ADTRAN Holdings Inc (ADTN) 2016 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to ADTRAN's third-quarter 2016 earnings release conference call. (Operator Instructions)

During the course of the conference call, ADTRAN representatives expect to make forward-looking statements which reflect management's best judgment based on factors currently known. However, these statements involve risks and uncertainties, including the successful development and market acceptance of core products, the degree of competition in the market for such products, product and channel mix, component costs, manufacturing efficiencies, and other risks detailed in our annual report on Form 10-K for the year ended December 31, 2015, and Form 10-Q for the quarter ended June 30, 2016.

These risks and uncertainties could cause actual results to differ materially from those in the forward-looking statements which may be made during the call.

It is now my pleasure to turn the call over to you, Mr. Tom Stanton, Chief Executive Officer of ADTRAN. Sir, please go ahead.

Thank you, Robbie. Good morning, everyone. Thank you for joining us for our third-quarter 2016 conference call. With me this morning is Roger Shannon, Senior Vice President and Chief Financial Officer.

I'd like to begin this morning by discussing the details behind our Q3 results and I will end with some comments on what we see for the future. Roger will then discuss our Q3 performance in more detail, and you (sic) will then open the call up for questions that you may have.

As we stated in our earlier press release, revenues for the quarter were $168.9 million, coming in ahead of expectations, up 4% sequentially and 7% year over year. Our total network solutions revenues, including both international and domestic markets, came in at $136.3 million, and total services and support revenue came in at $32.6 million, a new record for the Company.

Revenues from our domestic markets came in at $127.7 million or 76% of the total, and international revenues grew to $41.2 million for the quarter or 24% of the total. On a year-over-year basis, our domestic revenues were 7% as we saw strong growth in our services business and continuing momentum in CAF II and vectoring demand. These positive trends affected both our product and services revenues in both Tier 1 and Tier 2 accounts.

Our international business was up sequentially 42% and up 7% year over year. This quarter was helped by returning strength in vectoring shipments in Europe.

Moving down a little deeper, our access and aggregation category had a solid quarter, growing 18% sequentially and 17% over the same period last year. This growth was broad-based, as we saw growth both domestically and internationally.

Customer devices was down sequentially after a strong Q2, but that product category continues to perform well and as expected, growing nearly 10% through the first 3 quarters, driven by strong demand in ONTs and EFM.

During the quarter, we continued to make advances in our Mosaic cloud architecture. We started the quarter announcing our Mosaic SDN cloud platform and our Mosaic NFV operating system. That was followed by the introduction of our orchestration and control functions within Mosaic.

We exited the quarter having completed lab certification for our Mosaic software with one of our Tier 1s here in the US and continue to work with other Tier 1 and Tier 2 carriers in the US, Europe, and the Middle East. We believe ADTRAN is the only access vendor with SDN controls, software modularity, and application virtualization, and is leading the development towards the next generation of access architectures.

G.fast continues to gain momentum around the world, and we ended the quarter having been awarded primary supplier with yet another Tier 1 carrier in the EMEA region. Our awarded Tier 1 business here in the US continues to progress well as we formally exited the lab process and entered first-field applications for the first phase of deployment.

Finally, our NGPON2 initiatives continue to progress well as we bring to market the world's most advanced fiber access solution utilizing multi-wavelength TWDM PON, virtualized OLTs, and SDN control.

Looking forward, our focus remains on being the world's most comprehensive access solution provider, with clear industry-leading solutions in fiber access and ultrahigh speed copper. With the world's most complete access virtualization products, we believe we are well positioned to capitalize on the evolution in access as carriers around the world upgrade their infrastructure to meet customer demand.

I'd now like Roger Shannon to review our results for the third quarter of 2016 and provide some comments on the fourth quarter as well. We will then open the call up for any questions. Roger?

Thank you, Tom, and good morning. During my report, I will be referencing both GAAP and non-GAAP results. The GAAP results for this quarter include the impacts of the acquisition of various product lines from CommScope along with the impact of streamlining activities associated with our European operations.

ADTRAN's third-quarter revenue was $168.9 million, which is up 7% compared to $158.1 million for quarter three 2015 and up 4% from the $162.7 million we reported last quarter. Our network solutions revenues for Q3 were $136.3 million, down slightly from the $138.1 million for quarter three of 2015 and $138.5 million reported for Q2 of 2016.

Services and support revenues for Q3 of 2016 were $32.6 million, up 63% compared to the $20 million earned in quarter three of 2015 and 35% ahead of the $24.2 million reported for quarter two of 2016. Across our revenue categories, access and aggregation revenues for Q3 2016 were $120.6 million, up 17% compared to $103.4 million for quarter three of 2015 and up 18% compared to the $102.2 million for quarter two of 2016, as we had growth in both our US and our broadband services and internationally in our European access business.

Customer devices revenues for quarter three 2016 were $33 million, down 7% compared to $35.5 million for quarter three 2015 and down 19% compared to $40.9 million for quarter two 2016, mostly attributable to lower sales in our Ethernet portfolio of products.

Finally, traditional and other product revenues for quarter three 2016 were $15.3 million, down 20% compared to $19.1 million for quarter three of 2015 and down 22% compared to $19.6 million for quarter two 2016.

Looking at our geographic mix of revenues, domestic revenues for quarter three 2016 were $127.7 million, up $8.2 million from $119.5 million we reported in Q3 of last year and down $6 million from the $133.6 million in quarter two 2016. International revenues for quarter three of 2016 were $41.2 million, up 7% compared to $38.6 million in quarter three of last year and a 42% increase to the $29.1 million for quarter two of 2016. We published the reporting of each of these categories on our investor relations webpage at adtran.com.

For the quarter, we had three 10% of revenue customers, 2 domestic and 1 international, representing 26%, 16%, and 16%, respectively, in quarter three.

Our GAAP gross margins for the third quarter of this year were 44.9% compared to the 44.7% of Q3 of 2015 and 48.5% for quarter two of 2016. Our non-GAAP gross margins for the quarter were 45.8% after taking into account the impact of our previously mentioned European streamlining and product line acquisition.

Both results were negatively impacted by a material increase in warranty expense associated with a defect in a part provided by a third-party supplier. While we anticipate recoveries from third parties associated with this defect, we recognized expenses in the quarter associated with the remediation.

Total operating expenses on a GAAP basis were $65.7 million for quarter three of 2016 compared to $62.6 million for quarter three of 2015 and $64.1 million for quarter two of 2016. On a non-GAAP basis, our Q3 operating expenses were $63.1 million compared to $59.6 million in quarter three of last year and $62.3 million last quarter. The increases in GAAP operating expenses compared to quarter three of 2015 and quarter two of 2016 are a result of increases in variable incentive compensation and the European restructuring charges.

Operating income on a GAAP basis for the quarter just ended was $10.1 million as compared to the $8.1 million reported in Q3 of last year and $14.8 million earned in quarter two of this year. Non-GAAP operating income for quarter three 2016 was $14.3 million compared to $11.2 million in quarter three of last year and $16.8 million in quarter two of 2016.

As described in our supplemental information provided in our operating results disclosure, stock-based compensation expense, net of tax, was $1.3 million for quarter three of 2016 compared to $1.5 million reported in Q3 of last year and $1.3 million in Q2 of this year. All other income, net of interest expense, for Q3 of 2016 was $5.4 million, inclusive of a $3.6 million bargain purchase gain compared to $2.8 million for Q3 of 2015 and $1.6 million for Q2 of this year.

The bargain purchase gain associated with the acquisition of active EPON and RFoG assets from CommScope represents an estimate of the excess fair value of net assets acquired as compared to the consideration exchanged.

The Company's tax provision for Q3 of 2016 was $3.1 million or an effective tax rate of 20% compared to a tax provision rate of 35% for Q3 of 2015 and 37.8% for Q2 of 2016. The decrease in the tax rate is primarily attributable to additional R&D benefit recognized in the quarter, the tax impact to the bargain purchase gain, and an increase in international revenue.

Net income for Q3 2016 was $12.4 million compared to $7.1 million in quarter three of last year and $10.2 million in quarter two of 2016. Earnings per share on a GAAP basis, assuming dilution for quarter three of 2016, were $0.26 per share compared to $0.14 per share in quarter three of last year and $0.21 per share for quarter two of 2016.

Non-GAAP earnings per share for quarter three of this year were $0.26 compared to $0.19 for quarter three of last year and $0.25 per share for quarter two of this year. Non-GAAP earnings per share excluded the effect of amortization of acquired intangibles, restructuring expenses and stock compensation expense, and the bargain purchase gain from the acquisition. The reconciliation between diluted GAAP earnings per share and diluted non-GAAP earnings per share is provided in our operating results disclosure.

Turning to the balance sheet, inventories were $96 million at the end of quarter three, up from $86.9 million last quarter. The increase in inventory is attributable to higher field inventory awaiting network installation and acquired inventory from our acquisition of product lines from CommScope.

Net trade accounts receivable were $101.8 million at quarter end, resulting in a DSO of 55 days compared to 50 days at the end of the third quarter of 2015 and 50 days at the end of Q2 2016. The increase in DSO for the quarter was a result of higher international sales.

Unrestricted cash and marketable securities net of debt totaled $256.3 million at quarter end after paying $4.4 million in dividends and repurchasing just over 352,000 shares of common stock for $6.3 million during the quarter. For the quarter, ADTRAN produced $7.2 million of cash flow from operations.

Looking ahead, the book and ship nature of our business, the timing of revenues associated with large projects, and the variability of order patterns of the customer base into which we sell, and the fluctuation in currency exchange rates in international markets we sell into may cause material differences between our expectations and actual results. However, our current expectations are that Q4 2016 revenues will be down approximately 10%, slightly better than normal seasonal trends.

Taking into account the potential impact of account currency exchange rates and anticipated mix, we expect that our fourth-quarter gross margins on a GAAP basis will be slightly higher than our third-quarter GAAP results, which as described earlier were affected by unusual events. We are also expecting GAAP operating expenses for the fourth quarter to be slightly lower in Q3.

Finally, we anticipate the consolidated tax rate for Q4 to be in the mid-30% range. We believe the significant factors impacting revenue and earnings realized in 2016 will be the following: macro spending environment for the carriers' enterprises; currency exchange rate movements; the variability of mix and revenue associated with project rollouts; professional services activity level, both domestic and international; the timing of revenue related to the Connect America fund projects; the adoption rate of our broadband access platforms; and inventory fluctuations in our distribution channels.

With that, I will now turn the call back over to Tom.

Thanks very much, Roger. Robbie, at this point, we are ready to open it up for any questions people may have.

(Operator Instructions) Rod Hall, JPMorgan.

Thanks for taking my question. This is Ashwin on behalf of Rod. Just a clarification. Apologies if I missed this, but did you quantify how much of an impact did you have on gross margins because of the increased R&D expense?

We didn't break that out, Rod. And the reason is is we are still kind of in negotiations with some of these -- with a couple of the third parties.

So I would tell you it was over a point. It was materially over a point, you know, the impact in Q3. But we haven't broke out the exact dollar amount.

Okay. If we go back to the European restructuring you started in the quarter, is that mostly done? Or do you still -- are you still in the progress of making [maybe more construction]?

It's mostly done. There may be a little bit of cleanup from an accounting perspective in Q4, but it's mostly done.

My last question really is on services. Could you just comment on what's driving the strength here? Is it sort of related to any specific projects? How should we think about that going forward, primarily into next year?

Well, it is driven by projects. And I think you are probably aware of the projects. So CAF II is a big driver. We expect CAF II to continue on into next year.

We have fairly large vectoring deployments going on here in the US. We would actually, based off of what we are hearing and kind of what you are seeing in the press, we would expect that to actually be materially larger next year. So I would expect we should have strong services number in Q4 and I would expect it to actually grow next year.

Great. Thank you.

Doug Clark, Goldman Sachs.

Great. Thanks for taking my question. My first one is on the international piece. A nice sequential growth, and I think you commented on part of the improvement coming out of Europe in vectoring in particular. So is that growth attributed to the one historically large European customer?

And related to that, last quarter -- or actually for the past two quarters -- you've commented on increased competition or potential share shifts in that customer count. What are you seeing in that large European customer and what is your outlook, at least for the fourth quarter or near term? Is it going to remain at these levels, fall back, etc.?

So they were definitely stronger in Q3 than we planned. And yes, it is driven -- I would say we had other customers in the region, but the biggest piece was driven by the large vectoring rollout that we are doing there.

Really, I don't know if I would call it share shift. I think you had -- they were focused on different footprints at different times and that kind of benefited us in Q3. We think that will benefit us in Q4 as well. So, I mean, it's kind of what you said, which is just more focusing back on our product or on our region.

Okay. That's helpful. And then on G.fast broadly, you mentioned kind of the US win and another European Tier 1 win. There have been some large kind of notable customers, like British Telecom and CenturyLink, that have seemingly gone to other G.fast equipment providers.

Can you talk about the competitive environment? What you're seeing from a kind of share of wins? Are you still involved in those customers that I just mentioned that seem to have gone to competitors? Yes, [if you think about] --

Yes. You got two different types of things going on there. So the European customer you are talking about, yes, we are still involved, but I would say we are not in the driver's seat there. And they've been fairly transparent about their process.

We still have an ongoing trial there. We are continuing to try to show the benefit of our technology, which is materially differentiated from the other competitors there. But we still have a lot of work to do.

In the US, I think it's a completely different thing. I think the customers, including the ones that you talked about in the US, are still very much open. We are still doing things with that customer in regards to G.fast. I think it's just a matter of timing.

Great. And actually, and one follow-up on G.fast as well. The US customer in particular that's now exiting lab trials and going into deployments, can you give us a sense for kind of the size and scope and perhaps slope of the deployment there? Just to help level set us.

Sure. Let me be clear about this. I said first off is deployment, which means it's still kind of a trial phase. It's out of the lab, but it's actually out there with live customers now. And there's some kind of soak period with that before they really open up the floodgates and start deploying it en masse.

The reality is is I don't know; that's a customer that's very difficult to forecast. We are very happy that we got through this piece. The first phase of this -- there are at least a couple of phases to this. The first phase of this, of course, is kind of the gating item to start receiving purchase orders and things.

So we are happy with where it is, but I am not at this point trying to put a number -- definitely not a timing number. I think our belief on the total size of the opportunity hasn't changed, but I can't tell you exactly what it will be even in Q1 at this point.

Thanks for taking my questions.

Michael Genovese, MKM Partners.

So there was certainly a change in the international business compared to what we heard on the forecast on the last call. It sounds like it was all one customer in Europe, but I just want to check. Was there anything in Latin America that was different? And why did that project get pulled forward again, the European one, after getting pushed out?

Even the best carriers don't forecast well. And this was I think the demand outstripped what they were planning on. But the answer to your question is there were other customers that were up in Europe. But this one kind of led the phase.

You know, another thing going on there, though, is they have recently awarded business for their next phase of deployment. We did very well on that award and we'll start seeing that ship in 2017 as well. And that will be on top of whatever we are doing today for fill in on our existing devices.

Okay, great. And then on the Tier 3 market, reading between the lines, it seems like Tier 3 was probably pretty weak in the quarter. And is there a delay there? Are they waiting for the CAF funds, and should we expect Tier 3 to do better in 2017?

I hope so. We've always been a little reticent on Tier 3 because I think I think it's -- first of all, it's a very disaggregated customer base. And so everybody is thinking differently about what they want to do. I think the next point in time is in November.

Roger, you got --?

That's right. So the rate of return carriers, the Tier 3 carriers, are expected to make their decision in November between staying with the rate of return model or the alternative Connect America model, which is more of a price support model like the price cap.

So again, we've said that we haven't included any forecasts or expectations in this year's revenue for Tier 3, but as we've said before, we still expect it to be positive for next year.

At some point, it's going to kick in and they will have made all their decisions and they will have gotten their models right. But it's been definitely a protracted decision process.

Okay, great. And then final one for me right now. On the gross margins -- I understand there were some one-time issues in the gross margin in the quarter, but is there also still a significant regional mix difference? So the strength in international is part of the reason that we were down on the gross margins?

If you look at just gross margins, there are three buckets that we tend to look at. One is international versus domestic. And international is typically -- is lower, although that can be mitigated or exacerbated by the product mix that we ship into that region.

So when we ship chassis, we don't do so well on gross margin. When we ship line cards, we do materially better. But yes, the answer to your question is yes on that.

And then the other one, of course, is services, which has lower gross margin. It is accretive. There's very little operating expense associated with that, but does have a little gross margin. So as you see in our gross margins -- our services business picked up, but that hasn't been accretive to gross margin, although it's been accretive to operating income.

Just as a follow-up, then, just why is vectoring and/or CAF II have a richer mix of services than other businesses? I'm just trying to understand.

When they've started doing both of those projects, CAF and vectoring, we have -- I don't know if it's coincidence. I think probably it has to do with just staffing levels and being able to manage their run rate versus major projects like this.

But the customer base has been much more open to allowing third parties to go in and do the installations. And that's for both of those. We don't see that changing; in fact, probably accelerating next year. We'll see a --we are expecting very good business out of our services business next year.

Great. Thanks for all the questions.

Rich Valera, Needham & Co.

I have a follow-up on the gross margin. It sounds like you had over 200 basis points of what I'd consider nonrecurring items in the gross margin, the restructuring plus the warranty expenses. Yet you are only calling for a slight improvement from that GAAP gross margin, which includes those in fourth quarter.

Are you expecting any more warranty expenses in the fourth quarter? And if not, why wouldn't you get a little bit more of a bounce in the gross margin?

There may be, and we're trying to make sure that we are cautious around the potential impact of that. But we're also taking into account the fact that there will be a lot of push to get services done in Q4. So at the end of the year, everybody wants to get their projects done.

So you'll see our services business go up, and then we talked about the strength in Europe and the fact that that would drive it. So it's really probably -- I would call it more of a -- I don't know if you call it product mix, but definitely business mix, where you'll see a stronger Europe than we had anticipated and then you're going to see a stronger service profile. And then there is some worry about should we -- just making sure that we capture all the bases on the warranty recall.

Got it. So as we think about next year, presumably you are through this warranty issue, or at least if you look at the year in total, much less impact from it. And does the mix we think skew a little bit more North American, so in general, you have a -- I'm just trying to think about how we should think about gross margins as we head into next year.

That's a good question. So the reality is I really don't know yet. We'll definitely know on our next calls and be able to give some color on that.

But the drivers are, without a doubt, services is going to be higher next year. Our vectoring builds and product shipments in the US would be materially higher next year because there are some big projects that have been announced. We of course start shipping G.fast here in the US next year.

You know, the offset to that growth is -- I shouldn't say offset, but the counterbalance of that growth is we expect our European business to be materially higher because of the new footprint expansion that we were just awarded. So I really don't know the mix of that right now at this point.

Understand; a lot of moving pieces. Just following up on that international business. Any update on your Latin American customer, where you were kind of behind on some lab trials? Have you made any progress there?

It's two steps forward and three steps back. So, yes, we're still going through the lab process on two different -- really three different initiatives. Two of them are PON related. We still fully expect to be out by the end of this year.

That customer is going through its own work from a regulatory perspective. I think that it's fair to say that they are looking hard at the regulatory changes that have been made over the last six months which affect them. So it's a very -- sooner or later, the product has got to be built out. And I think hopefully sooner than later.

But from a lab perspective, we are good. We should be out this year, but I think it's still kind of an unknown as to when that spigot kind of turns back on.

Got it. Just one more from me. Can you say how much revenue you are expecting the CommScope acquisition to contribute in the fourth quarter?

Not a lot. I mean, we may be surprised there, but we are trying to be really cautious about that. The CommScope piece was very good for us strategically without -- it got us much better positioned with some US MSOs and got us into a position to be able to actually talk about our broader product set, not just the product set that was being sold there. So we are very happy strategically.

But we did not buy it for the revenue. We bought it to really kind of open the doors that we were trying to expand into. Having said that, there will be some contribution, but I wouldn't call it material.

That's helpful. Thank you, gentlemen.

And just one last piece on that. There are some awards that have not started shipping in that area that will start shipping next year that -- and I think there may have been actually press releases out about that and maybe even prior to our acquisition. So some of that stuff is still not cooked in yet as well.

Okay. Thank you.

Paul Silverstein, Cowen.

On the gross margin guidance, I just want to make sure I understood the guidance going back to Rich's question. I thought I heard you say that you expect it to be slightly better than the GAAP number, but are you saying that you expect it to be up sequentially, albeit slightly? What exactly are you saying?

On a GAAP basis, sequentially up from Q3.

All right. My larger question on gross margin is putting aside the near-term issues on the warranty, Roger, what's the reasonable -- what's reasonable peak that you could obtain over the course of the next year and longer term in terms of the trajectory? Any thoughts you can share with us? And then I've got a couple more, if I may.

We are really not going there yet. As Tom just mentioned just a few minutes ago, when you look at Q4 and the reasoning behind that guidance, there is some business mix related to international and as well as to the services.

But we will certainly give more of an outlook at the end of the year looking at 2017, but we've really not broached the subject of longer-term cyclical --

Some of it, yes, that's going to be very dependent on mix, of course, as you know, Paul. So the good thing is I think we came in thinking we would improve it 100 basis points to 200 basis points this year. I think we're still confident that -- and that would be for the full year and we're still confident we are able to do that. And then we'll take a look if there's any change to where that put us out, which it was kind of 45, 46, which was materially better than last year. And then we'll look again this year and see if there would be improvement as we go further into our planning process.

All right. And Tom, I heard you reference -- I think I heard you reference a new Tier 1 EMEA win. Is there any insight you can share with us? The nature, the magnitude, etc.?

Yes. That's a customer that is very -- in fact, there was a press release -- we have a couple -- one, we did in -- well, the last one and the one I was more referencing was the Zac move from a trial to deployment. And we were able to capture the prime award in that country, in Israel.

They are very bullish on G.fast. And it's not just G.fast. It also includes our Mosaic infrastructure. So we are -- that's a company that's fully utilizing what our capabilities are, so we are very happy about it. We haven't tried to voice any revenue projections on it, although I will tell you that customer periodically is a very strong customer for us.

All right. And then more generally, the [passive] unit growth and some number of opportunities abroad of various sizes, ranging from significant to perhaps less than significant. Can you give us an update in terms of opportunities that are out there that remain to be awarded and could be awarded over the next year, putting aside whether or not you win?

Yes, there are quite a few. I don't want to necessarily list them here for obvious reasons, but I would venture to guess most of them -- by far, most of them have not been awarded.

The last look I had, we were actually at this point in the quarter -- early in the quarter we were well ahead of 70. My guess is we're probably closer to 90 now of different trials going on with G.fast. So it's just very early in the cycle.

And then it's not just G.fast. I've mentioned in my notes NGPON2, where we are really making some great progress on the development side. Without a doubt, we have the most technically advanced product in the market, not just NGPON2, but in the access space in general. So there are a lot of opportunities out there and I would say a very small minority have actually closed.

It's obviously a huge number, but of those, how many of those would be potentially significant?

You know, you kind of asked a similar question last quarter, and I heard that. I answered it. So I think you know. Most of -- I shouldn't say you do know, but most of the companies are fairly public on what they are doing on G.fast. So there are enough for us to be excited about.

And then when you say less than significant, when you are talking about a Tier 2 carrier, two of those can really move the needle. So there are just enough of them out there that we are feeling really good about the access space right now.

All right. One last quick question. Can you show us what US product revenue did for the quarter?

I don't know if we did -- do we break out US in --

We haven't broken that. We broke out -- we break out obviously the product and the services and we break out (multiple speakers) --

Let me just see what we've broken out and we may -- we can follow up and see what we have broken out.

All right, I'll take the rest off-line. I appreciate it. Thanks, guys.

Simon Leopold, Raymond James.

This is Victor Chiu in for Simon Leopold. I just wanted to circle back really quickly on the international piece.

From your previous comments, it seemed that you applied implied that the growth from that customer wouldn't come back until 2017. And that's kind of how we forecasted and kind of built out our growth, in our model, at least. So I guess my question is how does the strength this quarter impact your previous thinking for 2017 growth from that customer?

I still think 2017 is going to be stronger. What we've seen here and we've had footprint awards historically, and what we've seen them do is to basically accelerate in this quarter and we'll see some next quarter as well accelerate into that existing footprint.

The incremental piece that we'll have next quarter -- and that can be lumpy. So you are absolutely right in that coming into Q3, we were less optimistic. And then as happens periodically in our business, the customer changed their mind.

Next year, though, we have an incremental piece, which is we have aggregate growth in the footprint that we are going to be deploying into. And this is kind of greenfield for them in relation to these products. So that will be on top of whatever we would historically have done.

Okay, that's helpful. And I just wanted to ask about the domestic portion. It seems that on the opposite side of that, had you not seen that international piece come back this quarter that domestic would've been quite weak.

So can you kind of give us a little more color around the specific -- which things in particular were the biggest drivers for that weakness? Was it primarily the G.fast trial?

I don't think I would characterize it as weak. From where we came into the quarter, we did better than we thought. And that better than we thought --

For the domestic piece you did better than you thought?

For the total Company. Right? And that better than we thought coming into the quarter on a revenue line, without a doubt. And that better than we thought coming into the quarter was driven predominantly by international. But that's not saying domestic didn't come in very close to where we were at.

Okay. That's helpful. Thank you.

Bill Dezellem, Tieton Capital.

Couple of questions. First of all, in the press release, you made a comment that ADTRAN's capabilities is proving to be a catalyst for change within the industry. Would you please expand on that comment?

Sure. So our Mosaic -- and I'm really speaking predominantly in that comment about our Mosaic architecture. So when I say we're leading the industry in kind of in an SDN-defined -- SDN platform, I'm serious. We have done a significant amount of work in virtualizing a lot of our functions and using SDN control and orchestration.

And customers want that, but not all customers are ready for that. And we've been able to build mediation layers that allow them to actually move forward with deployment under their existing OSes and get themselves ready to a transition to a true SDN-controlled network. And that's been very resonant -- resonating with our customers.

Now, there are customers where it doesn't -- it's not always -- there are customers that just flat-out aren't ready. And those, we will just sell our traditional EMS. And that's what we'll go to market with and that's what we'll actually compete with.

But I would say there's a large percentage of them that are ready to make this jump and they see us as an enabling platform. If you are in an RFP today and you are looking at G.fast or if you are in an RFP today and you are looking at NGPON2, you can do it the old way or you can do with the new way. And right now, we're making a lot of headway getting people to do it the new way.

That's helpful, thank you. And then I'd also like to tackle the gross margin, but from a slightly different angle.

The service gross margin was down sequentially and yet, the revenue piece was up. And so that seems a little counterintuitive when we would thought you would've had some economies of scale there.

You are right, but there's two things that affect that. One is mix. So we do engineering work. Honestly, engineering work typically comes in hindsight more at the front-end and it makes logical sense.

So you do a lot of engineering work upfront, which is typically higher gross margin. And then you do the installation work towards the back end of the quarter, which is lower gross margin. So I think we probably saw -- I know we saw an impact because of that mix.

And then when you are growing as fast as we are growing our services business today, you are not as efficient as you should be. And there's no doubt that we have some work to do on improving the efficiency as we scale up. So higher revenue in and of itself until you get a model that you can flow through and scale up is not necessarily a driver for higher gross margin at this point for us.

Great, thanks. And finally, I'd like to shift to the Israeli announcement. Would you provide a bit more detail beyond kind of what you just generally alluded to there in terms of not only size, but why you are selected. And I think it's a little bit unique for you to be the prime and what the implications are to that as us as investors?

I hate to call us being prime unique, because we do that in several of our customers, including some that we've talked about on the call today.

But one I think -- one of the big drivers, of course, was the Mosaic access architecture that I talked about, the SDN-controlled access architecture. So we have a customer there that very much wants to move their network forward and move it the way that they operate forward and be able to benefit from a next-generation design.

Our G.fast products themselves, which will be controlled under this architecture, we do have some differentiation. We're kind of agnostic in the chipset, where most people have chosen one or the other. And that gives carriers some comfort as to being able to have more control over where their roadmap goes.

And we do our normal thing, which is we try to just eke out every bit of performance we can out of the product. But I would say that the big differentiator for us was the Mosaic SDN piece.

Israel clearly has a high technology component, which I would presume would imply that they demand a high level of bandwidth. As we think about the size of that contract or business with them, how does that compare, say, with your vectoring customer in Europe?

The vectoring deployment in Europe with that customer is very broad. So I would not put it on par with that. I think that there are -- what we have seen before with this customer is that there are periods of time where we are hit with a lot of business and that may roll up into a particular quarter, but I wouldn't say that's the same size as what we're doing with vectoring.

Vectoring in and of itself I think is just kind of undersold. It's just now picking up at other carriers within Europe. Super vectoring is just starting to be deployed -- will start being deployed really next year.

And vectoring in the US has been nonexistent up until the Salt Lake City trials. And then you saw that carrier move forward in an accelerated fashion this year, which has benefited both our product and services. And they are moving forward aggressively next year as well.

Not only that, though, we have carriers in the Tier 2 space that for the first time are really pivoting and looking at that technology because somebody in the US has been able to make it work and has shown that it's actually economic to deploy and has actually got a sticky customer rate.

So G.fast is a great technology and we're very much behind G.fast. But I do think sometimes vectoring and super vectoring are maybe -- don't get the credit that they deserve. And really we are kind of pivoting towards the real sweet spot in vectoring in 2017.

Thank you, Tom. I actually said that was going to be my last question. But given what you just said, would you please highlight to us where you think vectoring is most appropriate versus where G.fast would be most appropriate? And why a carrier would deploy one in a certain situation and not another?

Sure. So if you want to deploy in the access network and you are trying to hit a bunch of home -- so you can use either technology, but the cost differential is pretty big. Vectoring -- the reach of vectoring is multiples of the reach of G.fast.

So let's say that you want to deploy a 100 to 200 megabyte network. Vectoring and super vectoring are -- do great. And the cost benefit of being able to deploy that architecture to get 100 to 200 megabytes to a customer beats anything else that's out there.

If you are trying to get to a gigabit of speed, that's where you look at G.fast. And of course, because of the range limitation, you are going to have to build that fiber much closer to the house or the MDU, which is why initial deployments are pretty much focused on MDUs because of their access and where those locations are.

And then as you try to move a gigabit to more single-family housings, you'll see it play into there. But it's really -- it's much more cost -- it's still cheaper than fiber-to-the-prem, but it's still substantially more costly than super vectoring is.

Does that answer your question? Okay.

It does. Thank you, Tom.

Tim Savageaux, Northland Capital.

First, just a real quick one. Not sure I heard you mention 10% customer exposure in the quarter. I wonder if you could cover that, and maybe note how many in US versus international or what you are comfortable disclosing.

I did mention that in the comments. There were three 10% customers: 2 domestic and 1 international.

It's what I would've guessed and sorry for missing that. To follow up, I want to go back on kind of the US versus international front, and especially if you expect international revenues to kind of hang in at this elevated level in Q4.

That is implying kind of the first year-on-year decline in your domestic business that we've seen in quite some time. And perhaps even more notable if you affect for services as well.

So I just want to focus back on that and see whether there is anything kind of developing from a trend standpoint with regard to the US market, whether this is just sort of a balancing out of strength earlier in the year. Or if you can talk about the dynamics and maybe outside of the CAF II and services stuff in the US market in general.

I think the premise there is that we started with maybe the confusion, because we've never said that the international business was going to be flat Q3 to Q4. And I'm looking at the chart. I would be very surprised if the US business was down on a year-over-year basis. So I'm just not coming from that same space.

We expect the international business to be stronger than we may have guessed in Q2. But I'm not expecting it to be sequentially flat. So I just think in terms of [normal pickup in business].

That helps a lot. If I can just -- since that was a quick one than expected, maybe I can --

I'm just not where you are, so that's -- I mean, that's --

Yes. Maybe I can toss one more in there, which is back on the kind of cable MSO-focused business and the acquisition. I assume the revenue contribution for Q4 is not zero, so I think probably low single-digit millions is something I would assume.

And in general, I wonder if you can describe the growth opportunity for 2017 in the cable space. How material that is, how material a portion of your current business that is. And that will be it for me, thanks.

All right. So the low-single-digit millions is the right way to think about it for Q4. The problem is is I don't have confidence in our forecast because the business is relatively new.

The amount of business that we do in the MSO space has predominantly been CPE business. It's been the stuff from our enterprise product set. So this is kind of -- I wouldn't say the first, but this is a material infrastructure change or the leader infrastructure -- more infrastructure products.

So it will impact that in a material way. We could be looking at tens of millions next year, but I think it's too early for us to try to forecast that in a firm way to you. We'll give you some color on that, though, on the next conference call.

Thanks.

Doug Clark, Goldman Sachs.

Just a follow-up actually to that first line of questioning there. On the domestic business, maybe not looking at fourth quarter, but in third quarter, it came down sequentially. And I'm just wondering how to reconcile that with kind of the building tailwind of CAF.

So assuming CAF and services did grow, which pieces declined in the third quarter on a sequential basis in the US or domestic?

Yes. So if you look at the product set, you will see broadband did well. We had traditional products, which is legacy -- what we used to call legacy products, but stuff like HDSL and some -- we do some powering shells and things like that. And we did see a material decline in that business.

But that's not a surprise. That's a very lumpy piece of business there. And if you look out a year over year, there's kind of some one-offs that happened in that business versus last year that made it materially different. That was the biggest piece.

The CPE business -- in general, it was down sequentially, although it's up fairly strong over the entire three quarters. We expect it to be up in Q4. But traditional products and a slight downtick in the CPE business.

Does that answer your question? I hope I --.

Okay, that's helpful. Yes, that does. And then one kind of model-related question. What would the non-GAAP tax rate have been kind of excluding the purchase gain?

It would have been 25%.

Okay. And just to be clear, that came in kind of below kind of your expectations of mid to high 30s% because of the higher international mix, is that right?

Yes, two things. A higher international mix, and as I mentioned in my notes, there was some benefit from an updating of our R&D tax credit outlook.

And that was a one-time item, meaning not to recur in the fourth quarter based on your guidance?

We are not projecting that yet in the fourth quarter. We're still working on that for the current year.

Okay. Great, thanks.

Okay. I think that was our last caller. So thank you, everyone, for joining us on our call. And we hope you have a great fourth quarter and we look forward to talking to you again at this time next quarter.

This does conclude today's program. Thank you for your participation. You may now disconnect.