much disk storage to achieve their business objectives.

But they do buy more of the higher-value, high margin system heads and associated software to manage their data.

Increases in our total capacity shipped may be less predictable for the next several quarters, as more customers recognize and adopt this value proposition.

But the bottom line is that FlexVol and FlexClone, two of the features in our latest Data ONTAP 7G operating system, are delivering measurable benefits and storage savings to customers.

And our FAS 3000 Series is the first product line to ship 7G as the default operating system.

FAS 3000 also had positive impact on the take-up of our SAN and iSCSI products this quarter.

Recall that this system was originally designed to be a quantum step forward for us in terms of bringing a highly competitive SAN product to market.

In Q2 a record 42% of revenue included block-based connectivity.

Of this 30% of revenues included SAN, and 18% included iSCSI.

The 6% difference between those numbers and the 42% I mentioned is the overlap where both SAN and iSCSI were included in one system.

A large percentage of these orders also included NAS.

Now with such compelling price performance leader in a mid-range it appears that customers are increasingly grasping the value of unified storage.

And NetApp is the only vendor who can offer all three protocols concurrently in one system, which allows customers to consolidate storage projects.

So again they buy fewer disks, but they buy the systems from Network Appliance.

We expect that FAS 3000 will drive increasing market share gains in the SAN market, and extend our leadership position in iSCSI.

With the increased use of SAN, particularly in a FAS 3000 Series, we also saw an increase in the mix of Fibre Channel drives purchased, especially on the 3050 systems.

Overall, Fibre Channel accounted for 55% of total storage capacity this quarter compared to 53% last quarter.

It appears that the bifurcation between the FAS 3000 with its ATA drives and our NearStore product has settled out.

NearStore held its own, accounting for 15% of revenues compared to 18% last quarter.

While the overall number of NearStore systems shipped decreased, the ASP for the NearStore systems increased, as did the average capacity per system.

So it is clear that NearStore is now being truly used for secondary storage.

And a strong software attach rates for business continuance, backup and disaster recovery products demonstrate this segmentation.

As a consequence of the effectiveness of the FAS 3000 with ATA for smaller configuration deployments, we have canceled our planed introduction of the NearStore R210.

On the high end, the FAS 960 continued its decline, as it has slowly subsumed by the FAS 3000.

But more interesting to us was that we saw a decline in the 980 this quarter.

When we looked into this further we found that some customers are buying clustered pairs of the FAS 3050s instead because the performance of this system is so remarkable.

The total number of clustered pairs we shipped this quarter increased sequentially by 47%.

Our high end refresh coming before the end of this fiscal year, will bring substantial increases in both performance and capacity, well above that of the 980, bringing back clear segmentation between our mid-range and high-end products.

Our low end FAS 200 Series continues to show strength, with FAS 270 units up over 10% sequentially again this quarter.

NetCash continues to chug along, again contributing about 3.5% of revenues.

With a product refresh this quarter, we expect to see increasing contribution from NetCash by the end of the year.

We completed our acquisition of Decru at the end of August.

The concept of securing data at rest really resonates with customers, and we have high expectations for this business.

However, we have also learned that we were a little aggressive in our near-term revenue targets, and here's why.

Our projections are based upon early success in the Department of Defense.

However, in the commercial sector we have experienced longer sales cycles.

The difference in sales cycles is attributable to the DoD mandated policies for data protection, which are very proscriptive about what data should be protected, and who should be granted access rights.

That simplifies adoption planning.

In the commercial sector it is clear that the customers want our products, but the deployment planning is protracted because they take longer to decide where in the infrastructure to deploy our solutions, and what level of security standards they want to have as an organization before they can implement the key management policies.

As a result, less than 1% of our revenues came from Decru this quarter, and we expect it to contribute about 1% again in Q3.

But I believe now we understand what it takes to complete an enterprise sale, and we are hiring dedicated professional services people to help streamline this process.

Looking at the business from a geographic perspective, the Americas accounted for almost 60% of revenue, and EMEA contributed 29% of revenue this quarter.

Both geos turned in very strong performances.

Asia-Pacific continues to be inconsistent, coming in a little over 11% of revenue, with Japan continuing to be weak.

At the end of Q2 we completed a search and hired a new Vice President to lead Asia-Pacific.

And we're investing to provide more support in that region.

Our indirect channel business grew to 60% of revenues in Q2.

Most notably the combination of Arrow and Avnet jumped to almost 13% of revenue, posting about a 40% sequential increase.

In addition, our federal vertical jumped up to 15% of revenue this quarter, and over 80% of that came through indirect channels.

And in our top enterprise accounts, over half of the revenue coming from our top 20 customers in the quarter came through systems integrators in partnership with our own NetApp salespeople using a very high touch sales model.

Each of these components contributed to the increase in indirect channel contribution.

As Steve mentioned, IBM was less than 1% of revenue this quarter.

As we have talked about previously, we do not expect material contribution from IBM for another few quarters, although the partnership is progressing very well.

In fact, just last week IBM announced the launch of their version of our 3000 Series mid-range products.

Our Spinnaker integration efforts also continue to progress.

Although we have about a one quarter delay in the release of ONTAP GX, our next generation operating system, due to the complexity of integrating millions of lines of code.

ONTAP GX will be the first operating system to support a true storage grid, using modular storage devices that can scale to hundreds of systems, within a single global name space.

Early on it will be targeted at high-performance computing environments, and it will be optional on our high-end systems.

In other words, we plan to offer 7G, our current operating system, as the standard for enterprise data centers over the next couple of years, with GX available on a parallel path as a customer option until they are fully converged.

We will show you this technology at our analyst day, which was moved to March 14 of 2006, due to a conflict with another large cap tech company's analyst day.

To wrap up, I believe we're back on track, driven in large part by the success of our leading FAS 3000 mid-range products.

We believe our FlexVol and FlexClone technologies are the beginning of a paradigm shift in the industry, using software to virtualize storage and improve utilization rates.

We will continue to work on growing across several different axes.

And we are pleased with the expansion of our core business, our SAN business, and of course our business with partners like IBM, Oracle, SAP and Symantec.

We plan to keep hiring people in sales, professional services and engineering to fuel our future growth.

You should expect us to return to the 16% operating margin target, which we believe optimizes both long-term revenue growth and EPS growth, and allows us to continue posting the strongest growth rates in our industry.

And with that, I'll now open the floor to questions.

But, again, I ask that you limit yourself to one question each.

If you have a follow-up question, please return to the queue so we may address everyone in a timely manner.

(OPERATOR INSTRUCTIONS) Harry Blount, Lehman Brothers.

A quick question, I guess, in terms of longer-term thought process.

It seems like with the launch of SQL Server, the IBM relationship, Decru, you got some new low end products coming later this fiscal year, what is it going to take to get back to a 30% topline growth?

I noted the numbers are still in the 25 to 28% range -- very healthy rates.

But I know the goal is to get north of 30.

What is it going to take to do that?

This is Dan.

It is really going to be really, I think, a question of sales productivity, shorter sales cycles, higher win rates, etc., more indirect channel contribution, all of which kind of fits into sales productivity model.

And we're very focused on ways of improving our overall field effectiveness.

I should point out that we are kind of hampered on the year-over-year comparison by about 2 points of currency movement.

And that would be just enough to push us over that 30 bar.

But I have to tell you we're very focused here on trying to get back over 30% growth year-over-year.

The entire Company is committed to that kind of number.

Tom Curlin with RBC Capital Markets.

Further to that topic, just given the comments on further penetration in the Fibre Channel SAN world, should we assume that some of these recent hires, both of whom are SAN industry veterans, suggest a greater focus on driving market share gains in Fibre Channel SAN environments, and perhaps even trying to evolve the culture internally towards that end?

I think that is a pretty good assumption.

A couple of employees pointed it out to me as well.

We still have a core orientation towards NAS, and we're not walking away from that.

But we have set our sights on gaining share in the SAN market.

Our strategy in a nutshell is continue to lead the NAS markets,

be the leader in the iSCSI market, and have a share position there that is about the same as in the NAS market, and then continue to gain share in the SAN space.

The real strategy in a nutshell is the unified storage message, right.

Any flavor the customer wants, he can get the best solution possible from Network Appliance.

But in order to make that a reality, we've got to be a real player in the SAN space.

Keith Bachman with Banc of America.

I was just hoping to understand a little bit more about the services run rate.

Could you talk about that?

It seems to be doing fairly well.

What are your expectations given the recent performance has it changed at all?

Combined with any color on the gross margins of that particular organization.

I think you should expect the service topline to continue its performance that you saw this quarter.

There's no reason why we expect service to slow down from a revenue growth rate standpoint.

We've got lots of new service offerings.

They are compelling.

We have been investing in professional services, and that organization is really starting to hit stride now.

As far as the gross margin is concerned, we do expect that some of the benefits that we saw this quarter, particularly the onetime credits, Keith, will not repeat next quarter.

I would look for next quarter's service margins to be in the mid-20s.

Ben Reitzes with UBS.

With regard to the new product that you are going to launch at the high-end, Dan, what kind of contribution are you looking for this product?

And could there -- how does it differ -- I know since it is higher end, you probably don't have as big a disruption risk.

And can you talk about how much contribution you expect this quarter, and then to the overall growth rate from this type of high-end foray -- maybe for the rest of the year?

The new high-end system will not contribute anything this quarter.

We are planning as always by the Q4 of our fiscal year, not the calendar year.

You won't see any revenue contribution to this quarter.

It will probably be coming to market, certainly on a limited basis, on our fiscal Q4.

But the objective of the product is to be a complete replacement for our high-end 960 and 980 pair.

You should expect two models, much like we did with the FAS 3000.

And it is going to be well above the 980 in terms of performance and capacity.

It is the first time we've had a 64-bit microprocessor, and it makes a big difference in terms of bus throughputs.

And I think you'll find it to be quite a behemoth.

And not expecting any disruption in this quarter from it?

No, in fact nothing in Q3 -- no disruption, no issues whatsoever.

But in Q4 I would expect to see, especially for some of our longer-term and more performance sensitive customers, that they will start adopting the 980 pretty quick.

But you know it really will be additive.

I think it is going to be an accelerator for Q4 to the extent that the customers decide to adopt it right away.

Don't forget there our typical sales cycle on new system is 60 days or something like that.

Testers really take an eval and make sure they're comfortable with the whole thing, and not necessarily commit to it until after they have been through that level of evaluation.

So even in Q4 I would expect the revenue to be somewhat diminimus.

But we will see.

It is going to be introduced in Q4.

Like I said, the performance sensitive customers will take it.

And we will just have to see what that equates to relative to incremental revenues.

Mark Kelleher with Adams Harkness.

I was just wondering if you could maybe talk about IBM a little bit?

The FAS 270 shipped in the quarter, and the 3000 is starting this quarter.

Yet we're only looking for 1% of IBM contribution next quarter.

What are sort of the touch points to look for in terms of ramping those products a little stronger at IBM?

The plan was always it would take them about a year to introduce the complete product line and to roll it out across all geographies and all channels.

It is a progressive type of roll out.

And I think the plan has always been they will build over time in terms of revenues in our mix.

We have no indication from them for this particular quarter, which is their fiscal Q4, what the volume might be on the FAS 3000.

It is brand-new.

And I think that is really going to be the first indicator.

At the end of this next quarter we can kind of give you an indication I think as to what the volume might look like going forward.

But to be very honest with you, this is a new product for IBM.

And we are all proceeding fairly cautiously with respect to all the support infrastructure, manufacturing infrastructure, etc.

And our expectations are that until all of that infrastructure is really smooth that you won't see a volume ramp coming through.

Richard Chu with SG Cowen.

I wonder if you can give as any metrics that give us a sense of how much progress you are making towards hitting your 70% large enterprise attrition goal a few years out?

I'm sorry, Richard, you dropped off at the end of that question. (multiple speakers).

I believe, perhaps a year ago, you talked about -- you tried to get to a 70% enterprise penetration mix.

And I wonder how much progress you're making towards that?

I don't remember the 70% number.

But I can tell you -- I recall the sign we put up -- and I think it was Rob Salmon put it up in the analyst meeting -- that we were really trying to go for one-third from top enterprise accounts, one-third from I guess what you consider to be the Fortune 2000, and about one-third from what you'd consider to be down to the Fortune 10,000 or 15,000 or whatever.

Our view is just roughly one-third, one-third, one-third.

So maybe that is where the enterprise number got to 66 or something.

This particular quarter enterprise revenues were up 10 points in a quarter over quarter comparison.

I think we're making great progress.

But you know our profile right now is just about 60% from enterprise accounts.

And that includes the top two segments I was referring to.

I should point out also that the government agencies are in there.

A number of our top enterprise accounts are either in DoD or in the civilian agencies.

Thank you very much.

We have done extremely well on the enterprise business.

And as I look at the mix, we are actually a little ahead of our plan relative to where we expected to be with adoption in the enterprises.

Laura Conigliaro with Goldman Sachs.

A question about operating margins.

You mentioned that your target for the year was still 15 8 to 16 4.

But yet if you use this quarter's operating margins and then basically back into the operating margins based on the targets you are giving for the next quarter, it would imply some pretty low operating margin levels for the fourth quarter.

Can you give us some further observations about what you are thinking as far as operating margins, and the hiring that goes along with that?

And are you being sort of very conservative just to see how it develops?

In other words, is there room for operating margin upside?

This is Steve.

So maybe I wasn't clear on the script part of the call.

But basically our objective is for each quarter we're looking at 15.8 to 16.4.

That is not to say the full year is going to be there.

We already have demonstrated a pretty strong quarter here that is obviously going to have an impact on the full year.

But I think that Dan made reference to it earlier.

We think that that level of operating profit is the level that optimizes our ability to grow the top line, and to turn in the fastest-growing EPS that we're capable of over the long haul.

Andy McCullough with Credit Suisse First Boston.

Just to follow-up on that question, Steve, I think you are pretty clear about the balance of growth and profitability, and how you guys are managing the business.

But in terms of the hiring initiative that you guys are talking about for the full year, how much of the remaining target do you plan to get done in the third quarter versus fourth quarter?

Steve again.

I think you're going to see it spread pretty evenly throughout the quarters.

We're looking to do about roughly 300 a quarter when we started the year.

And we're going to pick up the pace a little bit here, if we are capable of doing it, but that is certainly our plan.

And that is how we get the operating profit numbers we talked about.

Joel Wagonfeld with First Albany.

A follow on to those two questions.

Just looking out longer term, for this year I think at the top end of your ranges you're talking about EPS growth kind of roughly in line or slightly above revenue growth.

For next year should we assume that EPS would grow at or slightly below revenue growth?

And if so, at what point do you get, do you think, to a point where you start to see positive operating leverage in the model, or is it just kind of growth for the foreseeable future?

We will probably have to get back to you at our securities analyst conference in March with an answer to that question.

We haven't made any kind of disclosure or announcements about what our FY '07 plans are.

Right now we are concentrating on optimizing FY '06.

And we're working on our plans going forward.

That said, I would like to respond to the last clause you had.

I think it is all about growth for the foreseeable future and trying to maximize both our revenue growth, market share, etc, over that period.

I do think we are in a market which is undergoing significant structural change, market share shifts, primarily away from server vendors and towards pure play storage vendors.

And I think it is incumbent upon us to ensure that we maximize the opportunity to gain share over the next several years.

Furthermore, I think our history has shown that the way to optimize the EPS growth is in fact to drive the revenue line as hard as we can.

And if we don't reinvest enough in revenue maximization, then ultimately it shows, and not just slowing revenue growth, but even a slowing in EPS growth.

I don't see anything in the near term that would change that outlook whatsoever.

And then could I just ask a follow-up.

In terms of after this year you have talked about the services personnel a lot.

Going forward with the additional services personnel would it more on the R&D side or would it be investments in the indirect channel to further expand that.

We would be happy to put you back in the queue, Joel.

Rebecca Runkle with Morgan Stanley.

I love your discipline guys.

Shifting gears slightly, on the buyback front you more than doubled the amount you are spending on buyback over the last couple quarters.

Can you provide some more color in terms of expected levels going forward, especially given the incremental program that you announced today?

And I assume that going forward you'll still manage it along the lines of buying back enough to offset the dilution, but will you just comment there as well?

Rebecca, this is a Steve Gomo.

A couple of points of note.

First, the levels we're talking about have an unspecified time frame with them, just to make sure everyone has got that on the record.

The amount of cash we are talking about here is excess cash.

We have, in doing our analysis of what we can afford to buy back to make sure that we have enough cash for operations.

And we have a significant amount of cash left in what I will call strategic reserve for any strategic initiatives that the Company may want to pursue.

As we have looked at this, we set a strategy for ourselves that has -- it is kind -- two aspects to it.

The first aspect of it is our intent is to try and eliminate the dilution associated with employee stock options that are granted.

The second aspect of it is, is that on top of that, to the extent we have excess cash, which we do, we're going to go back and eliminate the dilution associated with acquisitions that we've made, both historically -- and then it will take several years to catch up there.

But then on go forward basis, we would like to do diminimize that, or bring that to a neutral position as well.

So that is the strategy behind what we are thinking about here.

And again, it is based on the fact that we have excess cash.

We have enough to run our operations, and a strategic reserve as well.

I should point out that we have gotten the question many times -- why not just make the acquisitions with cash?

And the answer is that is not necessarily our prerogative.

It is normally determined by the seller.

And it is really a function of tax consequences.

And in general the favorable tax treatment goes away if there is more than 20% cash in the mix.

So in general our sellers, if you will, prefer at least 80% of the transaction to be in stock, which they can time the ultimate capital gains transactions around.

Kevin Hunt with Thomas Weisel Partners.

Maybe you said this and I missed it, but did you say what the NearStore was in the quarter?

And you kind of had been talking about maybe that wouldn't be as strong growth in the future?

NearStore was 15% of mix.

That was down a little bit from last quarter.

NearStore had been ramping up really over the last several years.

I think it reached a peak of 18 to 20% in our mix, and dropped a little bit last quarter to 18.

And this quarter it dropped to 15.

What we have observed is that the smaller configurations that NearStore customers used to buy under the NearStore brand are now shifted to the FAS 3000 Series with ATA.

So I think up to about 8 terabytes kind of flipped in a hurry.

And so the consequence of the NearStore configurations are higher capacity, and they are also higher ASP.

But the low end units, as I say, just kind of bled off right into the 3000 Series.

We expect it to still be a very attractive product line going forward. 15% is a pretty good contributor.

Dan Renouard with Robert W. Baird.

I just wanted to circle back on Decru.

I am just kind of wondering if you could give us a little bit more detail on what happened in going forward?

It sounds like there was, maybe not so much execution issues it is just you didn't forecast very well.

Was there also some turnover on the sale side, or did you lose some people?

And then secondly, just on the government -- I know a fair amount of the Decru sales and probably the pipeline was tied in with the government -- the federal government.

Do you expect any impact from Katrina, budget issues, etc., did that have any impact?

And then just lastly, all related to Decru?

The Decru sales issue is really around predicting the sales cycle.

And the federal government, as I indicated, there are DoD mandates that really are fairly proscriptive around what data has to be secured, and what are the access rights associated with various personnel in various positions.

And what we found in the commercial sector is it is just a lot more decisions to be made on the part of the customer as to where they want to deploy it?

What day do they want to encrypt, and how they want to structure the access rights?

And that has taken a longer time than certainly we would have thought.

The DoD mandates obviously streamline the adoption process for the government.

And there is no such framework like that in the commercial sector.

So we were a little optimistic on sales cycles.

But I have got to tell you, the funnel is pretty wide.

And I believe it is still going to be a significant producer for us.

We're not backing off of the objectives we had for next year.

I think it is going to be accretive within the year.

Let me add to that.

It is Tom Mendoza.

First of all when we acquired Decru their primary salesforce was into the federal government.

They had just started to expand out of that.

The interest level outside of that is intense.

I personally just finished a five-week trip all around the world.

All around the world people have come to a couple of realizations.

One, the FBI put out a report in the last 60 days that 50% of all data theft comes from inside the firewall.

Everyone agrees they haven't protected that well.

Secondarily, Gartner Group put out a report advising all enterprises they should be encrypting their tape.

Decru, because it is an appliance, and you can do it at wire speed without changing a wrap -- everywhere I go they are considered the leading force to do this.

So we are in many, many, many accounts that are non NetApp accounts.

We have a lot of partners working with us to sell Decru into non NetApp accounts.

I think the entire field is so open that it is creating a lot of excitement.

We're very, very focused on maximizing this opportunity, but I don't think federal will necessarily be the big play over time.

I think financial services is going to be huge.

And I think it is more of a broad application than the original Decru emphasis was.

And we're going to take advantage of that.

This is Dan.

The other part of your question was about salesforce attrition.

We didn't lose anybody that we were interested in keeping.

And in fact, employee turnover at Decru in general has been very low, almost nil.

That said, I will point out that Dan Avida, the former CEO, has decided to move on.

And that that was not unexpected.

It was a little earlier than we had thought, but not by much.

And he has been replaced with somebody I think most of you know well, Suresh Vasudevan, who used to run all the core systems of product management, is now the General Manager of Decru.

One other point I will make, to give you some sense in terms of numbers.

Decru is actually less than .5% of revenue this quarter.

We expect that to double next quarter to the 1% range.

And probably go up to 1.5% the subsequent quarter.

So it is on a pretty good growth clip.

We were just a little more aggressive at the original outset than we think was warranted at this point.

The other thing I would tell you is a lot of the trials we're doing the vast majority are on very, very, very high profile accounts -- large accounts.

And what they're going to do is they're going to test it.

They're going to deploy some of it.

And then they're going to deploy a lot of it.

It is kind of like the early days of the pilot.

But the interest level in all the major accounts that I have called on -- literally all are interested in what Decru is doing.

And specifically because they are owned by NetApp.

Because almost all of the encryption companies that are out there are very small companies, and people are concerned about that.

So I think that we gave them a validation and a comfort feeling that they can do this because NetApp is behind them.

Clay Sumner with Friedman Billings Ramsey.

Congratulations on great results.

Can you go back a little bit to the issue of shipping more filer (ph) heads than disks this quarter.

Why will that return to a more balanced level in the future?

Just kind of where the factors that worked there?

Clay, this is Dan, and thank you for the congratulations.

I want you guys to remember that we are proof positive that there is such a thing in Wall Street as a one quarter problem.

That said -- I was waiting for the congratulations from somebody, just so I could use that line.

The phenomena of the head upgrade I think is what drove that particular mix.

We saw a significant number of system units shipped without disks, which are primarily intended for one of two purposes.

Either field upgrades of existing systems like 940s, or alternatively the addition of a second head to an existing system for purposes of clustering.

And that is the kind of thing that happens fairly quickly in an installed base.

It moves pretty fast.

And I would expect that mix to shift back to a more normal mix next quarter.

So people were clustering 960s.

I'm still not quite understanding you, Dan.

Yes, many stand-alone configurations, they were clustering -- let's say 940's with a stand-alone configuration.

They would upgrade that not just to a single stand-alone 3020 or 3050, but to a dual.

Glenn Hanus with the Needham Associates.

Any update on the competitive front with -- EMC have been -- had some fairly aggressive tactics of bundling software and services.

Any life of -- out of the HP?

Just kind of wondering anything incremental that you have been seeing out there are over the last three, three to four months?

No, nothing of that level of drama and intrigue.

In terms of frequency of competition, Hewlett-Packard was up in the mix this quarter.

Our win rate stayed very consistent against both EMC and HP.

So, yes, there wasn't really anything dramatic in the change.

I think the most notable statistical difference had to do with the frequency of engagement versus Hewlett-Packard.

I also think that -- and we said this before, but I can't tell you how many customers have said that they now believe it is a two horse race.

It is NetApp and EMC.

They are the two companies innovating.

And many, many of the wins we had this quarter into new accounts simply were customers turning to us and saying, using your technology, how would you accomplish my business problem?

We have now reached a role of many of our customers.

We are no longer just talking about gigabyte prices and things like that.

They are allowing us to use our creativity.

And that goes back to Dan's point, why the petabytes were down.

We are using our new technology, showing them -- giving them higher utilization rates.

And that has given us major wins.

I would say that the competitive landscape hasn't changed, except we're starting to tilt it in certain ways because of product.

An interesting to phenomena that is going to occur in the metrics is that you are going to see I think a flattening of the cost per gig, because the customers can achieve their business objective with fewer gigabytes.

And so even though we can come in under in terms of the dollar value of a deal, it is with less starch because of the utilization rates being higher.

So it is going to be an interesting transition over the next few quarters.

Les Santiago with Piper Jaffray.

Could you give us some more color on the overall spending environment in the enterprise and SMB markets for storage?

How much of this trend this quarter was pent-up demand due to the product transition versus maybe share gains, or maybe some kind of a seasonal uplift in demand that you are seeing out there?

This was a typical Q3 from my viewpoint.

So seasonality was such that August was very weak, and we saw a pretty good strength in September and October, particularly at our federal, which always surges in the end of the fiscal year for the federal government.

So September -- that trickles into October from our viewpoint because the lag time through the system integrators.

But you saw the federal business really surge in the mix.

I didn't see anything relative to pent-up demand, other than the end of the year buying rush in federal.

It seemed like a very normal kind of model.

I would say that on our federal team too, it was a spectacular execution by that team.

A quarter ago after we didn't do well, (indiscernible) came here and said there was some very specific issue where things had shifted.

And by God, they went and got them this quarter, and they did it.

Another thing I would point out to you is I really think we're taking real advantage of some of our key partnerships.

I had the great opportunity to speak in front of about 14,000 people at Oracle World last quarter.

That paid off for with a lot of deals this quarter, where people were not looking at NetApp.

Dan is doing a lot of speaking at major events at companies like Symantec, SAP.

Microsoft just did a cross country tour with us over the last two quarters.

Those type of things really are giving NetApp a whole different visibility and profile in the market.

And I think that has helped us.

Again, as long as we get to engage in more deals, given our win rates, things are going to go better -- continue to do well at the rate we're looking at.

Getting back to the essence of your question, I think most of this is share gain.

And I think federal, as Tom was pointing to, is a pretty good example.

Our federal business is outgrowing the market by a lot.

And if you look at this particular quarter as compared to one year ago, roughly 30% of the bookings this quarter came from civilian agencies, whereas last year it was a very small percentage.

So as we have continued to expand our focus on different segments of the market, we have continued to gain share.

I think is just that simple.

Bill Shope with JPMorgan.

Looking at the other verticals do you see any surprising strength in particular verticals, other than the federal government?

Yes, financial services continues to be pretty good.

And actually the energy sector this quarter -- it is really going to be in this quarter.

Somebody asked about Katrina, we shipped an awful lot of equipment to our customers in the oil sector in the wake of Katrina and all the rest of the storms.

But that didn't show up materially in last quarter's revenues.

I think it may have a spillover effect into this quarter's.

The oil sector continues to do well.

The financial services sector continues to do well.

We're continuing to do very well in automotive.

And it is -- there are not many weak sectors right now.

I think based on the numbers I saw the tech sector is still 18% of mix.

And they are continuing to buy and doing well.

So it is pretty strong across the board.

What percentage of the business is energy this quarter?

I don't know -- it is roughly about 5%, 5 or 6%.

On that order.

Aaron Rakers with A.G. Edwards.

A real quick question.

You guys had talked about the FAS 3000 Series ramping, and the fact that is driving some increased software -- add-on software rates.

You have now seen three quarters either at or above 35% contribution.

Should we think about that contribution continually expanding as you continue to ramp the 3000 Series, as well as new high-end solutions going forward?

I don't think so.

The advice I would give you is to model about 35.

There is probably an error bar around that of plus or minus 1 or 2%.

I think we're getting to a zone where you are going to see some oscillations just based on normal fluctuations.

But I really don't think there's a lot of upside over the 35%.

Steve?

I agree with Dan.

In our model that we have in front of us right here, we are modeling 35.

Kaushik Roy with Susquehanna.

Let me be the second person to congratulate you on the nice quarter end, guys.

The product margins have jumped almost 100 basis points to 67.5.

What are your expectations for the product margins going forward?

And how does it change as IBM contribution increases?

This is Steve Gomo.

If I look going forward, here is kind of my anticipation.

We're looking for, as I mentioned, next quarter our margins in total for the Company to come down more to the 61% level.

I think there's several forces at work here.

First, I think that service is going to back off from its current level, as I mentioned.

We're not going to have the benefits that we saw this quarter.

I think software, as we just talked about, we're modeling 35%.

And I think that based on our analyses we could see some fluctuations around that number.

So to go above 35% is to, I think, take a risky profile with your model, because it could end up just slightly below 35%.

If you take the software back to the 35% level that is half a point based on what we did this quarter.

And finally, I think if you look at the ratio of our system heads to our disk capacity, if you will, while it may be a little bit more favorable than it has in the past, it is not going to as favorable as we saw this quarter due to the onetime events that Dan was alluding to earlier.

That's another half a point.

I think it is pretty reasonable to assume that we get back to about a 51% Companywide gross margin for the next couple of quarters.

How does it change as IBM ramps up?

Obviously they're going to put a downward pressure on your product margin.

We already have that figured into that mix.

Paul Mansky with Citigroup.

Actually, just a clarification.

In the prepared remarks I believe you indicated that petabytes shipments grew.

But I believe Tom suggested they actually declined in the Q&A a minute ago.

I just want to make sure I'm not crossing wires in my model here.

Which one was it?

Petabytes grew by about 2 petabytes.

We saw more efficiency -- that is the amount of storage a customer has to buy for a particular business purpose declined somewhat.

Yes, average configuration sizes went down a little bit year-on-year.

But what you see is the utilization rates go up.

So the growth rate is going to slow in terms of petabytes.

I think this quarter it was 8% sequential, whereas in most quarters it has been running around 15 to 20, even higher.

And it is just because of higher utilization rates.

I misspoke, sorry.

No, I just want to make sure I wasn't modeling anything wrong.

Kevin Hunt with Thomas Weisel Partners.

I just had one question on the tax rate.

It was a little bit lower than it was.

And what should we expect on that going forward?

And maybe what are the reasons why it is lower?

Kevin, this is Steve Gomo.

The non-GAAP tax rate did not change.

It is still 18%, and we're projecting 18% going forward.

And you should be using 18% for your model.

The actual GAAP tax rate, particularly with the new accounting conventions, will tend to be a little bit more variable, as basically we have to take discrete items in the period that we recognize them.

In the non-GAAP tax rate we are spreading those items over an annual period so they have less of an impact.

On a GAAP basis now we are required to take them as we discover them.

This particular quarter we had a -- we completed a R&D tax credit study, and indeed confirmed that we're going to get everything that we thought.

So we took an adjustment against our tax reserve as a result.

And then there was a transfer pricing analysis that was also done that caused another minor adjustment.

That's what drove the GAAP tax rate down this quarter.

Harry Blount with Lehman Brothers.

One of the questions that I know will come up from customers is you guys had indicated, based on the timing of beta units, that there might have been some spillover effect from last quarter into this quarter.

Can you maybe help quantify that a bit?

See, I don't really think there was much spillover.

The backlog coming into this quarter was very similar to what we had seen in the exit of prior Q1s.

I understand that speculation.

I read a lot about it.

I've got to tell you, it was not true.

It was not well founded.

So there was no spillover affect whatsoever.

Richard Chu with SG Cowen.

I wonder whether you could expand on the decision to not do the R210?

And perhaps comment a little bit more on the Spinnaker delay?

The R210 was intended to be kind of a low end of the NearStore family.

And obviously the low end of that was the addressed adequately by the 3020 with ATA drives.

So it really became kind of a redundant product.

They would have overlapped too much in terms of the marketplace, probably created some confusion in customers minds.

So we just said we've got that space adequately addressed with a 3000 Series, and there is no reason to have a separate product in the portfolio.

On the delay on the Spinnaker thing we continue to make good progress, and I'm very pleased with the progress.

However, this is an enormous undertaking to create a new version of our operating system.

It is the integration of millions of lines of code into a single base.

You have two separate code bases kind of being merged together.

And so we have found that the time required to finish the project is about three months longer than we had originally forecasted, which on a 24 month plan strikes me as not too bad.

And with that this concludes the question-and-answer session of today's conference.

Back over to the group for any concluding remarks.

Again, I would like to thank you all for joining us in this conference call.

Before we end, I would like to really extend my thanks and appreciation to the people at Network Appliance.

At the end of last quarter I kind of sent out some messages saying we have got to pick up our game.

And I would like to thank them for increasing their focus this quarter.

I think you see it in the results, and I surely am appreciative of the way they reacted.

We will look forward to seeing some of you maybe at the NASDAQ opening on the 21st, the tenth anniversary of our IPO.

And certainly I look forward to seeing you with us next February 15 for the results of our fiscal Q3.

And at the analysts meeting, I would like to mark your calendar for March 14 in San Francisco.

So again, ladies and gentlemen, thank you very much, and have a great day.

Ladies and gentlemen, we thank you for your participation in today's conference.

This concludes your presentation, and you may now disconnect.