Infinera Corp (INFN) 2007 Q2 法說會逐字稿

Welcome to the Second Quarter Fiscal 2007 Investment Community Conference Call of Infinera Corporation.

(OPERATOR INSTRUCTIONS)

Now, I will turn over the meeting to Mr.

Bob Blair of Infinera Investor Relations.

Mr.

Blair, you may begin.

Thank you, and good afternoon.

I want to note that the forward-looking statements included in today's call constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

These statements address the financial condition, results of operations, business initiatives, views on our market, our products and competitors' products, guidance, growth plans and prospects of the Company in 2007 and beyond, and are subject to risks and uncertainties that could cause actual results to differ materially from such forward-looking statements.

Please refer to the Company's current press releases and SEC filings, including reports on Forms 10-Q and 8-K for more information on these risks and uncertainties.

Today's press releases, Q2 financial tables and investor information summary and a guidance reconciliation summary will be available today on the investor section of Infinera's website at www.infinera.com.

The Company undertakes no obligations to update or revise any forward-looking statements to reflect events or circumstances that may arise after the date of this call.

I would also like to discuss the non-GAAP financial measures that are included in this afternoon's press release and today's conference call.

In our earnings release, we announced operating results for the second quarter of 2007 that included invoice shipment results and the impact of non-cash stock-based compensation and warrant revaluation expenses.

These non-GAAP financial measures are provided to facilitate meaningful year-over-year comparisons.

Please see the exhibit to the earnings release for a reconciliation of these non-GAAP financial measures to the most directly comparable GAAP financial measures and for an explanation of why these non-GAAP financial measures are useful and how they are used by management.

On this call, we will also give guidance, including guidance for the third quarter of 2007.

in this guidance, we will include non-GAAP invoice shipment results and exclude the impact of non-cash stock-based compensation expenses from our results.

Again, we have reconciled the non-GAAP invoiced shipment revenue results to our GAAP revenue results on the investor section of our website.

We have not reconciled the other forward-looking non-GAAP projections that we will discuss today, related to gross margin, operating expenses, net results and earnings per share because we cannot readily estimate the impact of customer product mix on our lower of cost or market adjustments and gross margins on the impact of our future stock price on our future stock-based compensation expenses.

Of the remainder of today's call, we will be discussing our second-quarter 2007 results and our third-quarter 2007 guidance, excluding the impact of these items, and we'll refer to these results as non-GAAP.

I will now turn the call over to Infinera President and Chief Executive Officer Jagdeep Singh.

Good afternoon, everyone.

It's a pleasure to welcome you to Infinera's first quarterly conference call as a public company.

Joining me today is CFO Duston Williams, who will provide a second quarter financial report and an outlook for Q3 after my remarks.

We're pleased to have achieved the milestone of an initial public offering last month.

We're also gratified that many of you have demonstrated confidence in our long-term vision of Infinera as a leading systems company in the optical network industry.

Infinera has taken a contrarian approach to the market.

We're focused on delivering a highly flexible and highly leveraged solution based on photonic integration to a wide range of service providers.

The second quarter results and our customer win with Cox Communications, both of which we announced today, demonstrated continued steady evidence that the marketplace is embracing our unique approach and proprietary technology.

Cox Communications has selected Infinera to build its national transport network.

The 12,000-mile Cox national build extends from coast to coast and it will enable Cox to offer leading-edge services, including voice, video, data and wireless to residential and commercial customers.

Furthermore, as Duston's commentary will indicate, we achieved an operating profit in the second quarter on an invoice shipments basis, excluding stock compensation charges.

In addition to the IPO and the solid Q2 financial performance, we have had several other recent accomplishments worth noting.

First, we executed well against the strategy we articulate during the IPO road show.

On the sales front, we added five new customers in the quarter, bringing to 31 the total number of customers to whom we have shipped product.

I would emphasize that the win announced today with Cox Communications for a long-haul application is in addition to our previously won engagement with Cox in their metro markets.

Second, on the operations front, we performed well against our metrics of on-time unit shipments, product quality and customer satisfaction.

Third, we announced the availability of a new 40-gigabit Tributary Adapter Module, which enables customers to flexibly and economically deploy 40-gig services over their existing Infinera infrastructure, with no optical reengineering required.

Fourth, we announced our new 19-inch chassis, an important configuration for certain MSOs and Internet content service providers as they deploy Infinera gear.

And finally, we demonstrated a PIC with semiconductor optical amplifiers, or SOAs, thereby demonstrating the integration of an additional discrete element onto the PIC.

This provides further validation of the Moore's Law-like power of photonic integration.

I now want to take a few minutes to describe the long-term opportunity available to Infinera and why our technology and strategy are different and how our solution is winning in the marketplace.

As we all know, there continues to be a significant rise in Internet traffic, which has led service providers to build out the capacity of their networks.

The content, applications and intellectual property driving this traffic growth are well known, including lifestyle and workplace-changing applications like YouTube, MySpace, peer-to-peer and others.

To accommodate the increased traffic, customers tell us they're growing their network bandwidth at a rate of between 70% and 100% per year.

I would note that the network builds we see from our customers are success based.

That is, they're capacity additions in response to demand that carriers actually see, versus the build it and they will come model of several years ago.

So the growth is real, and appears to be sustainable.

Industry analysts estimate the total optical market is about $12 billion.

Infinera is addressing the DWDM, or dense wavelength division multiplexing segment of that market, which we estimate is $3.7 billion in size.

From day one at Infinera, our mission has been to use advanced technology to enable our customers to fundamentally improve the economics of optical transport.

In doing so, we sought to help service providers dramatically improve their business models.

We recognize that carriers would benefit greatly from more frequent digital access to the bits that move through their networks.

With more cost-effective access to those bits, service providers can do many more things with the information and potentially create additional and more profitable streams of revenue.

Until recently, service providers had the choice of deploying traditional DWDM equipment, with frequent digital access, but poor economics, or lower-cost all-optical analog DWDM equipment with insufficient digital access.

Neither of these approaches addresses the key issue of providing carriers with more frequent, low-cost access to the critical OEO conversion process in transmission of data over the network.

While the telecommunications industry was retrenching after the Internet bubble burst in 2001, Infinera was investing more than $300 million to build our vertically integrated model and to develop our competitive advantage to address this customer need for more cost-effective OEO conversions.

The result of this effort is Infinera's proprietary photonic integrated circuit, or PIC, the world's first and only commercially deployed large-scale PIC.

Our development of the PIC reduced over 60 discrete components onto a single pair of optical packages, thus dramatically reducing the cost of the key components in the OEO conversion process.

As a vertically integrated systems company with component systems and manufacturing expertise in house, we don't sell the PIC alone to customers.

Instead, through the collaboration of our systems and components experts, we've designed a value-added system around the PIC called the DTN.

The DTN enables carriers to deploy OEO conversions as frequently as they need them over a Digital Optical Network.

As a result, carriers no longer have to avoid the expensive OEO conversion process, provided they can deploy it more often, more cost effectively, monetize it to their advantage and digitally process traffic as often as they would like.

The advantages of the Infinera Digital Optical Network over traditional analog models or all-optical networks are compelling to customers.

First, there are real and tangible OpEx savings.

This has turned out to be our biggest differentiator in the sales process and our strongest weapon in overcoming the power of incumbent competitors.

Infinera's OpEx benefits help neutralize pricing as the main lever in the sale.

Reduced OpEx is achieved in a number of ways, with enhanced network flexibility, simplified network planning, improved reliability, lower repair costs, scale in 100-gigabit-per-second increments, speed of installation and reduced OEO conversion costs.

And secondly, customers are able to add new revenue streams through access to new cities, speed of provisioning services and service differentiation with new features.

Another key consideration is that the Infinera Digital Optical Network does not require new infrastructure installation by our customers.

Customers can deploy a full nationwide installation of Infinera gear over their existing fiber plant, or they can add our gear on a route-by-route basis.

In addition, our DTN is inherently flexible.

The design of our Digital Optical Network decouples the line or transmission side and the tributary on the services side, allowing Infinera gear to be deployed in any capacity on the services side, without touching the line side.

For example, our recently announced 40-gigabit TAM enables customers to deploy 40-gigabit services over their existing 100-gigabit Infinera network infrastructure.

By contrast, all of our competitors have a transponder-based solution that ties the line side and tributary sides together in the transponder, providing no ability to scale services at higher speeds without impacting the line side.

These advantages resonate with customers, and as a result, Infinera is winning lots of new business.

Our customer strategy has been to focus on bandwidth wholesale carriers, then expand to cable MSOs and Internet content providers, and finally to the RBOCs.

Our international plan is to take current products and expand into Europe and Asia.

Our product plan is reinforce our strength in the core transport space and then expand deeper into the metro edge and Ethernet markets.

Today's Cox announcement reinforces our execution to this plan on the customer front.

We had indicated during our IPO road show that we had won business with three of the top five cable MSOs in North America, and this included metro deployments at Cox.

As previously mentioned, today's announcement applies to a new win with Cox for their long-haul national network.

So we're clearly winning the long-haul metro [core] markets, the segments we focused on most aggressively from our inception.

As a result, we have seen rapid share growth.

Today, we believe Infinera is shipping more 10-gigabit long-haul ports than anyone, equating to over one-third of the world's 10-gigabit ports.

It's also important to note that we're winning in the marketplace against different players in each major segment, with the same Infinera solution, giving us important product leverage across the board.

Based on the compelling economics of Infinera's solution, we displaced incumbent vendors in every deal we won, a powerful statement, given there are inherent advantages of incumbency.

As we look at our markets and focus, we are extremely pleased with our position in the wholesale bandwidth carrier space.

I would note yesterday's announcement that XO Communications has selected Infinera for a major capacity addition to its nationwide network.

This expansion, including 800-gigabit-per-second of additional capacity on a coast-to-coast basis, represents a 200% increase in XO's network capacity.

Second, we have won significant business with the cable MSOs and leading Internet content providers, some of whom may well become the world's largest users of Internet bandwidth in the years to come.

And finally, we view the RBOC market as an important long-term opportunity on par with our other major segments.

We believe that the same economic advantages of Infinera's Digital Optical Networking that apply to our other segments apply to the RBOCs, as well.

However, the RBOC sales process is inherently longer, typically lasting several years, so investors should view this opportunity on a longer-term basis than some of our other initiatives.

We believe we have broken out of the competitive pack by giving customers a compelling economic solution, one that enables a Digital Optical Network and revolutionizes their ability to deliver services in an era when those services are proliferating.

We're pleased with our progress to date and we look forward to keeping you informed in the future.

With that, Duston will now provide his financial report.

Duston?

Thanks, Jagdeep.

It's obviously a privilege to report on the financial operations for Infinera's first quarter as a public company.

Before I go into the Q2 results, because of our unique revenue recognition accounting, I thought it would be useful to cover how we measure the true financial health of the business internally, and how we will be analyzing and reporting some of these results to our external constituents.

I'll then provide an overview of Q2 actual results.

I'll follow that up with an overview of Infinera's business model.

Hopefully, this model will provide you with some direction as to how we see the business progressing and maturing through the next few years.

Finally, I will conclude with our outlook for Q3.

My commentary on future calls will not be as lengthy as today, but I believe it's very important that our investors and potential investors understand the complexities associated with how the revenue recognition rules apply to our business, and how the true economic reality of the business can be analyzed in a straightforward manner.

We also believe it's equally important that you understand how we view our business model and how we think our existing technological leadership will continue to enhance our business model in the years to come.

Infinera's revenue recognition accounting is somewhat unique.

Like many companies, we fall under and are governed by software accounting rules, specifically SOP 97-2.

However, in Infinera's case, to date, we have not established VSOE, vendor-specific objective evidence, or the fair value for the multiple elements of a sale, and are therefore required to ratably recognize our entire revenue base over what is currently a 1.2-year period.

It is important to note that it is solely the lack of VSOE for the multiple elements that is causing the requirement for ratable revenue recognition, rather than any guarantee of future product enhancements.

Our cost of sales are also recognized ratably over the same period.

However, our business model does result in selling some of our common equipment components at a loss.

These losses cannot be deferred and are recognized in the current period.

In addition, all of our operating expenses are recognized in the period that they are incurred.

The result of SOP 97-2 revenue recognition accounting is that our GAAP results may not be the best measurement of the performance of the business during our most recent accounting period.

To mitigate this, Infinera analyzes its business based on invoice shipment results.

Wall Street analysts have also modeled our business based on invoice shipments.

We define an invoice shipment by the following criteria.

A PO has been received from the customer, the product has been shipped and accepted by the customer, the customer has been invoiced and has paid or will pay within our normal payment terms.

For ease of calculation purposes, we calculate invoice shipments by simply taking the current period GAAP revenue, and adding back the period-over-period increase in deferred revenue balance.

We calculate the invoice shipment cost of sales in a similar manner.

We believe that reporting the invoice shipment results will give investors a more complete view into the true economic reality of Infinera's business, and will allow investors to analyze the business the same way as management does.

Our current Q2 earnings press release provides further detail of the invoice shipment methodology, and explains how to reconcile these results back to our GAAP results.

The following analysis of our Q2 results is based on invoice shipments and excludes stock-based compensation and warrant revaluation expense.

Please see the GAAP to non-GAAP invoice shipment reconciliation which is attached as an exhibit to today's earnings press release for a reconciliation of these results to our GAAP results.

Invoice shipments in Q2 totaled $69 million, versus $66.7 million in Q1.

International sales were 16% of invoice shipments in Q2, versus 11% in Q1.

In Q2, we had two 10% or greater customers, with the largest customer accounting for 48% of revenue on an invoice shipment basis.

In Q1, our largest customer accounted for 57% of invoice shipments.

Turning to gross margins, they were 37%, versus 35% in Q1.

A favorable customer mix and product mix helped account of the quarter-over-quarter increase in gross margins.

Operating expenses for the quarter were $23.8 million, versus $28.7 million in Q1.

A majority of this variance related to a more favorable commission structure and lower R&D project cost.

Some of these project-related expenses will roll into Q3.

Operating income for Q2 was $1.8 million, versus a loss of $5.4 million in Q1, reflecting better-than-expected performance on revenue, gross margin and operating expenses.

Other income and expense for the quarter was a favorable $0.9 million, which included a gain on the sale of assets of $1.1 million.

Net income for the quarter was $2.7 million, versus a loss of $5.4 million in Q1.

Turning to the balance sheet, we were pleased with the overall performance for the quarter.

Cash and cash equivalents ended the quarter at $197.4 million.

We had net proceeds after expenses from the initial public offering of $190.2 million.

We used $0.8 million of cash from operations, paid off $29.3 million in debt and had capital expenditures of $3.6 million.

DSOs were 36 days, inventory days were three, and accounts payable days came in at 51 days.

Over time, we believe that our business model will reflect the true value and advantage that can be derived from photonic integration.

Our business model has already begun to prove itself in a very short period of time, and we expect continued improvement in the years to come.

Our revenue growth over the last two years has been outstanding.

We expect to see continued revenue growth into the future, but obviously at somewhat lower growth rates.

After 2007, and for the foreseeable future, we could see annual growth rates in the 25% range.

I believe it's important to point out, although we see good revenue growth continuing on a year-over-year basis, the quarter-over-quarter revenue growth may be impacted by several factors, including the timing of large customer deployments of Infinera gear, acquisition of new customers and general market conditions.

Therefore, the quarter-over-quarter revenue growth could be somewhat uneven and growth may not always occur in a linear manner.

From an invoice shipment gross margin point of view, we expect to see continued improvement in the years to come, but with some potential volatility on a quarter-over-quarter basis due to customer and product mix.

We believe Infinera is in a unique position due to the tremendous value proposition of photonic integration and our vertical integration to not only enhance our gross margins on a go-forward basis, but also to continue to help drive our customers' business models through greater levels of efficiency and profitability.

As we move into 2008, we could see invoice shipment gross margins in the high 30% to low 40% range, and then migrating to 45%-plus.

Based on some early data points and our continued ability to further leverage our photonic integration advantages, we believe that we can drive margins in the 50% range over time.

We will continue to make significant investments in the business for years to come.

Operating expenses should average 35% of revenues, beginning in 2009.

Regarding the operating expenses, it's important to note that there are projects outside of our current product roadmap that have a high potential ROI that are currently not funded.

If and when we are fortunate enough to exceed our revenue and margin projections, we will make decisions at that point in time as to the use of these incremental margin dollars, that being either to fund additional projects, flow it to the bottom line or some combination of both.

These decisions will obviously be based on what we believe will provide the maximum shareholder value over the long term.

The summary [of announced] above would suggest a long-term operating model of 50% gross margins, expenses equaling 35% of revenues, with operating profit of 15%.

As we look into Q3, the demand for Infinera products continues to be strong.

As Jagdeep noted in his remarks, we are very pleased with the overall customer traction and have plans for several new deployments during the quarter, including some of our new products developed specifically for the metro environment.

As you might expect, the timing of the equipment acceptance and invoicing of these new customers and products is sometimes difficult to gauge.

While we are assuming some of these new engagements will be invoiced during the third quarter, while others are expected to occur in Q4.

These new shipments will carry a higher percentage of common equipment, and therefore have some limited negative impact on our margins.

With that as background, I would like to offer the following guidance for Q3, based on invoice shipments, which excludes any stock-based compensation expense, invoice shipment revenue of $68 million to $72 million, gross margins of 35% to 37%, operating expenses of $27 million to $28 million, again, excluding stock-based compensation, net result of a $1.5 million loss to breakeven.

Based on estimated average diluted weight of shares outstanding of $92 million, this would lead to an EPS between a loss of $0.02 and breakeven.

Operator, would you now please open the call up for questions?

Thanks.

(OPERATOR INSTRUCTIONS)

The first question is --

We have not heard the question.

The first question is from Brant Thompson.

Please state your affiliation.

Goldman Sachs.

I was wondering if you could just talk a little bit about the OpEx ramp that you're seeing into this next quarter.

And you talked about kind of how that was going to trend, but if you could give us an idea of how much the OpEx this quarter being low is going into next quarter, so we could just have an idea of what the kind of normalized, I guess, run rate there is, number one.

And number two, Jagdeep, could you just talk a little bit about the metro products that you're looking at shipping into the next quarter, and how you see your progress in that space progressing?

Thanks very much.

Yes, Brant, Duston.

Let me take the first part of the question there on expenses.

There's some roll into Q3, but it's really not a significant piece of it.

I think if you look at the expenses quarter-over-quarter and based on our guidance for Q3, there's probably three or four things in there worth mentioning.

We are hiring additional sales personnel in the quarter, and that's proven out to be pretty successful from a revenue generation perspective.

So we're pretty comfortable with those investments.

We will also see some more lab trial gear go into the field into customer sites, for demos and trials in the quarter.

And then also we are planning on a fair amount of R&D headcount adds in the quarter, with some other project-related expenses.

So it's really those three or four things that account for a majority of the increase.

There's other SG&A-related stuff, but that's really not a real significant piece of the total.

And regarding the metro question, as you know, the DTN today competes quite effectively in the metro space, and where we compete, as you know, is in networks where there's a sufficient amount of bandwidth density.

What we have announced is a smaller chassis, a 19-inch form factor chassis, which many of our customers in the metro, for example, our cable MSO customers, tell us is really important for their applications.

And so what we continue to do is additional product development like that that make the product a better and better fit for lower-end applications in the metro space.

Great, thank you.

Marcus Kupferschmidt, please state your affiliation.

Lehman Brothers.

Good afternoon, guys.

Hi, Marcus.

Two questions.

I guess, [understand a bit] more OpEx, and then I'd like to talk about the metro product.

In terms of this new metro product, so just to be clear, is this a skinnied down version of today's DTN, or are you adding more functionality?

Is this somewhat of an MSPP-like box, right now, and then I'll follow-up with my OpEx question.

Yes, we've said on the metro [front] is the product we've announced is in fact the 19-inch form-factor version of the DTN, and to do that you need to also shrink a few of the line cards to get them down to half height and so on.

Over time, we expect to announce additional functionality in the metro side.

What we said on the IPO road show is basically where we play is where there's a sufficient amount of bandwidth density requirement.

What we intend to do with our product line over time is lower the threshold of bandwidth at which the product proves in economically.

The 19-inch chassis is a key step in that direction.

Many of our customers consider that a hard requirement.

Over time, additional functionality will be added.

It's not intended to be an MSPP-type box in any way.

We play in the WDM and bandwidth management space, but we don't consider ourselves to be in the MSPP space.

Right, and before I ask about OpEx, can you just quickly explain to us why there's more common cards, and why it's a lower-margin sale than a regular DTN?

No, Marcus, my comment there was it's just the same on any deployment.

You first start out with more common as a percent of what you'll end up with and then over time we'd layer in line cards and additional TAM, so no different there than other deployments.

Right, and then in terms of the OpEx, when you think about your engineering spending today, how much of that is on new products, new enhancements, versus the basic DTN box as it is today and your enhancements to that?

And then, thinking about OpEx, or sales and marketing, how is that going to scale with your revenue growth, since it seems to be lumpy here over the last couple of quarters, with not a lot of correlation?

Sales and marketing, obviously commissions continued to ramp up with revenues, obviously, as we progress through the year.

Headcount, what we've done there is, as I say, when we've added sales headcount, there's been very good success trailing with that regarding revenue increases.

So what we've done there is as we're exceeded -- similar to my comments, as we've exceeded our expectations from a revenue and margin perspective, we kind of look at the business and say what do we want to do with those incremental margin dollars and in Q3, for instance, some we flowed to the bottom line and we put back a fair amount invested into the business, both into sales and R&D.

A lot of the R&D dollars are pointed to add-ons to the enhancements, I guess, of the initial what you'd call the DTN system as of today.

Thanks.

Hasan Imam, please state your affiliation.

Yes, Thomas Weisel Partners.

Hello, guys, and congrats on your first quarter.

Thanks, Hasan.

On the OpEx, obviously there's some volatility there.

Trying to get to the bottom of that.

You mentioned two points there, or two sources, of the favorable move.

One was commission structure and the second was R&D project related.

Can you shed some light on the commission structure, is that a transient phenomena or something longer term?

And then this $27 million, $28 million guidance, should we think of this as kind of a steady state?

Sure, let me take the commission structure, without trying to get too detailed here.

The commission plan for 2006, we've since changed it, but it was on a different structure and we had a very, very successful year in 2006 and -- far ahead of our expectations.

Therefore, obviously, commissions were higher than we thought.

But the way the plan was structured in 2006, it was based on what was booked at the end of 2006, but it wasn't paid until those bookings actually shipped in Q1 of 2007, and those were at the accelerated rates of 2006.

We won't see that issue going forward, because we've changed the structure of the plan, so that's a one-off thing that's kind of some historical stuff that we had a bigger pop in Q1 and then similar dollars were paid at a lower rate, quite honestly, in Q2.

Expenses, again, I would look at it going forward more based how the business does, and again, if we're fortunate enough to exceed our estimates, we'll take a look internally to see what the right thing for the business and shareholders are over the long term.

And in some cases, the payback of putting it back into the business is a pretty appealing scenario.

(Inaudible) question.

Thank you for that.

Your revenue guidance range of $68 million to $72 million and gross margin range of 35 to 37, can we think of the range as related to these new customer deployments?

In other words, if we move to the higher end of the range, does that mean the lower end of the gross margin guidance?

I wouldn't necessarily correlate those to each other.

So no, I don't think the high or the low really has nothing necessarily to do with each other.

There's been argument out there that your product costs may be less compelling downstream in metro than long haul, and therefore you would first make your footprint in long-haul and then expand into metro.

Cox appears to have gone in reverse.

Maybe, Jagdeep, you could shed some light on why this has been the case at this particular customer and what it means for your ongoing trials?

Thank you.

Sure, so, again, the entire DTN, as we've pointed out in the road show, has been proven in a full range of networks, long haul and metro.

Really the only common characteristic of all these networks is that they have a sufficiently high amount of bandwidth that they needed to transport from point A to point B.

If the bandwidth the demands are great enough, then the Infinera approach proves in over any approach that we're familiar with.

So, as we pointed out again during the road show, and DTN currently has deployments not only in the long haul, but also in the metro.

We specifically said, for example, that we have three of the top five cable MSOs in North America as customers, and as you might expect, many of those customers are deploying us in the metro applications.

The Cox application is interesting, because many of those cable MSOs that initially or in the past were focused on delivering cable service within their metro markets have also recently chosen to think about deploying long-haul networks to connect their various metro markets and to deliver business services across that long haul.

Many of them in the past were leasing circuits from other carriers and have chosen to build their own networks.

And what the Cox win represents really is that somebody is building out a brand-new greenfield network and has a choice of going with any vendor in the industry, has concluded that the Infinera approach is more compelling than other alternatives they have, and that's why we think it's a very significant announcement, above and beyond the fact that they, as we pointed out, already have Infinera gear deployed in the metro.

Next question?

Thank you.

Ajay Diwan, please state your affiliation?

It's J&W Seligman.

I just read through the Cox press release and I'm trying to get a sense of how big the win is, and it talks about a 12,000-mile national network build.

So, I mean, is there any way for you to scale this in a broad range?

And, also, can you tell us the timing of this?

How long will it be for them to roll this out, so size and the revenue over how long?

Let me go ahead and take that.

So the size of the build is pretty much what's specified in the press release.

It is a 12,000-mile national build out, so it's basically -- it's an industrial strength long-haul network, not fundamentally different from other long-haul networks in North America.

It's a big network build.

In terms of timing, we really haven't said anything in terms of what the timing of the deployment is on that front.

Is there any kind of industry benchmarks about, I don't know, cost per mile or something?

And then, I think you said this was a greenfield build, so they have obviously an existing network with an incumbent supplier, but this has got nothing to do with them touching that piece of the network.

It's just completely new build out.

Is that right?

That's right.

What the press release said, as you saw, is that this is a new long-haul network that they're building out to deploy a range of new services.

What we've said before regarding timing of a typical new long-haul build is the Infinera product is capable of being rolled out on a national basis in North America, and on the order of a couple of quarters, which we think is faster than other approaches.

And of course, once the network's rolled out, then there's ongoing bandwidth additions that are deployed in response to traffic growth in the network.

And there's no reason to believe this network would be fundamentally different from that.

Got it.

Well, congratulations.

Thank you.

Thanks, Ajay

(OPERATOR INSTRUCTIONS]

Ehud Gelblum, please state your affiliation.

Hi, thank you.

It's JPMorgan.

Hi, Jagdeep.

Hey, Duston.

A couple of questions if I could.

First of all, just following up on Ajay's comments about the Cox, that was really impressive.

My guess, just looking at the size of the network is it's somewhere in the $20 million, $25 million range, you're saying over a couple of quarters.

So it sounds pretty sizable.

I want to see if that kind of correlates, because I'm looking at your guidance for next quarter, and when are we going to start seeing that come into the invoice revenues?

I would have thought then the guidance may have been a little bit higher on a sequential basis.

But, related to that, can you give us a sense of the lumpiness of your revenue base right now, specifically Level 3?

Because if I try to put my guess, $20 million, $25 million, of the Cox deal into your next couple of quarters, does that mean that Level 3 may be lumped down a little bit?

And therefore -- I'm just trying to correlate all that with a gross margin that seems to be going down when perhaps I would have thought that when Level 3 goes down, gross margin goes up.

So if you can just talk about the lumpiness of your revenue base, specifically the lumpiness of Level 3 over the next couple quarters and how Cox kind of fits in over the next couple of quarters into that, that would be very helpful.

Yes, Ehud, obviously we're not going to comment not the size of the Cox deal there.

We will ship some of that equipment in the current quarter here, and then some of it just depends on not only for them but other customers, quite honestly, is you've got acceptance periods, especially with new customers.

So the exact, as I say, timing of these, when they become invoice shipments, there is some movement there, whether it's Q3 or Q4.

Regarding Level 3, I think we've been pretty clear all along that we see them coming down as a percent of the business going forward into the second half of '07.

It did come down in Q2 and we expect it to come down somewhat going forward, also.

Again, the margin play and trying to correlate it to customers and stuff like that, again, when you have new deployments, you're going to have common equipment shipping at lower margins and, as you know, as we add onto those deployments with DLMs and TAMs, higher margins over time.

So I wouldn't necessarily try to correlate all that stuff.

But with new customer and fairly large potential customers, then you're going to have common equipment at lower margins.

So, we're clear to assume that you're implicit in your guidance and your expectations for at least next quarter, and probably the quarter after that, that Level 3 goes south of about 48% of revenue, roughly stays down there?

Yes, we've said that they'll tail off to some degree in the second half of the year.

Where do you expect them to be next year?

We'll have to see once we end '07.

Okay, great, but wherever you end this year -- last question -- you just expect it not to go above that.

As you backed all additional revenue, you think that Level 3 will stay at or below, going forward, of your run rate at the end of the year on a percent basis?

Most likely, but we'll have to see how it plays out at the end of the year.

Thanks so much.

Okay.

Marcus Kupferschmidt, please state your affiliation.

Lehman Brothers.

My follow-up question was, we're talking about this new product starting to drive some business, can you give us a sense of what kid of contribution you could see in the back half or in 3Q here for this new metro product?

No, we haven't started breaking out --.

We've been asked in the past to break out the metro portion of our business.

We'll do that at some point in time, but I think we've got to get a little more consistent pattern to the revenue and metro deployments before we start doing that, because they're going to go up some quarters.

It's going to go down some quarters, and at some point, we'll do that.

But we're probably not ready to do that yet.

Just trying to see if there's anything to figure out how much pressure it's putting on the gross margin, since you talked to us about that as being a lever here in the next quarter or two.

Yes, I'd just look at it as new deployments with the common equipment from a margin perspective.

I wouldn't look into anything more than that.

But similar to any other new win you'd get from any other new customer.

Correct.

Okay.

Thanks.

Okay.

Okay, with that, let me go ahead and wrap up here.

So, in closing, we're excited about the outstanding opportunities for growth in the digital optical network space for Infinera.

Our team is focused on continuing to deliver a strong financial performance in the years ahead as we establish ourselves as a leading systems company through our differentiated approach to the markets we serve.

I thank you all for joining us today, and we look forward to reporting our progress in our next call.