Ciena Corp (CIEN) 2017 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Ciena Corporation Q3 2017 Earnings Conference Call.

(Operator Instructions) As a reminder, this conference call may be recorded.

I would now like to introduce your host for today's conference, Mr. Gregg Lampf, Vice President of Investor Relations.

You may proceed.

Thank you, Crystal.

Good morning, and welcome to Ciena's 2017 third quarter review.

With me today is Gary Smith, President and CEO; and Jim Moylan, CFO.

Steve Alexander, our CTO, will join us for the Q&A portion of the call.

This morning's press release is available on National Business Wire and ciena.com.

We will also post to the Investors section of ciena.com an accompanying investor presentation, including certain highlighted items from the quarter being discussed today.

In our prepared remarks, Gary will touch on highlights from our fiscal third quarter and discuss management's view on the market.

Jim will then provide detail on our financial performance as well as guidance regarding expected future results.

(Operator Instructions)

Before turning the call over to Gary, I'll remind you that during this call, we will be making certain forward-looking statements.

Such statements are based on current expectations, forecasts and assumptions regarding the company that include risks and uncertainties that could cause actual results to differ materially from the statements discussed today.

These statements should be viewed in the context of risk factors detailed in the most recent SEC filing.

Our 10-Q is required to be filed with the SEC by September 7 and we expect to file by that date.

Ciena assumes no obligation to update the information discussed in this conference call, whether as a result of new information, future events or otherwise.

Today's discussion includes certain adjusted or non-GAAP measures of Ciena's results of operations.

A detailed reconciliation of these non-GAAP measures to our GAAP results is included in today's press release.

This call is being recorded and will be available for replay from the Investors section of our website.

Gary?

Thanks, Gregg, and welcome, everyone.

We posted solid results for our fiscal third quarter this morning.

Once again, we've outpaced the competition and delivered differentiated financial performance.

Quarterly revenue was up 9% year-over-year, adjusted net income and EPS grew nearly 30% from the same 9-month period last year and we delivered 14% adjusted operating margin in the quarter.

Importantly, I think this quarter's performance demonstrates that our strong competitive position, underpinned by our diversification and innovation, continues to enable us to grow faster than the market and our peers, capture share and increase our industry leadership.

A couple of quick highlights from the quarter that I think really illustrate these points.

With respect to geographic regions, business with our largest service provider customers remains very healthy, particularly in North America.

We had 2 10% customers in Q3, AT&T and Verizon.

And we expect to finish the year with those customers absolutely in line with our expectations that we had for the year.

Asia Pacific has been a solid growth vector for our business and we expect that to continue.

In Q3, we added another one of India's top 3 service providers as a customer with a new 100 gig metro rollout, boosting our confidence in the sustainability of this market as a long-term opportunity for Ciena.

And finally, EMEA and CALA are relatively stable and performing in line with our expectations for the year.

With respect to our portfolio, we continue to see evidence that our competitors are struggling as we force the pace of innovation.

Many of them simply do not have the capacity, scale and channels to market to compete.

As a result, they are demonstrating behaviors that are not sustainable and their solutions portfolios are rapidly falling behind.

By comparison, our significant investment capacity enables us to be ultra competitive with time to market of best-of-breed platforms and capabilities.

A great example of this is our WaveLogic Ai chipset, which is now generally available and has begun to ship.

And we believe that we have significant opportunity to capture additional share globally as it comes to market in its various forms over the next few months.

In fact, we have already taken initial orders for Waveserver Ai from 2 major customers in Q3, 1 in North America and 1 in CALA, further demonstrating and strengthening our #1 position in all DCI categories tracked by analyst firm, Dell'Oro.

It's becoming a very clear that the vendor selections of our customers and the weakening financial results of our peers continue to rationalize our fragmented competitive landscape towards a smaller number of successful strategic players that truly have global scale with obvious winners and losers emerging.

Before I turn it over to Jim, I'd like to touch on our guidance for the fourth quarter and provide some clarification and color around the variables that are impacting our near-term outlook.

During Q3, we began to see a couple of small variables in the North American marketplace.

While these factors have no bearing on the overall health of our business and the underlying demand drivers, when taken together, they impact our near-term outlook.

First, we now believe that some U.S. government-related business will not materialize in our fiscal second half as we had expected.

We believe this is largely due to the same uncertainty that several other companies in our broader space have recently referenced.

Second and to a lesser degree, we've also seen some softness in orders from a handful of regional service providers in North America really related to very customer-specific challenges, including things like M&A and integration activities.

These recent spending dynamics may have taken slightly longer to impact us than others because of our global scale and broadly diversified business.

That said, we believe these are likely to be relatively short term in duration.

However, even with these slightly lowered expectations for Q4, we expect to continue to significantly outpace the competition in the market, with North America as a strong contributor.

In fact, we've grown revenue in North America by nearly 6% year-to-date versus the first 9 months of fiscal 2016.

And consequently, we've increased our market share in North America this year by approximately 5% according to Dell'Oro.

In closing, I'd like to make one additional, but I think important point, about our performance relative to the market.

Based on the latest available data through the first half of calendar 2017, we believe that the overall market is growing in the single low digits.

For example, Dell'Oro recently stated that the optical transport market is currently experiencing negative growth, falling 3% year-over-year in calendar Q2 '17.

The analyst noted that all regions declined except for Asia Pacific, which is consistent with what we're seeing in the marketplace.

And in North America alone, where we remain very strong, the same analyst firm has indicated a 9.6% decline on a rolling 4-quarter basis, which compares and contrasts to our year-to-date growth of 6% that I mentioned earlier.

Against that backdrop, our Q3 performance and really our expected performance for the year proves that our strategy to diversify our business and out-innovate the competition is clearly working.

Our position is the only industry player to participate in all key areas of the market.

Systems, software and components sets us apart from others and gives us a clear edge in this evolving environment.

As a result, we are able to deliver differentiated financial results and very clearly are taking market share away from our competitors.

Lastly, I'd like to extend our concern and support to those affected by Hurricane Harvey.

It is our top priority to ensure that our employees, customers and partners are safe.

We are in constant contact with our customers and are ready to mobilize our service teams on the ground to assist with every level of technical support they may require to maintain their network operations at this critical time.

With that, I'll hand over to Jim.

Thank you, Gary, and good morning, everyone.

Q3 was an excellent quarter on both the top and bottom lines.

As Gary mentioned, our successful execution is a direct result of our strategy to diversify our business and maintain innovation leadership, both of which are advancing our competitive position.

There are several highlights from the third quarter that I'd like to mention.

To start, we continue to win in DCI.

Q3 was our largest quarter ever for Waveserver, keeping us well on track to deliver over $100 million in revenue this year for the platform.

We added 14 new Waveserver customers in the quarter, bringing our total to 56.

We also had a strong quarter in Packet Networking, which is up nearly 25% sequentially and where we added 6 new customers during the quarter.

In subsea, we secured a new turnkey cable system build in the Caribbean with our partner TE SubCom during the quarter.

This is in addition to 2 earlier wins through this partnership.

Those were for trans-Pacific submarine cable systems, one of which also involves 2 global web-scale companies.

On the software side, we now have 25 Blue Planet customer engagements, 4 of which have converted into live deployments and 8 that are in active trials.

And finally, our component vendor partnerships are progressing nicely.

We recently began shipping initial samples for lab evaluation and customer field trials.

That progress enabled us to deliver the solid Q3 financial results we posted this morning.

Total revenue was $729 million.

Notably, this reflects a nearly 10% sequential increase in North America.

It's clear that our diversification is helping to offset the market conditions that Gary mentioned.

And our overall order flow in Q3 was about equal to revenue.

Q3's adjusted gross margin was 45.5%.

Adjusted operating expense was $229 million.

We achieved $102.5 million in adjusted operating profit in Q3 and we delivered 14.1% adjusted operating margin.

We also continued both to delever and to strengthen our balance sheet and took additional steps to reduce the dilutive impact of our convertible notes.

During the quarter, we repaid the remaining $185 million in outstanding principal on our 2017 converts in cash.

We also completed a successful exchange offer that will give us settlement flexibility upon the conversion of our 2018 convertible notes.

We intend to settle the face amount of the new notes in cash in order to reduce our shares outstanding.

The revised calculation for EPS is a bit complicated, so please refer to the IR section of our website, where you will see how to perform the calculation for your modeling purposes.

For example, for stock prices above about $20 per share, you should not remove the entire 14.3 million shares of the underlying notes from the calculation.

Also interest expense remains in the calculation of diluted EPS until the notes mature.

In Q3, we generated $50 million in cash from operations.

We ended the quarter with approximately $850 million in cash and investments reflecting the repayment of the 2017 notes.

And finally, adjusted earnings per share in the quarter was $0.51.

On that point, we've become much more profitable, having delivered over 20% CAGR on our adjusted EPS over the past 5 years.

It's clear now that the financial community is looking more closely at our EPS for valuation purposes, so you will begin more about our expectations for this performance metric going forward.

I'll now turn to guidance for our fiscal fourth quarter.

As noted in the press release, we expect revenue in Q4 to be in the range of $720 million to $750 million.

We expect Q4's adjusted gross margin to be in our target mid-40s percentage range.

And we expect Q4 adjusted operating expense to be approximately $240 million.

In closing, I want to emphasize that our view of the fundamental demand drivers for our business are unchanged and that we are well positioned to continue gaining market share.

As Gary mentioned, we have recently seen a decline in spending on optical gear in the North American market according to one industry analyst.

However, we do not expect that decline in spending to continue.

The underlying growth in bandwidth demand continues and the drivers for that demand remain in place.

Things like movement to the cloud, the Internet of Things and with fiber deep and 5G on the way, we believe that growth in the North American market will resume and we are well positioned to benefit when it does.

And remember, we are experiencing growth in North America even with current market conditions.

We have a solid strategy that will drive short- and long-term profitable growth and we will maintain our focus on revenue diversification and innovation leadership.

In a global market that we expect to grow in the low single digits, we intend to continue outperforming our competition by growing faster than the market and improving our profitability.

We have a strong track record of delivering this level of performance as you can see in our consistently improving financial results over the last several years.

Operator, we'll now open the line for questions.

(Operator Instructions) Our first question comes from George Notter from Jefferies.

I guess I wanted to start by kind of digging into the guidance a little bit more on the top line for the October quarter.

So are you guys seeing any sort of demand overhang associated with the pending shipments or current shipment of WaveLogic Ai-based products?

I guess I'm wondering if there's an eval cycle that customers are going to go through that can potentially kind of slow down your opportunity in the near term?

Is that part of the narrative at all in terms of the guidance?

George, this is Gary.

I don't think so.

Well, I can point to some examples of that.

But overall, I think the release has been on schedule and people are understanding of the capabilities of it.

So I think whatever overhang there was, it's been long baked into our expectations.

So I really wouldn't point to that.

I think this is very specific to a number of sort of small variables in North America.

We've been able to kind of drive through these things during the year because of the diversified nature of our business.

But when you get 3 or 4 of them at the same time, we're clearly not immune to it.

So as Jim said, as we look to next year, we think it's going to be a good year.

These things -- these dynamics, I think, may affect us for a couple of quarters here.

But overall, I think we feel pretty good around next year.

And your point around the Ai, we're just beginning that cycle.

So that clearly will play into a good 2018.

Got it.

And then I guess as part of that, I'm just curious about the timing on shipments for Ai in the 6500 platform.

Can you just talk exactly when you expect to ship that?

And then also how long would it take for that to start to kind of show up in the financials and get traction with customers?

I think you will -- whilst we'll shipping a little bit towards the end of this quarter, mainly it will be Q1 and into Q2.

Q1 is always a challenge for us, given a straddling the end of the year here.

But I would think the simple answer to your question, George, would be Q2 next year.

Got it.

And then do you expect eval cycles to be significant then or not?

Not particularly.

I would think, given our history and given the well-documented engagement with these customers, I would not expect it to be too challenging.

We're already beginning to ship some of it at the end of this quarter.

You've got Q1, which is generally quiet.

But I would think as we get into Q2, that would start to ramp.

So I'm not too concerned around the eval cycles on this.

Our next question comes from Simon Leopold from Raymond James.

First, just if you could just give us a quick clarification on the 2, 10% customers you told us, AT&T and Verizon.

Could we get the contribution percentages from each?

Yes, I'll get that.

Okay.

While you're looking, I'll shift over to what I wanted to dig into is the trend within Packet Networking segment.

It was strong this quarter, which was a nice, pleasant surprise.

Just want to get an understanding of the applications and how we should think about that trending.

My guess is that it's both cable and telco-type customers.

To what degree are the applications related to enterprise versus mobile backhaul?

And when we think about the longer-term trend for Packet Networking, should we think of initiatives like FirstNet or 5G as contributing to improving revenue for Packet Networking over, let's say, the 2018 time frame?

If we could get a little bit more color as to how they think about the drivers for that particular business.

Yes, Simon, it's Steve Alexander here.

So a simple answer is it's a lot of everything you just mentioned, right?

The various marketplaces are all interested in higher speeds at the edge of the network.

I would also tell you to take a look at the success we've been having in India in particular, right?

That's a massive build-out going on in that country.

And it's really the first time they been able to put high-speed capacity out in the packet format.

So there's drivers that are really across all of the industry segments.

And how do you see that trending in October?

Is it sequentially down after such a strong July quarter?

Is that kind of implied in this guidance?

Simon, I would tell you, look, there's always ebbs and flows in the deployments, right?

As you go through the next couple of quarters, you'll see some of the build-outs continue on.

There's others that are just starting.

And there's some that are going to trail down.

And I would just tell you it's kind of normal business as usual.

The overall trend (inaudible) in India is up.

And the trend towards higher capacities at the edge of pretty much all networks, right, you've got the phenomenon of if you were Fast Ethernet, you're going to Gigabit Ethernet.

If you were Gigabit Ethernet, you're going to 10.

If you were 10, you're going to 100.

And that just drives demand across all the market segments.

And the Packet Networking portfolio was built in expectation of those trends.

And should we think of FirstNet as a catalyst for this business?

I wouldn't tie it specifically to FirstNet.

I think all the carriers that are out there that are participating in this explosion in terms of bandwidth demand are trying to build kind of consistent common architectures.

And I would look at FirstNet as a use case of the kind of architectures that are being put in place.

But to your point earlier, it's driving from the enterprise demand, it's driving wireless, it's driving from people who are putting in place deep fiber, right, in the MSO side.

It's coming from all segments.

Thank you, Simon.

And I'll give you your answer now.

AT&T was about $121 million or 16.6%.

Verizon was about $83 million or 11.4%.

Our next question comes from Rod Hall from JPMorgan.

I wanted to start off, I guess, and see if you could talk a little bit or maybe quantify the government impact in the guidance.

It sounds like it's most of the guidance (inaudible).

But how close to consensus would you have been without this government change?

And then can you also talk a little bit about how you expect that to move after the October quarter?

Do you think that it's just delayed?

Do you think that it will materialize in the next 2 to 3 quarters?

And then I have another follow-up to that.

Rod, this is Gary.

Yes, I know a lot of folks have talked about the obvious sort of government constraints over the last couple of quarters.

We, as you know, have a broad relationship with multiple opportunities in the government, particularly related to Fed and DoD, that kind of stuff.

And we did expect -- even seeing that earlier in the year, we did expect an uptick in the second half, which is typical of that.

Our view now is that, that is probably not going to happen.

And that's really reflected in the guidance.

I'll give you a sort of a perspective in terms of proportionality of it.

If you take the sort of consensus have been $770 million and the midpoint of our current guide at about the $735 million, obviously the $35 million sort of delta, I would say more than half of that is related to specific expectations around government business that we do not think will materialize.

These are known projects.

Some of them are established networks.

We think this is a budgeting issue that is affecting widespread government spending.

I would think, Rod, that as we get to next year, they will start to free up.

They have to be resolved at some point.

Okay, appreciate the color.

And so when you say more than half, should we be thinking like 60-40 split between that and these regional volumes you talked about.

Yes, I mean, just to rough out, if you -- and again, there's a lot of variables to it.

But if you roughed it out at sort of $25 million of the $35 million, you'd be pretty close to it.

And then the other thing I would say is even in the sort of regional providers, which constitute the other sort of $10 million, so small number on $700-and-something million.

But again, some of that can also be the cascading effect of the government withholding sort of programs as well.

Okay.

And then I wanted to also -- software is still a pretty small percentage of the total revenue.

But I'm just curious what you guys think you could achieve next year in terms of software percent of total revenue.

I mean, it's like 2.5% now.

Is it possible that could get to the mid-single digits?

Or where do you think that ends up next year?

Well, we're just now developing our 2018 plan, Rod.

And we think that software will grow for us.

And remember, we have a couple of different vectors there.

We have our base network management system, which we think will show nice program progress next year.

We have the Blue Planet set of applications on top of our new control system.

And we have software-related services.

And I'd expect it will grow.

But as I say, we haven't done our 2018 plan as we sit here.

Our next question comes from Stanley Kovler from Citi Research.

My first question, I just wanted to ask about the overall market growth.

It seems like before, you were thinking that the overall market could grow more than mid-single digits.

And now we're hearing that at least some aspects of the market are growing in the low single digits.

And I guess it's safe to assume that you still expect to gain share there.

Is that something that from a market perspective, we should expect to continue into 2018?

And then I have a follow-up.

Yes, Stan, we started out the year, we thought, around the mid.

We've lowered that to the single.

I think most of the industry analysts are consistent with that.

And in fact, if you take a basket of the industry analysts, they're all saying that North America will decline about 10% year-to-date.

But then as I said, we're still taking share and growing in that.

Asia-Pacific is the one sort of standout to that.

And principally, that's been China.

And in our view, India has been the standout country for that.

We think that next year, a little early to tell.

But we think it should be improved next year.

But I would say that this environment is also an opportunity for us to continue to take share and really to push down on the smaller subscale competitors that are struggling.

And I think whilst it's been challenging in its own right, it really is an opportunity for us to change the industry structure as we go forward.

And just a follow-up, you have some very interesting updates during the quarter.

Just to refresh people about Coherent technology, we hopefully get your thoughts around how much you can push that to down market and to shorter links?

And what do you think the time frame would be for that type of development?

Sure.

So again, this is Steve Alexander.

So clearly, Coherent has a lot of places it can play very effectively, right?

It's pretty much the best way to do what you might call maximum capacity, maximum reach sorts of applications and such.

And we've talked about it in the past.

We're doing the modem work now with our partners.

And following on to that, there's initiatives that focus on shorter reach data center interconnects, 400 gigs and above, right?

So we do believe that it has widely applicable market.

And we're going to continue to basically drive the innovation curves that way.

Our next question comes from Patrick Newton from Stifel.

I guess, Gary, following up on kind of your prior comment on India, APAC.

I think it's up greater than 50% year-to-date.

You're not pointing to APAC for any of the softness in the 4Q guide.

I'm curious, given that you inked another top 3 customer, for 100 gig, how should we think about the growth potential of APAC and specifically India in the next several quarters?

Yes.

I mean, I think, Patrick, you're right.

We're not pointing to India in any way, shape or form in the guide for Q4, except that obviously we're going to continue to grow in Q4.

I would remind folks of that.

And in fact, it will be our best-ever revenue performance in a quarter on the midpoint of the guide.

India, as you said, is up about 80% for the first 9 months of the year versus '16.

Obviously, you're going to have some fluctuation quarter-to-quarter.

It was down slightly in Q3.

We expect it to be up again in Q4, so as a positive part of the guide.

And as we turn the year, particularly securing the third carrier there, we've got all top 3 carriers now plus the government, plus some of the smaller regional carriers in India, I think we're incredibly well placed going forward.

And I think that talks to the sustainability, I think, of the opportunity both in terms of our presence and relationships with these customers and also the market dynamics that we're all seeing and reading about in India.

And we do expect Asia-Pacific to be most likely our fastest-growing region over the next couple of years.

It's not just India, although India is a stellar part of it.

Australia, we're doing extremely well with Telstra through the Ericsson partnership.

Japan has opportunities for us.

We're going to be attacking in some other countries in the region.

So Asia-Pacific will be a place of growth for us for the foreseeable future.

Great.

And then Jim, just shifting to gross margins with that mid-40s target in fiscal fourth quarter.

Given that we've seen several sequential fourth quarters up and down over the past couple of years, can you help us put a finer point on how we should think about trends on a sequential basis?

And then as we kind of look out into FY '18, is it still fair to say that gross margins should have a net tailwind, given the combination of mix and software playing into your model?

I feel really good about our progress in gross margins.

We've done a lot of good things.

We've taken the cost down of our goods.

We've expanded our scale, so our overheads are declining as a percent.

We've done better with respect to our quality, so our warranty has done very well.

And we're in the later stages of a number of projects in which we were attacking.

But as you know, gross margin is a very difficult number for us to predict in the near term with certainty because things change as we move through the quarter.

I feel like our sort of range for gross margins is mid-40s.

I think we're going to see some movement around that range.

But I feel good about that range.

Haven't done planning for next year, we're still working through all that.

I would say that we still feel good about a 15% operating margin target for us.

That will take some improvement in gross margin over the next few years.

I can't make a prediction for 2018 at this point in time.

But I do think it will improve over time.

Our next question comes from Paul Silverstein from Cowen.

I was going to ask you a single question.

But given that none of my peers have, I'll asked you multipart where the parts have nothing to do with each other -- first off, can you tell us with the 32 percentage points of the U.S. or approximately 32 percentage points that was in AT&T and Verizon, how much is U.S. federal?

And how much is U.S. regional as a percentage of that revenue?

Paul, it's tough to break that down.

Obviously, I don't have the stats to that.

I mean, basically think about it like this.

There's a lot of components to that element.

You've got some DCI in there.

You've got enterprise in there.

You've got government.

You've got a lot of elements in there.

And to split all of that out into that granularity is challenging.

What we tried to represent here, I would think about it generally as the Tier 1s, I think, are very consistent and are doing well.

And we're seeing some softness in really what I would characterize generally as the sort of regional, smaller-type carriers.

It's not a huge part of our business, Paul.

It isn't.

But in aggregate...

Gary, apologize for the interruption.

But what I'm trying to understand on this question is $35 million, if I assume that's CenturyLink and some other Tier 1s beyond AT&T and Verizon are a healthy portion of the remaining 30 percentage points or so, it seems to me I guess, it would be 32 percentage points or so of the total revenue.

It sounds like the U.S. fed and regional -- I'm trying to do the math.

It sounds like the $35 million miss would be 25 to 35 percentage points of the total revenue you did from those 2 groups or thereabouts.

I mean, it sounds like it's a significant...

Let me come at this (inaudible) in a different way, Paul, without -- and we could take this offline if we get to the detail of it.

But let me give you another sort of -- our government business generally is about $100 million to $125 million a year.

That's where the -- directly into the government.

And what -- and the statistic I gave you was that of the $35 million delta between original consensus and where we're now forecasting, about $25 million of that is the delta from expectation.

Okay, all right.

Let me move on.

I'll take the rest of it offline.

On DCI and Web 2.0, how much revenue concentration -- I assume there's a fair amount of concentration at this point, given the size of some of your customers.

And can you tell us what is a typical project deployment cycle?

Again, I don't think there is a typical project deployment.

I mean, basically the dynamics of the industry in the DCI market is you've got the 4 big content players out of North America.

You've got their content players, the smaller content players relative to the spend in some of the other countries coming out.

You've got those content players spending in North America.

But increasingly, their growth is outside of North America.

So things like submarine, if you look at the transatlantic submarine ownership of capacity now, it's owned by the top 3 content guys.

It's not the carriers anymore.

So it's quite a diversified dynamic across the global revenues.

All right.

Gary, pricing environment, a number of your competitors have cited price pressure.

Are you seeing that to any full extent?

Has the rate of price erosion changed for you?

Listen, it's always a challenge, where I think you've got a number of smaller competitors that are desperate and are doing some unsustainable things.

But really -- and that remains challenging.

That's nothing new.

We've seen that before.

So I don't think there's anything appreciably different in that dynamic, except that they get more and more desperate as their revenues continue to shrink.

All right.

And finally, I know there's sensitivity about any particular customer.

But on AT&T, is there any signs of weakness?

Notwithstanding the strength in the quarter, any signs of coming weakness tied to the Time Warner close or for any other reason?

No.

Our next question comes from Dmitry Netis from William Blair.

A couple of questions.

Just going back to that single-digit growth rate as we sort of model out '18, I understand you don't have all the puts and takes yet for next year.

But just help us understand, is that the overall market, which includes various speeds?

And if we were to separate the 100 gig or 200 gig out of that number, what the growth may look like.

And if you could help us understand what your exposure is today to the 100 gig, 200 gig market as a percent of total sales or products, however you want to define it.

And also if you would, remind us where the web-scale is as a percent of sales both on direct and indirect basis.

So I would -- we'll step through that.

There's multiple elements to that.

I would think about the growth dynamics less about the technology pieces frankly and more about the geographic demand with the exception of the DCI space.

I think what you're seeing overall as a dynamic is Asia-Pacific growing and really stability in Europe, probably 0 growth in Europe, down a bit in CALA and down in North America probably for the year.

And I think I would think those are the overall macro dynamics.

As we talked about, I do -- our best view that we have is that may improve, we think, in North America for -- as we go through '18, probably the second part of '18.

But what I would say, and the reason that we're obviously continuing to grow in a down market, is we're exposed to all of the good stuff.

The strong Tier 1 relationships, they've been very solid.

I would expect them to continue to be solid going forward.

The DCI market, we have #1 market share there.

It varies quarter-to-quarter.

But typically, about 10% of our revenues go direct with those guys, and then considerably more than that comes indirect with them through third-party relationships.

India, we have #1 market share in India now.

And that is obviously, we think, the fastest-growing market.

Subsea, we're #1 market share with about 40%.

We think that's going to continue to grow.

So I'd look at, first of all, the geographic splits on growth as the most important.

Secondly, within that, the sectors, obviously DCI and countries like India.

And then Steve, do you want to talk a little bit about the sort of 100 gig dynamics?

Sure.

So just to fill in the blanks on -- because you asked about the rates and such, right?

So clearly, 40 gig is in decline, right?

It's basically a channel fill business for the areas that deployed it.

WaveLogic 3 was a 200 gig product.

Ai is a 400 gig product.

And when you look at who takes what flavors of things, typically carriers have been at around 100, the web-scale guys were 100 and rapidly went to 200, right?

We expect they're going to continue with that kind of pace.

And again, I'd go back and point you to we're able to drive the pace of innovation here, right?

And every time we are able to increase data rate, that drives cost out and the rest of it, right?

So we're well placed to ride this continuing curve of demand.

Yes.

And we had a particularly good quarter with the web-scale companies.

We've said historically that we've been in the range of 5% to 10%.

We are still in the range of 5% to 10%, but we're toward the upper end of the range and a very strong quarter.

It was up sort of 21% sequentially or something like that.

So Jim, is -- if I just look at web-scale and include the indirect business, would we be in that 35% to maybe heading to 40% range now?

Is that -- have we broken through that threshold or...

I still think we're in the 15% to 20% range today.

I'm sorry, 15%, okay.

1-5 to 20%, which is the range that we've talked about historically.

The (inaudible) is the direct sales, probably an equivalent amount is indirect.

Okay.

I guess I mismodeled it with the enterprise and kind of cable and stuff, which would get you into that 1/3 of the business, right, so...

Our next question comes from Doug Clark from Goldman Sachs.

First one is on AT&T.

It looks like it was up pretty nicely sequentially in the quarter.

I'm wondering if that has any impact from -- I think the last quarter you talked about some metro edge project wins with them.

So was there any contribution from those?

Or is that still to come?

Most of that is still to come.

But what I would say is that people were inquisitive about whether we were going to see an uptick with AT&T in the second half.

And we have seen it and we expect to continue to see it as we move through the year.

The business with AT&T is in great shape.

Okay, got it.

Okay.

And then alternatively, I know you kind of mentioned some of the lumpiness in Asia-Pac and India in particular.

But off of a very strong quarter last quarter, it did step down pretty significantly sequentially.

Was that India?

Or is that other areas of Asia-Pac?

And if you can kind of comment or help level-set us on what you believe is the kind of the typical seasonality.

Is it fairly consistent with what we see in North America for how India should behave kind of in a typical calendar year?

I would say that our experience of India so far is that the seasonality stuff really doesn't sort of seem to pervade, given the sort of dynamics of the marketplace.

It's going to ebb and flow depending on the build-outs.

And I think you saw a bit of an ebb in Q3 again.

And when you think about the Q3 performance we had, India was down in Q3.

It was expected and it was built into our guide.

But I think it helps build this overall picture of a globally diversified player with multiple customers exposed to the higher growth markets.

So we expect actually India to be up in Q4 and very strong for the year.

Year-to-date, it's about 80% growth.

Yes, it is going to be lumpy though because it is a project-based business, where we do the deployment.

And we need to get to acceptance of a project before we can take revenue.

That means it's, by its very nature, it's going to be deployment.

If you look at our deferred cost of sales, India is a big piece of that.

Our next question comes from Tejas Venkatesh from UBS.

I just wanted to clarify your earlier comment on Tier 1 telco customers.

I wanted to clarify that AT&T is still on track for a flat to slightly down year, implying a very strong 4Q.

And then do you expect Verizon to be a 10% customer for the year?

We do expect a strong Q4 with AT&T.

As far as Verizon being a 10% customer, it was a 10% customer in Q1.

It was not in Q2.

It was in Q3.

And we're doing very well with that customer.

We sell just about everything we have on the truck to Verizon.

I won't make a prediction about it being a 10% customer in Q4.

But we're doing extremely well with them.

We're proud to be associated with them.

Sounds good.

And then any thoughts on EMEA spending?

When we spoke a quarter ago, you had started to see some signs of improvement.

Are you seeing that continue?

I would say that EMEA overall, I'd describe as sort of stable for us.

And obviously, we think we're well placed with some of the tier -- when we talk about Tier 1s, by the way, we obviously always focus on North America.

But it's our Tier 1 coverage globally.

And we've got some very good Tier 1s in Europe, Orange, BT, et cetera.

So we actually think we'll see some growth as we come out of the year in EMEA.

And I think we should be pretty well positioned as we go through 2018 in EMEA as well to see a little bit of growth there.

Our next question comes from Alex Henderson from Needham.

The first question I wanted to ask you is the exchange rate swings that we've been seeing.

Is it having any impact on your business?

And how do you see that playing out?

We've seen a bit of impact on our OpEx because the Canadian dollar has strengthened considerably.

The European currencies have strengthened a bit.

So a little bit of effect on our OpEx, although we've managed that pretty well.

And that's about it for us, so far.

A second question, if I could, on the DCI market.

Obviously, the Ai has higher bandwidth per wavelength.

But are you seeing price pressure as they build volume?

Those guys are notorious for pricing very aggressively.

Can you talk a little bit about sort of the (inaudible) trade-off between higher bandwidth of your system versus the pricing pressure?

I'll take the pricing, then hand to Steve for the dynamics of the platform.

Listen, there's a notion around the DCI market.

Listen, I think they're obviously very demanding and knowledgeable customers with a lot of spend and leverage.

But frankly, that's no different than the Tier 1 carrier market as well.

It's not dissimilar dynamics, so we're not unused to that.

Yes.

And Alex, I'd point to the fact, keep in mind, we're able to drive cost out of networks, right, the way we do with Coherent, with WaveLogic, right?

So for the same kind of investment in chip, if I can double the data rate, I effectively cut the cost for that bit in half, right?

So we're actually able to drive the kind of cost reductions that not only the web-scale but the carriers want to see, right?

That's one of the inherent benefits of being able to just go faster for wavelength is your cost per bit starts to drop, right?

So we're able to, again as we're driving the innovation curves here, we're able to drive cost out of networks.

And to Jim's earlier points, we've been able to maintain good margins in the process.

We do believe WaveLogic Ai sets a new standard.

And we think it's going to position us very well with a whole range of customers, including the DCIs and Tier 1s and regionals.

So it's a big step forward for us.

It's not in our numbers yet.

It will start to appear in our numbers in Q4 and should have a pretty significant effect next year.

One last question, if I could.

The India are also notorious on pricing and margins.

I think you've said in the past that margins in India are roughly comparable.

Are you seeing any change in behavior there or change in pricing there that would be causing a mix shift to a lower margin environment in that geography, which is it's notorious for?

Yes.

What I'd say there is we -- there are a number of new projects underway in India.

And we will compete and hopefully we will win.

It is -- I'm not sure that you can state that India overall, margins are lower.

But as we've said often, early period projects are going to have lower margins than later period projects because of the mix of gear that you sell.

Also we're going to cost-reduce our product over time.

So we'll see some effect of the attacking margins that we're going to see in India.

But we don't see that, that's going to be a huge effect on our overall result.

We've got a lot of good things happening in our gross margin.

And we'll have some lower margin early period projects in our margins.

Our next question comes from Tim Savageaux from Northland Capital.

A couple of questions and sort of related or I think maybe related.

Jim, you gave a sequential growth number for the cloud space.

And I guess, my question was, was that -- well, assuming that's for the entire direct and indirect.

But however you want to slice it, can you estimate a year-over-year growth number either direct or indirect or both for cloud either in the quarter or year-to-date?

And then bigger question, do you feel like there might be some relationship between the cloud strength that you're seeing and the regional service provider weakness, which is to say, at some point, does a lot of the kind of do-it-yourself network building on the part of the cloud guys start to put pressure on wholesalers, I guess, in terms of selling wavelengths, dark fiber, what have you, anything to it, a potential dynamic there?

And that's it for me.

Tim, let me try and -- this is Gary.

Let me try and talk to the -- it is tough to break out the applications because obviously we don't always get visibility to exactly the sort of application.

I would say though that the growth of Waveserver is really a bit of a proxy for that DCI space and the kind of growth to it.

It's not an absolute fit.

But frankly, that platform is really addressed into directly into that market and for the DCI sort of interconnect, both with the content guys and the other folks that are connecting data centers as well.

So I think as that settles down, I think that will be a pretty good proxy to talk about that.

The second element you talk about is an interesting one, which is really -- as more of the cloud is stood up, for want of a better description, does that impact the flows of traffic negatively or positively to the regionals?

I would say that our view on that is probably not at this stage.

What we're seeing in some of the challenges in North America in some of the regional carriers are pretty specific to each of those carriers.

And when you think about the cloud market generally, we're either connecting content-to-content or content-to-users.

And frankly, these folks have the content-to-users connectivity piece.

So even if you look at the dynamics of cloud, they should actually benefit from that.

I think to your point, Tim, around the wholesale piece of point-to-point connectivity, that may come under pressure as increasingly the content guys are building their own direct data center to data center connectivity.

You could see some change in flows there.

And the only other thing I'd add, Jim, is that we've said this over and over again, don't build a trend based on a quarter.

We point out things that happens in a quarter because we think they're meaningful.

We're having great momentum with the web-scale players.

We're #1 across the board of the applications.

And we have not yet gotten Ai to market, which we believe will put another marker on the board for us and help us gain market share.

So feel really, really good about that set of customers.

If I could just follow up for 1 second, that's kind of why I was asking for a year-to-date cloud growth number, including the other stuff as well.

I assume Waveserver in the triple digits.

And so I'm not trying to model that ad infinitum but I would love to try and tie you down on (inaudible)

If you look at the direct web-scale customers, Tim, here's another way which I think might be what you're getting at.

It's up about 20% sequentially...

Quarter-to-quarter.

Quarter-to-quarter.

We don't have the year-to-date number right handy.

We have it here -- or we'll try to get it before the call ends.

Our next question comes from Tim Long from BMO Capital Markets.

Just 2 if I could.

First, on the subsea business, could you just update us?

It sounds like another win in the quarter.

How meaningful is that now?

And how long do you think that cycle can last?

And then secondly, Jim, it seems like OpEx came in a little bit lower than I thought for Q3 and a little bit lower for Q4.

So anything structural to the better OpEx performance in the second half of the fiscal year?

Or is it just likely a little bit lower revenue impact on those numbers?

Tim, let me take the first part of that.

I think what's happening in the submarine cable space is we're moving, I think, from an upgrade environment, where basically you've got the cables around the world being upgraded.

And clearly, we disrupted that space.

And we're now #1 market share with about 40% of that.

The relationship specifically with TE is designed for what we now see as a shift in not just upgrades, but people are having to build new cables just for sheer capacity demands around the globe.

And what we're seeing in these first few wins with TE is really a validation of that relationship and our ability to get into a new part of that market for us, which is these new cable builds, which typically we didn't have much visibility to.

It was a complete turnkey project.

And so we actually think of that, that market over the next few years will be, excuse the pun, quite buoyant.

Starting in '17, we're already seeing that ride the way through the next 2 to 3 years.

And the lead times on these cables are obviously very long, given the logistics involved on it.

So we think we're well placed to continue to consolidate our position there.

And on OpEx, I would say that we work very hard on controlling OpEx.

We do believe that we must continue to invest in this business, and we do.

We spend about $400 million a year on R&D all-in.

And we'll continue to do that.

There's nothing particularly exceptional in those numbers.

It's just we're controlling it as best as we can.

And it is one of the things that is in our control.

We've done -- over the last 5 years, we've actually done better on OpEx than we predicted on many, many quarters.

Thanks, everybody, for your attention today.

We look forward to connecting with everybody over the course of the coming days and weeks.

Ladies and gentlemen, thank you for participating in today's conference.

This does conclude the program, and you may now disconnect.

Everyone, have a wonderful day.