Gogo Inc (GOGO) 2017 Q1 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Gogo Inc. First Quarter 2017 Earnings Conference call. (Operator Instructions) As a reminder, this conference call is being recorded. I would now like to turn the conference over to Ms. Alva, Vice President Investor Relations and Treasurer. Ma'am, you may begin.

Thank you, and good morning, everyone. Welcome to Gogo's first-quarter earnings conference call. Joining me today to talk about our results, are Michael Small, President and CEO; John Wade, Executive Vice President and COO; and Barry Rowan, Executive Vice President and Incoming CFO. Immediately following the filing of the company's quarterly report on Form 10-Q, Barry Rowan will become Gogo's Chief Financial Officer.

Before we get started, I would like to take this opportunity to remind you that during the course of this call, we may make forward-looking statements regarding future events and the future financial performance of the company. We caution you to consider the risk factors that could cause actual results to differ materially from those in the forward-looking statements on this conference call. These risk factors are described in our earnings press release and are more fully detailed under the caption Risk Factors in our Annual Report on Form 10-K and other documents we filed with the SEC.

In addition, please note that the date of this conference call is May 4, 2017. Any forward-looking statements that we may make today are based on assumptions as of this date. We undertake no obligation to update these statements as a result of new information or future events.

During this call, we'll present both GAAP and non-GAAP financial measures. We include an explanation of adjustments in other reconciliations of our non-GAAP measures to the most comparable GAAP measures in our first quarter-Earnings Release.

This call is being broadcast on the Internet and is available on the Investor Relations section of Gogo's website at ir.gogoair.com. The earnings press release is also available on our website.

After management's remarks, we will host a Q&A session. And now it's my pleasure to turn the call over to Michael.

Thanks, Varvara. Good morning, everyone.

We had another outstanding quarter with record revenue of $165 million, an increase of 17% from Q1 2016. Service revenue grew 16% in CA, North America, 30% in BA and more than doubled in CA-Rest Of World. CA-Rest of World service revenue was bolstered by rapidly growing passenger take rates. Our record results in the BA segment were driven by continued strength in both ATG, aircraft online and ARPU.

We also had another important 2Ku win during the quarter with Virgin Australia. Our technology is already on board Virgin's aircraft and is outperforming the competition in that market. We continue to see momentum in our sales pipeline coming out of last month's Aircraft Interior Expo in Germany.

Moving beyond our recent news, I want to address questions circulating in the market about our satellite capacity. Let me be clear, we have more than enough capacity now and for the foreseeable future. And we have the most capacity -- satellite capacity dedicated to the aero market.

We also have the lowest bandwidth cost structure in the industry for the following reasons. First, our 2Ku antenna is far more efficient than any other competitors', due to its proprietary design, which yields nearly twice as many megabits per megahertz of capacity.

Second, our superior capacity utilization, due to our ability to lease capacity where, when and in the quantity needed. And third, our ability to capture performance enhancements and cost reductions from satellite innovation over time. This is accomplished because of our open architecture, multi-source leasing strategy.

We have access to dozens of satellites positioned at key points on every continent. We have signed deals with 9 satellite providers, including our most recent agreement with SES to lease all the capacity on AMC-4. With this new agreement in place, SES will actually move the satellite so we can better serve flights over the Pacific and to and from Alaska and Hawaii. This flexibility, depth of capacity and redundancy does not exist in the Ka-band.

It is far more cost effective to lease satellite capacity rather than making large and long time -- long-lead time investments in owning a few satellites. Our leasing strategy is becoming more compelling each year as the rate of innovation in the satellite industry continues to accelerate. And because of 2Ku's open architecture, we offer our airline partners technological advantages generated from the billions of dollars that companies like SES, Intelsat and SoftBank are investing in satellite technology. 2Ku's compatibility with the most satellites across multiple providers is the best way to future-proof an airline's in-flight connectivity technology decision.

Our success-based leasing strategy is needed to cover high air traffic cities like London, New York, Chicago, San Francisco, Atlanta and Seattle. These areas cannot be serviced by 1 or 2 global satellites. They require multiple satellites with both wide-beams and spot-beams. Likewise with runway secured cost-effective capacity for more distant destinations, such as Honolulu, Anchorage, Sao Paolo and Sydney, that are not at a complete covered with Ka.

Add it all up, and 2Ku is delivering the best performance in the industry, characterized by 3 numbers: 15, 98, 98. This means 15-plus megabits per second speed to connect the passengers; 98% coverage of global flight hours and 98% service availability. This is the performance we are delivering today to 2Ku aircraft around the world. We are driving continued improvements across all 3 performance metrics. With the new modem in HTS coming this year, we will see rapid improvement in speed on our satellite network.

Wrapping up, we're delivering what airlines care about: 15, 98, 98, which means high speeds everywhere they fly, all the time. With that, I'd like to turn it over to John Wade.

Thanks, Michael.

During the first quarter, we signed an agreement with Airbus, to offer factory installations on the A320, A330 and A380 aircraft. 2Ku is the first next-generation in-flight connectivity technology to be selected by Airbus for factory installations. Our airline partners can now talk directly to Airbus, all with 2Ku.

The first A350 factory retrofit should take off later this year. Starting in 2018, they expect to see -- begin seeing OEM factory-installed aircraft. This new agreement provides a path towards OEM installation on all of Airbus aircraft. To put this in perspective, there are more than 17,000 Airbus aircraft on order today. And more than 13,000 are from the A320 family of aircraft. This designation opens up significant new markets to us.

Also, 2 weeks ago, it was announced that we will install 2Ku on Delta's new Bombardier C-Series aircraft at the factory.

During the quarter, we also made progress in 2Ku installations. We now have completed more than 170 aircraft installations on 8 airlines, including Aeromexico, Air Canada, British Airways, Delta, GOL, JTA, Virgin Atlantic and most recently, Virgin Australia. As we mentioned in our Q4 call, the pace of installations this year will mirror that of last year, a slower installation pace at the start of the year with a significant ramp up after Labor Day.

Certification in aircraft engineering capabilities are a challenging, and often-underestimated aspect to succeed in our industry. We are obtaining the additional STCs we need. We recently received 2Ku STC for single-aisle Airbus, A319s, A320s and A321s, doubling our STC coverage to around 2/3 of the world's 2Ku applicable aircraft. Given this, and other STCs, we expect to hit our guidance of 450 to 550 2Ku-installed aircraft by year-end.

With every new 2Ku aircraft taking flight, our brand experience is a resurgence, as airlines and passengers can now see first-hand the incredible Internet experience that ample capacity provides.

Turning to Business Aviation, as Michael mentioned, we had another outstanding quarter in our BA business. I'm excited to announce that Gogo Biz 4G is flying on our Challenger test aircraft. Our 4G service launch is expected later in Q2, on track with our expectations. And STCs are in progress to satisfy the many orders we've already received for the Gogo Biz 4G products. As an update on our previously announced next-generation ATG technology, we just, this week, achieved speeds of 134 megabits in the lab. We're on track for deployment in 2018.

Now I want to turn it back over to Michael, to introduce our new CFO, Barry Rowan.

I want to welcome Barry to Gogo in his first Gogo Earnings Call.

Barry has extensive experience in Finance and Corporate Strategy, and a long track record of value creation. He served as a CFO at 3 public companies, including Vonage and Nextel Partners. What distinguished Barry from the other candidates we interviewed, was both his financial acumen and his ability to be a strategic partner to me and the rest of the leadership team. I also want to thank Norm Smagley for building a world-class finance organization that will serve as a solid foundation for Barry, moving forward. Now I'd like to turn it over to Barry to run through the Q1 numbers, Barry?

Thank you, Michael, for the kind introduction. And good morning, everyone. I am thrilled to be joining the Gogo team at this very exciting time for the company and the industry.

Let's get right to the numbers.

We had a great first quarter. Total revenue was up 17% to $165 million. Service revenue grew 23% to a record $146 million. Adjusted EBITDA of $11 million included $9.4 million in incremental spend for next-gen ATG development. Our adjusted EBITDA margin, excluding the ATG milestone payment, reached 12%, up 200 basis points from last year. We are reaffirming all of our 2017 and longer-term guidance, which can be found in our earnings release.

Now turning to segment results. CA-North America service revenue increased 16% to $97 million, driven by an increase in aircraft online to over 2,700 and stronger ARPA. For the quarter, ARPA grew 6% to $142,000, driven by 14% growth in passenger connectivity revenue and more than 70% growth in passenger-entertainment and connected-aircraft revenue.

Our connectivity take rate increased to 8.3%, up from 6.5% a year ago, demonstrating the benefit of the ATG offload from 2Ku and the potential of the multi-payer strategy. We expect ARPA growth to continue throughout 2017, driven by: first, a significant increase in available bandwidth as we upgrade North American aircraft to 2Ku; secondly, the continued expansion of the multi-payer strategy; and thirdly, a stable regional jet count.

CA-North America segment profit of $11 million included the $9.4 million spend for next-gen ATG development I mentioned earlier. The segment profit margin excluding the ATG milestone approached 21%, up nearly 500 basis points from last year. As discussed on the fourth quarter 2016 call, we expect a total of $20 million of incremental expense in this segment in 2017 for the next-gen ATG development.

Now turning to CA-Rest of World, total revenue for the quarter more than doubled to $10 million, driven by strong growth in ARPA and aircraft online. Aircraft online increased 281, up 44 versus the prior year.

We continue to expect roughly 150 2Ku installations for ROW in 2017. Our CA-ROW 2Ku awarded, but not yet installed aircraft, was approximately 650 at quarter end. For the quarter, we generated annualized ARPA of $202,000, up 45% from the prior year, driven primarily by higher airline paid and third-party paid usage. While we expect some ARPA dilution later this year as new international partners launch service, it is clear that abundant capacity and the multi-payer strategy are already driving strong growth in revenue per aircraft.

Rest of World segment loss for the quarter, increased to $27 million from $20 million in the prior-year and was up $2 million, sequentially. Although quarterly revenue was strong, it was offset by the planned increase in expenses, primarily for new airline launches. We continue to expect 2017 to be the peak loss year in this segment, with improvement in 2018 as ARPA growth and increased aircraft online drive better ROW service margin.

Now turning to Business Aviation. Service revenue for the quarter was up 30% to a record $40 million, driven by 18% growth in ATG aircraft online to over 4,300 and a 12% increase in ATG service ARPU to nearly $2,800 per month. BA equipment revenue of $16 million was down $3 million from the prior-year but up $1.2 million versus the prior quarter. We will start to recognize some of the $5.7 million of deferred revenue under the 4G sign and fly program midyear. As John noted, there's great customer interest in this product and we look forward to its commercial launch this quarter.

Segment profit was up 29% to $26 million, representing a 46% segment-profit margin. This is up 6 percentage points from last year, driven by 30% growth in high-margin service revenue, which represented 71% of BA revenue for the first quarter.

For the company as a whole, first-quarter consolidated cash CapEx of $59 million was $35 million higher than the prior year, due to increased airborne equipment purchases to support 2Ku installs. We ended Q1 with $439 million of cash on hand. This includes $68 million in net proceeds from the issuance of additional senior secured notes in January. Q1 has seasonally high working capital, driven primarily by the timing of interest payments. The cash used in the quarter is consistent with our full-year expectations.

To wrap up, our strong financial performance in the first quarter coupled with increasing 2Ku installation capabilities have made for a great start to 2017. We expect the momentum to continue.

To end on a personal note, I wanted to say how much I look forward to working with you, our investors and analysts, in the months and years ahead. I know many of you from my previous CFO experiences and I look forward to hearing your perspectives on Gogo in the coming weeks.

Operator, we're ready for our first question.

(Operator Instructions) Our first question comes from Ash Birla with Dougherty and Company.

Can we talk about the next-gen ATG spend? When does that $9.4 million -- how much is left to go and how much does that come off later in the year?

Yes, Ash, it's Barry. As we said on a previous call, we expect the next-gen ATG spend to be $20 million in OpEx, of which $9.4 million was recognized in the first quarter. The rest will be spread throughout the year, but in much smaller milestones. On the CapEx side, there is about $50 million, which we have not incurred any of -- any yet, but that will begin later in 2017 and spread from 2017 to 2019.

Great. Thanks, Barry, and welcome to Gogo. I just have a few follow-ups. Michael, SmartSky raised 170-some million dollars. Have you guys seen them, at all, in the supply chain, in the dealer network, at all?

Ash, I'm going to throw that one to John Wade.

What we're seeing really, in the supply chain right now, is a continued strong interest in Gogo Biz 4G. As we've mentioned before, we've got a significant order book already for that product. And I will say, right now, all our views are that we're going to have a very strong year with 4G.

John, just to follow up little bit on that. I mean the -- as you know, we have talked to them, it's been 4 years since they've been saying they will build a network. Do you guys have any color, and I know that is a question probably for them, but if you can provide any color, where you see -- you have 200 towers, how many towers do they have, and where do you see them? And do you think that the $170 million is enough to build an ATG network?

I think in terms of where their network will build out, you really should probably ask them on that. I think that our view in terms of what it takes to build and operate a -- I don't think our networking requires more money than that, we can say that based on our own experience. I think we can also say that with the developments that we've seen, as recently as this week, as I mentioned during the script, we're already seeing performance in excess of 100 megabits a second in the lab. If you couple that performance with the license spectrum we have with our own ATG solution, we really think that this is going to be the superior solution. And we think the market is going to vote with its checkbooks that, that's the one it's going to want to go with.

And just one last one, Michael, maybe 2 last ones. If you have 136 megs, why do you need 2Ku in North America? And the second one is Alaska. Did we see a little bit of different tone in your voice, like I think when you mentioned Alaska and Hawaiian routes? I mean, have you -- can you share what they are telling you, because, I know that they have an RFP out? So -- and does Telus, does their network cover Alaskan routes?

So, 2 questions. With over 100 megabits per second on a next-gen ATG, why 2Ku, and it's ultimately coverage is fundamental. Most mainline aircraft fly to the Caribbean or Mexico or Hawaii, and so to maintain coverage, need this satellite solution. Some TV and gate-to-gate solutions -- there is also benefit to satellite versus air-to-ground. Of course, air-to-ground is great for smaller aircraft with smaller antenna. But we will match the speed on air-to-ground with 2Ku.

On the second side, on Alaska, again we don't comment on individual airline negotiations, although we do acknowledge as Alaska said publicly, they are looking for a satellite upgrade. To make this clear in the script, we have the superior solution, and particularly to Alaska and Hawaii, we have rock solid coverage there; we just moved AMC-4 -- we just contracted for AMC-4. So we'll have tremendous coverage over the Pacific, including routes to and from Hawaii and Alaska.

Our next question comes from Philip Cusick with JPMorgan.

First on, I guess following up on next-gen ATG, what gives you confidence, Michael, in launching that in 2018? And should we think of that as a launch for your test bed plane or commercial launch, when do you start spending for STCs for that? And can you talk about the pace of OpEx from here? Obviously the $9 million, I think John already addressed this -- of 20, it sounds like it's going to slow down quite a bit from here? And then second, Barry, can you talk about the due diligence you did on joining Gogo before coming in, and what got you excited about coming on board?

So our next-gen ATG, it will be a commercial launch in 2018. What gives us confidence is we've been working on the technologies that underlie this for a decade. This isn't a new effort for us, we announced it last year. And so far we're hitting all our milestones against that effort, right on time and right on budget.

And Phil, to your question about the due diligence I did and the reasons for my joining, and it's great to be working with you again. I did a lot of due diligence, I wasn't actually initially interested. I wasn't looking for a position, I was quite happy.

Then as I had initial conversations with Michael and with the board, I could see the opportunity here. So I did -- a lot of my due diligence had to do with the economics, particularly the fundamental customer economics.

I would summarize my reasons for joining in 3 areas, in response to your question. You asked kind of a personal question, so I'll give you a bit of a personal answer related to the business, the people and then what I think is my ability to contribute.

On the business side, it's a business I can get very passionate about. As you know, I've been helping build technology companies for over 30 years, and around 20 years in communications-related space, and I find this very interesting and exciting business. But more fundamentally, I think, and importantly, I should say, the fundamentals here I think are very strong. Aircraft connectivity is going to be a given, people are going to want to connect in the aircraft, the penetration levels are low and with the increase in number of planes over time, that makes for a great opportunity and Gogo has the leading market position. So I think the fundamentals which often carry the day at the end, are very strong.

I think it's also a good time to be joining Gogo, with the new technology in 2Ku coming out and the next-gen ATG, the customer economics are very compelling there, as I looked at those and really got underneath the covers there. So I think that is -- makes for an exciting time to be here. And also, I think the evidence in the market of the receptivity of 2Ku, based on the backlog, I think meaningfully derisks the company over the next couple of years as we aggressively install that backlog. So I think with that significant market opportunity, Gogo's leading position and a strong backlog, I think the company is in a good position to execute on those compelling customer economics.

And Phil, to pile on here. I can say -- I didn't say it in the script, one of the primary reasons I wanted to hire Barry is he actually asked me more questions than I asked him during the interview process -- he wanted to get to the bottom of it.

So I did a lot of that, including kind of developing my own investment thesis, looking at the company from that standpoint. So the business was clearly an attraction at this time, in particular.

I think secondly, on the people side, from my first conversations with Michael to meeting the board, the rest of the leadership team, it strikes me as a team with a high degree of competency and a deep commitment. People are very passionate about making this work, and very committed to working through the inevitable ups and downs of building a business like this. And I think there's also a deep sense of character and integrity, that I found myself feeling like I could really feel some chemistry here and be an integral part of the team.

And then finally, in terms of my ability to contribute, that's very important to me, I feel like I could make a difference. This is the fourth public company I will have served as a CFO for. I think the last 2 are particularly relevant, in that they were companies with strong fundamentals that frankly were under varying degrees of significant pressure, in the case of Nextel Partners and Vonage. And at Nextel Partners we were able to grow that from breakeven EBITDA to $750 million in the 4 years that I was there; and Vonage, we were able to achieve a comprehensive turnaround, that is now still being executed by the team that's still there. So I feel like all the experience that I've had will enable me to dig in and be helpful to the company. So hopefully that gives you a sense of both the reasons and the level of my excitement for being here.

Our next question comes from the line Robert Gutman with Guggenheim.

ARPA in the ROW segment got a big boost last year, and it remains at a high level, continues to actually get a little higher. Aside from launches that could cause temporary headwind there, is this level sustainable? Or are there any temporary sponsorship programs supporting this level?

This level is sustainable. When you have the bandwidth, you can grow ARPA. And the primary driver is Japan Airlines moved to an airline-paid service for the entire flight. And we saw rapid, rapid increase in its passenger take rates and ARPA is a result of that; it's a fundamental driver here. And you'll see similar such programs appearing at multiple airlines going forward, and we now have the capacity available to accommodate those.

Later in the year, as we launch some of the additional airlines, like British Airways and Air France, those new planes will dilute the numbers, but the existing aircraft, this is highly, highly sustainable. Growable too.

And one follow up. If you, on the Airbus line-fit approval, does that apply to any of your existing customers in terms of their future deliveries? And is it opening up opportunity, or has it yet opened up opportunity for negotiations with new customers?

In answer to the first one, no. All of our current backlog is OEM retrofit or true retrofit. In terms of opening up opportunities with current and existing customers, absolutely. It's actually, I think, a very significant moment for Gogo, in terms of true life operability on both Airbus and Bombardier. I think you can expect to see additional customers taking the opportunity to take their market-leading performance of 2Ku on their factory-delivered aircraft.

Our next question is from Simon Flannery with Morgan Stanley.

This is Landon Park on for Simon. Just wanted to start off with the 2Ku installs. It seems like you guys installed about 20 in April; is that a good run rate to use until the ramp in September? Or how should we think about that?

Month-to-month, there's going to be variability based on airline, aircraft availability. But in even things like Easter Weekend in April, actually affects the aircraft availability and route planning. So I wouldn't think -- look at any particular months right now as indicative of what's going to happen post-Labor Day when the airline, aircraft are traditionally available for installation. So it is going to be bumpy here over the next few months, but really look for more -- the end of Q3, Q4 for the run rates that we're expecting to achieve. The issue -- sorry, the thing we needed to make sure we have in place is STCs and we're very comfortable we're going to have the STCs in place in time for our Labor Day kickoff.

That makes sense. And then looking at CA-NA, just the -- can you take us through what you guys are expecting on the take rate? Obviously, very solid number there, and do you think that the 6% ARPA growth that you saw in the quarter, is that, a sustainable run rate? Or how do you think that will trend throughout the year?

The 6% is indicative of where we stand now. As we bring more bandwidth, we're going to have room to accelerate ARPA. And we're just beginning to get the bandwidth benefits of the 2Ku planes, so we announced we've now crossed 100 planes installed with Delta, and those were all upgrades from ATG, creating more capacity on those planes clearly, but also offloading the ATG network. So it's what we're doing now and actually, as we keep the 2Ku going at some point, we should be able to even improve the 60% ARPA growth.

On take rates, which, I'm sorry, I missed the other half of your question. They will keep growing, because we have the capacity to take on more customers and begin segmenting the market beyond just going after the hard-core business traveler. We're now able to find ways to bring in the leisure traveler too.

And just one last one, just on the SES deal. I believe they indicated that they would begin recognizing revenue towards the end of second quarter. Just wondering how we should think about that, on the expense side for you guys?

Yes, the SES capacity there is consistent with our plan, and it's consistent with our objective to get to over 50% service margin in rest of the world, as we're doing today in North America. And don't forget our capacity is highly portable, so if we fly more in 1 place, we can reshift the capacity that we have, particularly with our Intelsat and SES contracts. So we get to move it around. This capacity is highly fungible, globally.

Our next question comes from John Hodulik with UBS.

Maybe a couple of questions on some of the new capabilities. First Michael, on the new modem. Could you remind us what the difference in, sort of, speeds or performance you expect? And how quickly will this sort of make its way into the base, if it's not -- hasn't started already? And then similarly, on the next-gen ATG side, I imagine once it becomes available it'll go into all new installations, but how should we see this sort of penetrating the existing base of both commercial and business users?

Sure, John. So the new modem, it's actually -- the production model of the new modem is on the Jimmy Ray, our test aircraft; and it removes the bottleneck the modem had previously been, and so we're now seeing in the zip code of 70 megabits per second on wide-beam satellites and over 100 megabits per second on spot-beam satellites. And that will go into the marketplace second half of this year, we'll begin installing them in actually at a quite rapid rate. And that new modem not only benefits 2Ku, it'll benefit the Ku planes.

One additional benefit of the new modem is on the reverse link; it is the slower link for everybody, but we get real improvement on the reverse link, and that's going to show up in a very real way in the passenger experience, and passengers have really, really noticed that.

Next-gen ATG won't go on every aircraft, well even the ATG aircraft. It'll likely go on the regional jets and the larger business jets, some percentage of larger business jets. The antenna is about 30 inches long, it's like a low -- only 4 inches tall and 4 inches wide on the bottom -- it goes generally on the bottom of the plane. And so those planes will now have the ability to get the very fast service, and also those planes, again, will further offload the existing ATG network to the extent most of their traffic does. So 2Ku and next-gen ATG, good in their own rates, but tremendous offloads of the ATG network.

One quick follow-up, if I could. Agreed with the last question that the 8.3% take rate was stronger than we expected. Can you talk about how you're bringing in these -- the sort of nonbusiness customers? Is it just rejiggering the price? Are there other sort of partnerships? Just as similar to what you said, what you have a T-Mobile, on -- in store for us, or how are you going after that segment of the market?

We go after this in conjunction with our airline partners and it varies by airline, and sometimes we do it in conjunction with third parties.

The 2 leading reasons that it's going now, the take rate's going up, is there's T-Mobile, there's a significant number of new customers coming into the franchise. We have very few T-Mobile users as existing customers, and so when they offered it for free, those were virtually all incremental new customers.

On Alaska Airlines, we introduced the messaging app, so basically primarily for i-Message is the predominant use and that's bringing in some additional users.

In addition, the dilution from adding a whole bunch of new RJs ended, for all practical purposes, so we weren't diluting the take rate that way. So those RJs are continuing to grow, and the main line are continuing to grow, but there's no dilution of new planes anymore -- new RJs.

Our next question comes from Lance Vitanza of Cowen and Company.

Just a couple quick ones. The first, can you given any update on the 2Ku install time? I might have missed that during the prepared remarks.

Sure. Well we continue to set the record here and we have now achieved installations with experienced crews of under 3 days, which as far as we know, is by far the fastest installation of our large radar systems. So it's going great.

Great. And then on the AMC-4 capacity deal, I know -- was it all capacity, or was it all remaining capacity on that satellite that you locked up?

All capacity.

Okay, great. And then any update on the American Airlines, the install? I know you guys talked about that at length on other calls, but are we still just thinking about it the same way?

Yes. We are predicting that we'll start to see the first in deinstallations happening late this year and then continuing over the next few years. And that's the way we're looking at that.

Our next question comes from Andrew Spinola with Wells Fargo.

I wondered if you could just give me some additional color on the cost-of-service trajectory in your commercial business. You've got a very substantial ramp of 2Ku in North America this year, but if I look at the cost of service in that segment it was down sequentially and flat year-over-year, while on the international business it was up $5 million sequentially, and it doubled year-over-year. I'm just wondering why we're not seeing a bigger ramp in North America, and maybe, is that relative to the mix of the different types of cost of service? What are some of the drivers there?

So again, as we emphasized in the script, we have the ability to get capacity -- lease capacity, when we need it, where we need it and the amount we need, which is by far the efficient way to acquire capacity, versus buying a whole bunch up front, by owning a satellite and then having to sell it.

Or having to architect where the capacity may go way in advance in a satellite design years before you know you where you need it. So we have the efficient way to acquire capacity. There'll be a lot of decisions along the way, purchase decisions, but frankly, where cell sites are built on the ground, isn't the question you generally ask operators there.

So I think -- and I also made the point about the fungibility of capacity, also the portability of the capacity. If we buy more like AMC-4, we will now say to SES and Intelsat, "We need fewer spot-beams in that region. We want to light up more in the other areas." So I think that the net result is we're able to match capacity purchases with the planes on a very, very real way.

Michael, I know it's a good point, the fungibility of the capacity and just the nature of these, the new networks where you've got a managed network as opposed to just buying pieces of transponders. Just wondering, in the complexity of that type of accounting, is there any capacity that's serving North America that's being expensed through the Rest Of World segment?

No, we can trace capacity to aircraft quite well and we can trace -- and we have our -- we allocate revenues based on where the aircraft flies, so I -- we think that works very well.

And just to stay on the topic, 600 basis-point improvement year-over-year in gross margin in commercial, North America, was there -- is this a sustainable level, is there any onetime benefit that maybe drove the cost of service down? Anything that you can give on that result?

I -- there are no onetime issues I'm aware of.

Okay. Last question for me, just on the balance sheet. The deferred Airborne lease liability declined $25 million sequentially, I think it's a high-teens rate. That's a number that's been going up pretty -- every quarter, so I'm just wondering what changed there, why did that decline?

I'm going to let Varvara take that one.

Thanks, Michael. Hi, Andrew.

So we did have one of our international customers, specifically Japan Airlines, that basically migrated from one service model to a different service model, and that, with itself, it just changed some of the geography on the balance sheet. Fundamentally, nothing really changes in this particular case other than we will now start recognizing a little bit of the equipment revenue, as well as the cost of equipment, on the face of the P&L. As we mentioned on the fourth-quarter call, that's the elements of our airline-directed model accounting, and that's what's happening with Japan Airlines.

Our next question comes from James Breen with William Blair.

Michael, can you just, or John, can you just talk about what you're seeing in the 2Ku planes early on? Obviously we see the blended number for ARPU and for penetration growth, but maybe something a little bit more specific to the planes that you have up and running at this point?

What we're seeing is a phenomenal customer experience. It's like people, I guess, are used to using the Internet on the ground in their homes, and in their offices. We've seen the entire plane, for all practical purposes, get on it at one time and use it. Streaming, works great. We summarized the experience because we're trying to put rigorous performance metrics around this, and be able to communicate them to you is that 15_plus megabits per second, 98% coverage of global flight hours, and 98% service availability. And not only that, that's what we're delivering today, and it's getting better every day.

Is there any way you can look at, in terms of planes that came online, maybe 2 months ago, that you can quantify the jump that you see in either ARPA or take rates on those planes, sort of over the course of the first couple of weeks of being in service?

That's a -- we do see benefit there. I think what you're going to see is the ARPA benefit is something -- critical reasons we saw 6% ARPA growth in the first part of that wouldn't have been possible without the 2Ku, nor would the 8-plus percent take rate have been possible without 2Ku. So not only do we get to see the benefit on those planes, but we get to see the offload of the ATT network.

But fundamentally, it is very clear to me the capacity, additional capacity, is the key to growing our revenues.

I think the other part -- it's Barry -- I would add to that, is if you look at ROW and the very substantial ARPA grow there, that it's a smaller base, so you see the impact of the airline-directed model there, and really the increased usage as a result of the airline being able to offer free service to the passenger, but still a benefit to Gogo as a provider.

Our next question comes from Philip Cusick with JPMorgan.

I just wanted to follow up on that last question. Can you talk at all about the experience that people who are getting on JAL? What's the pricing look like, what's the usage look like? And I think there was an unlimited video effort there for a while. What are people seeing today?

On JAL, we have gone to an airline-directed model where the airline pays for that capacity, and it causes the accounting change that Varvara described. And we are seeing -- we can't disclose all the take rates due to our agreement with the airline, but we -- they actually fly 777s with 500 seats on very short flights, and that means that by -- almost by definition, everybody's using the service all at the same time. If your flight's only an hour long, it's not spread out.

And we are seeing a tremendous number of users on very large aircraft. And the service is performing admirably, the customer satisfaction's off-the-charts good. So I'll say it again: we're entering an era of much greater bandwidth abundance and that is going to do very good things for our financial statements.

Our next question comes from Ash Birla with Dougherty and Company.

John, in your prepared remarks I think you had mentioned or kind of alluded to that you have a line-fit on Airbus all aircrafts -- all aircraft, and I think when you guys filed the press release the A350 wasn't included, so do you have an A350 line-fit, apart from what you're doing with taking the plane to a different hangar and then putting it?

No, not today. What we said in the prepared remarks was we were going to have true line-fit as part of the production line for 320, 330 and 380 (sic) [A320, A330 and A380]. We are anticipating getting line-fit on the 350 after it comes out of the industrialization phase, but that would be at a later date. In the interim, we have the A350 service-closing approach, and the -- which is their line-fit retrofit, if you will, and the first of those aircraft deliver to Delta late this year.

And Michael, I know a lot of questions have been asked on take rate, just one last one. This 8.3%, were there -- apart from T-Mobile, were there any other promotions, like Google or Facebook or any other promotions going on?

The only other significant change is -- that drove the year-to-year improvement was the messaging service for Alaska, although I will say that T-Mobile was the larger factor of the 2. And then there is some underlying growth and take rates across the base besides those 2 -- I want to say 2 programs.

Alaska is only 5% of your aircraft anyway, so okay.

I'm showing no further questions at this time. I would like to turn the conference back over to Michael Small for closing remarks.

Thank you. We're pleased with the quarter, most pleased that we're starting to see the benefits of the additional bandwidth 2Ku is bringing to bear here, both directly and through the offload, and that started showing up the increased ARPA that you saw this quarter. So thank you all for joining, and look forward to our ongoing dialogue with you. Have a great day.

Ladies and gentlemen, thank you for your participation in today's conference. This concludes the program, you may now disconnect. Everyone, have a great day.