Good morning, ladies and gentlemen, and welcome to the second quarter 2019 Gogo Inc. earnings conference call. At this time, all participants are in a listen-only mode. Later, we'll conduct a question and answer session, and instructions will follow at that time. If anyone should require operator assistance, please press star then zero on your touchtone telephone. I would now like to introduce your host for today's conference, Will Davis, Vice President of Investor Relations. Mr. Davis, please proceed.
Thank you. Good morning, everyone. Welcome to Gogo's second quarter 2019 earnings conference call. Joining me today to talk about our results are Oakleigh Thorne, President and CEO, and Barry Rowan, Executive Vice President and CFO. 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 press release filed this morning and are more fully detailed under the captioned risk factors in our annual report on Form 10-K and 10-Q and other documents we have filed with the SEC.
In addition, please note that the date of this conference call is August 8, 2019. Any forward-looking statements that we 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 the call, we will present both GAAP and non-GAAP financial measures. We include a reconciliation and explanation of adjustments and other considerations of our non-GAAP measures to the most comparable GAAP measures in our second quarter earnings press release. This call is being broadcast on the internet and available on the investor relations section of Gogo's website at ir.gogoair.com. The earnings press release is also available on the website. After management comments, we will host a Q&A session with the financial community only. It is now my great pleasure to turn the call over to Oakleigh.
Thanks, Will. We're pleased to announce a very strong quarter, exceeding our own revenue, adjusted EBITDA, and cash flow expectations. Quarterly service revenue grew in all three business segments and reached $174 million, up 9% over Q2 2018. Adjusted EBITDA exceeded our expectations and grew to $37.8 million, up 100% over the prior year period, driven mostly by service revenue and IBP-related cost savings. There are also a couple quote unquote "good guys" that helped adjust the EBITDA in the quarter, and I'll describe those in just a minute. Free cash flow also exceeded expectations at negative $3 million in the quarter versus negative $35 million in the year-ago quarter, and negative $37 million for the first half 2019 versus negative $144 million in the first half a year ago. I'm very proud of my Gogo teammates for delivering such a great quarter.
We've worked hard together over the past year to improve our operations and achieve our financial goals, and we're on track with both. Thank you very much. I'm going to quickly touch on strategic developments and then dive into the quarter. I'll leave the heavy number lifting to Barry. As I look ahead, I feel very good about where we are strategically. As I've said before, our model is to grow as demand for in-flight connectivity grows, and I think that demand is poised for accelerating growth as airlines increasingly look at providing free IFC to passengers and as the aviation ecosystem looks for cheaper and faster ways to access operational data. A key enabler of our ability to scale and meet this demand on our satellite network is our open architecture, asset-light operating model.
Key enablers of our ability to scale in the ATG world are our proprietary spectrum and our ATG infrastructure, which can provide redundancy and cost savings as we deploy Gogo 5G. Turning back to satellite. Today, we work with 12 satellite providers and use 30 satellites to create a seamless global network. Because we use multiple satellites from multiple operators and operate in the Ku-band, where there are some 200 communication satellites, we can layer capacity where it's needed because 80% of the world's air traffic flies over only 20% of the Earth's surface. We can provide more redundancy than closed Ka-band constellations, which is especially important given the increasing risks posed to satellites by space debris, and we can scale as demand grows with the provision of free in-flight Wi-Fi. Today, Gogo supports two airlines providing free Wi-Fi to passengers.
In May, Delta conducted two weeks of market tests on 55 daily flights of 2Ku-equipped aircraft. In the quarter, we made substantial progress on our plans to ramp up operational support for airlines to provide free Wi-Fi to passengers. A key benefit of our asset-light model is that we have the flexibility to move to new, higher quality, lower cost technologies as they come along in the future. Towards that end, we're working with satellite partners in several new technologies. We're working with a GEO operator on a satellite specifically designed for aeromobility, and more specifically, for efficient utilization with our 2Ku antenna. We're working with other satellite operators on smaller, more versatile software-defined satellites that could vastly enhance capacity utilization. We're working with future LEO providers to see if we could potentially employ their networks to enhance latency and reduce costs.
Given our position as the leading provider of broadband in-flight connectivity, and given that in-flight connectivity is one of the fastest-growing markets for the satellite industry, we feel we are very well positioned to partner with satellite companies as they develop these new technologies. On the ATG front, we made substantial progress on our Gogo 5G product in the quarter. We're partnering with a leading U.S. 5G solutions provider and are nearing completion of the system design phase of the project. We'll be building our network on the latest 5G technology and be able to deliver higher throughput and lower latency for a better passenger experience than other potential ATG products. We're starting to talk about 5G to airlines for their regional fleets and talk to business aviation owners, operators, and dealers about 5G for their aircraft, and we're getting a very positive response.
We remain on track to deliver this product in 2021 and are very excited about the value it could create for our company and our partners. Now let me turn to the quarter. This was our fourth straight beat and raise quarter and our second highest ever adjusted EBITDA quarter, which has led us to raise adjusted EBITDA guidance once again. The biggest drivers of over-performance were improving service revenue at CA-NA, continued cost improvement from our IBP plans, and lower than anticipated Satcom expenses as a result of more efficient network management. As I mentioned before, we did have two good guys in the quarter. First, we had guided to a much lower second quarter adjusted EBITDA, partly because we anticipated American Airlines concluding its shift to the airline-directed model at the end of Q1.
As it turned out, that got delayed until the end of Q2 and resulted in an above forecast $7.5 million contribution to Q2 adjusted EBITDA. A second good guy was associated with the renewal of our contract with American Airlines, which resulted in an additional revenue and adjusted EBITDA benefit to the quarter of $5.1 million. We'll not have the benefit of those $12.6 million in good guys going forward. Q2 was our second quarter of positive combined CA-NA and ROW segment profit and the second quarter of positive service margin in CA-ROW. We made significant progress in reducing our cash burn in the quarter and are reiterating our guidance for at least $100 million improvement in free cash flow for the year, despite having an extra interest payment in the year, and still expect to produce meaningful positive free cash flow in 2021.
In the quarter, we also refinanced our $690 million senior note and the $162 million convertible stub, which pushed the majority of our maturities out to 2024 and has improved our strategic flexibility. We plan to supplement our liquidity with a $30 million revolver, which Barry will discuss in a moment. Given our improving free cash flow trajectory and our improved maturity schedule, we do not expect to need new capital to finance our operations or our strategic investments before reaching positive free cash flow for the year in 2021. I do want to be cautious about Q3 and Q4, however, as we will not get the benefit of the good guys, as I mentioned a moment ago. We will begin to ramp Satcom spend as more 2Ku aircraft come online, as usage grows, and as we ramp in anticipation of significantly more demand in 2020.
We expect to incur increased investments in key programs like Linefit and Gogo 5G in the second half. I'll leave it to Barry to discuss how these trends impact the numbers. Now let me turn to the business segments, starting with some comments on the combined CA segments and then diving into rest of world and NA separately. Service revenue growth was strong in our commercial airline segments, and we are now guiding towards the high end of our prior revenue guidance for both segments. Take rates also grew in both segments over the prior year, and the combined profits of the CA segments were positive and ahead of expectations for the second quarter in a row, despite the deinstalls. In fact, this was the first quarter since the deinstalls began in earnest a year ago that we had growth of aircraft online in the quarter.
Because those deinstalls have now been completed, we expect growth in aircraft online to continue throughout our forecast period. At the end of the quarter, we had more than 1,202 2Ku aircraft online, a net increase of 109, and we had a total fleet of 3,134 commercial aircraft online, an increase of 81 over the end of Q1. Even though we installed more than 100 aircraft to 2Ku, our backlog held steady at approximately 900 aircraft as existing airlines added to their orders. 61% of our 2Ku backlog is in rest of world and represents great growth potential, 39% is in North America and predominantly represents upgrades from ATG that we believe create an operating growth opportunity. We also had some positive developments in contracts in the quarter.
T-Mobile renewed their contract for a year. American Airlines renewed their 2Ku contract that was expiring in September. Now let me comment on the Boeing 737 MAX situation. We've been able to complete seven installations. We still have 12 in our installation schedule, including one linefit, and we've removed eight from our installation schedule for this year. In total, we have a backlog of 36 MAXs, which includes the seven installed because we do not count an aircraft as online until it is producing revenue. The bigger impact has been the airlines holding back on other aircraft that were in our install schedule, as they need to use those aircraft to fill in for MAXs that they cannot fly.
Our MAX installation and line fit schedule could be at risk depending on decisions by Boeing, the FAA, and the airlines, and those are out of our control. In other OEM developments, service bulletin installations on the 787-9 continue at Boeing, and on the A330, A350, and A380s at Airbus. Line fit on the A320neo family continues on plan for a first line fit installation mid-next year. In the quarter, we had new inductions on the Alaska Airlines 737-800 and the Cathay 777-300. Despite the MAX, we're on track to meet our prior guidance of adding 400 to 475 2Ku aircraft online this year, but this could be at risk per my earlier comment about the impact of the MAX. We're excited about the potential of our CA business for a couple reasons.
First, global wireless usage trends are solid and improving and will drive demand for free Wi-Fi on aircraft. Second, the addressable market is large and relatively untapped, with only about 35% of the aircraft globally installed with a broadband IFC product today. We believe there will be 18,000 new or retrofit aircraft installed with broadband over the next decade. Now let me turn to CA North America installs briefly. We had 92 gross additions, up 50% from 61 in Q1, but down 17% from 111 in Q2 2018, which was an unusually strong gross addition quarter. Net additions, that is net of deinstalls, were up 31 for the quarter versus down 139 for Q1 and down 31 for Q2 2018. I'm going to leave the impact of those installs on revenue for Barry to cover in just a second.
Now let me turn to the CA Rest of World installs, where we again had a strong quarter with 50 gross additions versus 55 for Q1 this year and 47 for Q2 2018. Compared to this quarter, we expect ROW revenue to grow in both the Q3 and Q4 of this year. I'll add that I didn't give a deinstall number for ROW because that's very minimal. We are focused on driving to profitability in our Rest of World segment by installing our backlog, ramping ARPA, reducing program costs as line fit programs are completed, and better utilizing SATCOM capacity over time. Let me turn to our business aviation segment, where results were not as strong as we anticipated due to lower equipment revenue as a result of the impact of FAA ADS-B installation mandates on our dealer channel.
Outside of equipment revenue, we had a very good quarter, experiencing record service revenue, record ATG ARPU, and record ATG aircraft online. Result of the equipment shortfall, we're lowering 2019 revenue guidance for BA to $290 million-$300 million from our beginning of the year guidance of $310 million-$320 million. Despite our equipment sales shortfall, BA ATG plane count grew to 5,462, up 542 aircraft or 11% from Q2 2018, an increase by 114 aircraft or 2% from Q1 this year. Let me dig into the equipment revenue issue in a little more detail. We shipped 186 ATG units in Q2 this year versus 281 in Q2 2018. Some of which can be attributed to a very tough comp last year, shortly after our AVANCE product was launched, and we had a very large backlog to fill due to pent-up demand.
We also misjudged the impact of ADS-B. Let's talk about that for a moment. ADS-B stands for Automatic Dependent Surveillance-Broadcast, and is an initiative the FAA launched a decade ago to improve air traffic safety, and it requires aircraft owners to install ADS-B equipment by the end of this year. Despite the mandate being out for 10 years, many owners procrastinated, and the MROs and dealers are now packed with planes trying to complete the install by year-end. These installs are both crowding budgets for the IFC and also literally crowding out shop floor space as dealers are booked with installations. Though aircraft must be installed by the end of the year, under the mandate, we believe that many will lapse into next year, crowding shop floors through Q2.
Our view of what is happening with ADS-B installs is confirmed by other companies with exposure to the BA aftermarket. Last quarter, we also expressed some concern about the OEM channel. However, we're seeing a recovery there and feel that we should be back on track by year-end. In fact, so far, six OEMs have initiated line fit for our new AVANCE product, and three more are in the process of doing so. On the AVANCE activation front, we're up to 629 L5 customers and 254 L3 customers activated and billing, with 34% of those customers purchasing streaming plans. We remain excited about the opportunity in business aviation. It represents a large, unpenetrated market. We have an exciting new product pipeline, and it provides a resilient recurring service revenue stream with low fixed costs from our proprietary ATG network.
Let me conclude my comments by saying that with the exception of the ADS-B issue, we had a very strong first half. The half was accentuated by a few good guys, but even without those, it was very positive and positioned us well for re-accelerating growth next year. With that, I'll turn it over to Barry to do the numbers.
Thanks, Oak. Before reviewing our detailed operating results, I'd like to highlight our financial accomplishments for the quarter. Adjusted EBITDA of $37.8 million approximately matched last quarter's $38 million, the best in the company's history, and was well ahead of our expectations. Each of our three business segments posted gains in service revenue for the quarter, and this revenue performance was complemented on the cost side of the business through operational execution and Satcom efficiencies. We're again raising adjusted EBITDA guidance on this call. The midpoint of our new guidance implies 55% growth in adjusted EBITDA for the year. Gogo's cash flow, and therefore our cash position, is also running well ahead of plan. For the first six months of 2019, unlevered free cash flow improved by $144 million versus the same period last year and was positive for the third consecutive quarter.
We're projecting unlevered free cash flow to be positive for the full year 2019, and we are on target to improve free cash flow by at least $100 million over 2018. This is a particularly important achievement considering that our net cash interest expense for the year will be $40 million higher in 2019 versus 2018, primarily due to making three interest payments on our senior secured debt during the year due to the timing of our refinancing. This improved cash flow performance is the result of the very strong adjusted EBITDA performance achieved during the first half of the year and improvements in net working capital. While our cash position is ahead of plan, we want to ensure that we have ample buffer capital to support the business.
As we previously disclosed, the indenture for our 2024 senior secured notes gives us the flexibility to enter into a $30 million revolving line of credit with an additional $30 million available based on future performance and leverage covenants. We are in the final stages of negotiating the principal terms of the $30 million asset-based revolver and anticipate completing the transaction within the next few weeks. While we expect to have the facility in place by the end of the third quarter, we do not plan to draw on it at closing. Including this facility and with our cash position well ahead of plan, we are forecasting a minimum total liquidity balance of at least $100 million throughout our planning horizon. We continue to target achieving meaningfully positive annual free cash flow in 2021.
Based on our current plans and projected cash flow trajectory, we do not anticipate requiring additional capital except as needed to refinance our debt maturing in 2022 and 2024. I will now turn to a discussion of our second quarter operating results, beginning at the consolidated level. Total consolidated revenue was $213.7 million for the quarter, down 6% from a year ago due to lower equipment revenue. Total service revenue grew 9% to $173.7 million as we saw growth from all three business segments, despite the impact of the American Airlines deinstallations. Our near-record adjusted EBITDA of $37.8 million was driven by strong service revenue growth, lower operating expenses, and lower than expected Satcom costs. I'd like to provide some context for the way we see adjusted EBITDA playing out for the first and second halves of 2019.
As you know, we achieved exceptionally strong adjusted EBITDA for the first half of this year, totaling $76 million, which substantially exceeded our expectations. These results include the benefit of non-recurring revenue from two different airline partners in both the first and second quarters, with the accompanying adjusted EBITDA benefit Oak described for the second quarter. As we look to the second half of the year, we are projecting increases in Satcom spend to support growing usage, as well as planned expense increases for key programs such as Linefit and 5G. These factors contribute to our expectation of lower adjusted EBITDA for the second half, as implied by our guidance. We continue to expect the momentum we are building this year to carry into strong adjusted EBITDA and cash flow improvements in 2020. Now let's move to a discussion of the business segments, starting with Commercial Aviation.
For CA-NA and CA-ROW combined, aircraft online increased from 3,053 to 3,134 sequentially. As Oak mentioned, this represents the first increase in AOL for the combined segments since the second quarter of 2018, as net aircraft additions in both CA-NA and ROW offset the planned deinstallations, which were completed during this quarter. With the deinstalls now behind us, we expect AOL for CA-NA and CA-ROW combined to grow for the balance of this year and for at least the next couple of years as we install our backlog and achieve new airline wins. Again, on a combined basis, CA-NA and CA-ROW achieved modestly positive segment profit for the second quarter in a row due to both stronger service revenue and lower non-Satcom expenses.
We continue to expect to reduce our functional expenses, which excludes Satcom, for these two segments by approximately $45 million this year. This represents 60% of the $75 million in annual savings we expected by 2020, and we are on track to achieve this targeted cost structure. Importantly, our lower than expected Satcom expense is the result of greater network efficiencies achieved through the deployment of new technologies and our increasingly sophisticated network management capabilities. With the growth in service revenue and the benefits of operating leverage on this reducing cost structure, the combined bottom line performance of CA-NA and CA-ROW is accelerating ahead of our expectations, as Oak described. Turning now to CA-NA. Total revenue for CA-NA in the second quarter was $105.7 million, a decline of 12% from the prior year period, wholly due to lower equipment revenue.
Service revenue increased to $96.4 million, up 1% from the prior year period, despite the American Airlines deinstallations. Because the second quarter results include the benefits we've described, we are projecting CA-NA service revenue to decline and reach a flat bottom in the second half now that the deinstalls are behind us. Equipment revenue of $9.3 million was up from $4 million in the first quarter, but down from $24 million a year ago. This was due to lower total 2Ku installations and a shift in the mix of installations from the airline-directed model to the turnkey model as compared to the second quarter of 2018. Take rates in CA-NA increased to 12.7% versus 11.2% in the prior year, a 14% improvement. Net annualized ARPA increased to $136,000 from $113,000 a year ago, which reflects the second quarter revenue benefits we've described.
Excluding these positive impacts, ARPA would have grown modestly year-over-year. Consistent with our expectations for service revenue, we expect ARPA to decline in the third quarter. CA-NA segment profit increased to $24.2 million, modestly above the $23.5 million previous record achieved in the first quarter of this year and is significantly exceeding our internal expectations. The primary drivers of this outperformance include stronger service revenue, lower than expected Satcom expense, and reduced operations expenses resulting from the IBP programs we initiated last summer. Now let's turn our attention to CA-ROW, which delivered total revenue of $36.7 million, up 9% year-over-year. Service revenue grew 49% to $22.6 million, driven by an increase in aircraft online. We anticipate continuing strong service revenue growth in ROW as we install the roughly 560 planes in backlog.
Equipment revenue was up modestly from the first quarter, but declined 23% to $14.1 million from $18.5 million a year ago. We completed more installations in ROW than a year ago, but the shift in mix from the airline-directed model to the turnkey model resulted in lower equipment revenue being recognized. You'll recall that equipment revenue is based on the number of installations done under the airline-directed model. We were particularly pleased with the increase in take rate ROW delivered for the quarter, as it was despite the higher percentage of aircraft online from new airlines, which more than doubled from 21% in Q2 2018 to 46% this quarter. Take rates for ROW grew from 13.2% to 13.5%, reflecting a 16.1% take rate for seasoned airlines and an 8.6% take rate for new airlines. Importantly, annualized gross ARPA grew year-over-year for both seasoned and new airlines.
Gross ARPA for this quarter was $209,000 for the seasoned airlines and $85,000 for new airlines. In the aggregate, gross ARPA was essentially flat sequentially. Again, an achievement given the growing mix of aircraft from new airlines. Segment loss in CA ROW improved 29%, from negative $24.5 million to negative $17.3 million versus a year ago, as we benefited from continuing improvement in Satcom utilization, lower OpEx from our IBP initiatives, and operating leverage. I'll turn to a discussion of our BA division. Total revenue for BA was down 4% to $71.2 million as a result of lower ATG equipment shipments. Service revenue increased to $54.8 million, up 14% from Q2 2018, while equipment revenue decreased 37% quarter-over-quarter. You'll recall that BA had particularly strong equipment revenue in 2018, up 34% over the prior year on the strength of the AVANCE platform, which was introduced last year.
This year's slowdown comes on the heels of meeting the strong pent-up demand for that product, which is the most successful new product launch in BA's history. The softer performance versus expectations is largely attributable to the timing delays due to ADS-B Oak described. We highlighted this issue last quarter and expect it to continue through the first half of next year. ATG Aircraft online grew 11% over the prior year, and ARPU grew over 2% year-over-year. BA's segment profit decreased to $31.3 million, primarily due to the decline in equipment shipments. Segment margin of 44% remains healthy despite the revenue shortfall and is in line with the expectations we have set due to a higher mix of service revenue and strong cost management by the BA team.
Segment profit of $64.8 million for the first half of the year is down from $69 million for the first half of last year. While we are managing the CA expenses carefully in light of this softer equipment revenue, we are continuing to invest in new products and technology, including the exciting 5G network we announced in June. In spite of these investments, we expect CA segment profit to be higher in the second half of this year than it was during the first half, based on our expectation of increased revenue. I'll now turn to a discussion of our 2019 guidance. Total consolidated revenue in the range of $800 million to $850 million is unchanged. We expect CA-NA revenue to be at the high end of the previously provided range of $355 million-$380 million, with approximately 5% from equipment revenue.
We also expect CA-ROW revenue to be at the high end of the previously provided range of $135 million-$150 million. However, we expect approximately 40% of revenue to be from equipment versus the guidance of 30% we had previously provided. We are decreasing our BA revenue guidance to $290 million-$300 million versus our prior guidance of $310 million-$320 million, based on the reasons we have discussed. We're raising our adjusted EBITDA guidance to a range of $105 million-$115 million. This is an increase from the $90 million-$105 million we provided on our last earnings call and up from the guidance of $75 million-$95 million we provided on our fourth quarter 2018 earnings call in February. We remain on track to achieve at least $100 million improvement in free cash flow versus 2018.
Again, this is despite absorbing additional net cash interest payments of $40 million during 2019. Finally, we maintain our estimate for an increase of 400 to 475 2Ku aircraft online. As Oak mentioned, we continue to monitor the impact the 737 MAX program delays may have on our installations. Before opening the call up for Q&A, I want to take a moment to join Oak in thanking all the team at Gogo for their talent and dedication. It's because of their relentless focus on execution that we are able to continue delivering quality service to our customers and financial results for our shareholders. Operator, we're now ready for our first question.
Thank you. Ladies and gentlemen, if you have a question at this time, please press star then one on your touchtone telephone. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. Again, that is star then one if you would like to ask a question. Our first question comes from the line of Ric Prentiss with Raymond James. Your line is now open.
Thanks. Morning, guys.
Morning. How you doing, Ric?
Good, thanks. Obviously encouraging news on the take rates, both CA-NA and CA-ROW. Where do you think that heads? Obviously previously you had what we thought was pretty conservative expectations on take rates, but, and Oak, you also mentioned you have two airlines on free. Delta did their trial, but what are your thoughts on how we should view the demand side of this equation, on the take rate side?
Obviously the move to free will drive take rates up dramatically. Our take rate today has a blend of both paid browsing sessions, paid streaming sessions, and free messaging in it. If things go free messaging kind of goes away and we expect a very large jump in the browse sessions, and potentially a jump in streaming sessions. It's an order of magnitude larger, Ric. We see 30% take rates on free fleets in Asia right now, and they could be higher in North America where there are longer routes.
Yeah. Yeah, go ahead.
I was just going to kind of do a double click on that, Ric. If you look at take rates under the steady state model with 2Ku being installed, it's a little bit different in North America versus ROW, as we described a bit. In North America, as 2Ku gets more fully installed, we do expect to see those take rates improving over time. As that capacity grows and it's not constrained. In ROW, there is this combination effect of seasoned airlines and new airlines. As I mentioned, there's quite a disparity, almost two x between the current take rates for the new airlines versus the seasoned airlines. We do expect those take rates for the new airlines to grow pretty significantly up, as they get seasoned to approximate what we're seeing for the take rates overall in that part of the world.
That makes sense. On the cost side, we were obviously pleasantly surprised by the SATCOM cost. You've called it out a couple of times on the call, but how should we think about modeling that? It's probably one of the hardest ones I've found to model for you guys. As we think about the take rates going up, how you manage that cost side. How should we think about the SATCOM cost going forward, besides just what you mentioned for the second half of 2019?
First, it's important to underscore what's going on structurally here. We are seeing the benefits of the technologies and the network management that we're putting in place that does drive less SATCOM requirement than we had previously expected. That's a good thing. We see that continuing. In terms of how to model that going forward, clearly SATCOM expense will grow as usage grows. There are a couple of drivers of that. One is that we do see absolute levels of pricing per megahertz coming down around the world. That has been true historically, and we expect that to be the case. We also see ourselves getting better utilization in SATCOM, particularly in Rest of World. As you know, there's both a coverage requirement and then a capacity requirement for SATCOM.
We do have worldwide coverage, and as we put more planes flying in that network, we will see, as we have already, a significant benefit in improved SATCOM utilization. As those planes get added, then you'll add capacity as it comes along. I think the way to think about that is that we see data margins generally north of 50% in North America. The costs in Rest of World continue to be higher. We'll see efficiencies over time from that. Those data margins will only grow as we see that better efficiency and the reducing costs, but not to that level in the short term that North America is seeing.
Okay. Glad to hear you're making progress on the 5G next-gen ATG by picking a U.S. provider. You mentioned a couple of times that there'd be some higher costs. How should we think about the cost of that 5G impacting 2019 and 2020, both on OpEx and CapEx?
It started now a bit. It's in the order of a few million dollars for the back end of this year. The spend ramps during the course of next year. It is comprehended in the guidance that we have given. In terms of the CapEx side, it's comparable on an aggregate basis to what we had laid out before for the LTE-based network.
Okay. Thanks, guys.
Thank you. Our next question comes from the line of Phil Cusick with J.P. Morgan. Your line is now open.
Yeah. A couple of follow-ups there from Ric's questions. First, you said Gogo 5G, you have a solutions provider. Is that the vendor that you'll use? How many sites do you envision at kickoff?
Yeah. In our business, it's a fairly specialized business, so when we develop a new network, we can't buy things off the shelf. We have to actually design the equipment with our vendors. We are partnered with an American U.S.-based 5G, I would call them an infrastructure company, in designing and developing the radios and antennas, et cetera, for the new system. Like we've been partnered with ZTE in the past, we now have an American vendor in that role. In terms of towers, I'm trying to remember if we thought 200, Barry?
Yeah. 150 to 200.
Yeah, 150 to 200, yeah, when we roll out.
Grow as required from there.
A lot of those towers will probably be the same as we use today for our ATG network. We'll get some synergies there because we've already got sheds and generators and things like that. Obviously, we'll be paying more rent at those towers because we'll be taking more space. There are some cost synergies in the network rollout.
Good. Can you give us any sort of results from the Delta unlimited trials? How did speeds hold up? What did you see in terms of capacity issues?
Well, the trial was really aimed at market acceptance of free. It wasn't meant to test our system. Going to full free, we obviously have things we need to grow, like our satellite capacity and other things to be able to really provision that. In terms of how the test went from a market perspective, I think that's a question Delta should answer. It's their role to announce what they're doing with the free Wi-Fi, and it's our job to support them operationally. I'll leave it at that.
Okay. Lastly, any update, Oak, on how you think about the strategic and competitive ecosystem for in-flight broadband globally? What are you seeing in terms of both demand and then what are you seeing from your competitors? Thanks.
In terms of demand, we see the trend of free Wi-Fi will drive a tremendous amount of demand.
Will actually be good for our industry. Our industry's been kind of the dog everybody likes to kick for a couple of years. With demand growing, I think we'll have a lot of growth industry-wide. There's been always a lot of speculation about some consolidation in our industry. I think that that speculation continues. Most recently, you have Inmarsat going private, which I think in partnering with Panasonic, I think that's an indicator of a type of consolidation. I think we expect that there will be more consolidation or there may be. We want to be in a very strong position to play a role in that if it happens.
Just to clarify, anything happening on the RFP side? We've seen a kind of a freeze on new airlines for quite a while.
Yeah, there are lots of RFPs. There aren't many awards. We are seeing a lot of RFPs. We're deep into some processes that we've literally been working on since I got here, from before I got here. We hope to be able to conclude some of those, but nothing to report at this time.
Okay. Thanks very much.
Thank you. Our next question comes from the line of Simon Flannery with Morgan Stanley. Your line is now open.
Thank you. This is Landon Park on for Simon. Just wanted you to touch back on the Delta free trials. Are you able to comment on any sense of a timeline there for an ultimate decision or any future trials that have already been set? Secondarily, can you comment on any ongoing strategic discussions around potential investments or transactions on that front?
I'll start with the second question. Right now, we're really focused on operationally supporting airlines that want to go free, and we're going to be focused on that, I think, for a while. Not very focused on strategic transactions at this particular moment. In terms of timeline for Delta, you really have to talk to Delta about that. I think you got to understand what your role is in the world, and our role is not to be commenting on what Delta's doing with free IFC.
All right. Thank you very much.
Thank you. Our next question comes from the line of Louie DiPalma with William Blair. Your line is now open.
Good morning, Oak, Barry, and Will.
How are you, Louie?
Great.
Not bad. What are the engineering challenges for Gogo's 2Ku system to offer free Wi-Fi at such a large scale that has never been done before with 98% availability? Are there necessary hardware modifications, or is the formula as simple as increasing capacity via purchasing more bandwidth from your partners, SES and Intelsat?
We've actually got 12 partners, yeah, that's it, Louie. I mean, the good news about 2Ku is that we sold it as future proof, it really is future proof. The ramp up on the aircraft for our 2Ku aircraft is very minimal. On the satellite capacity, in satellite capacity, yes, we have to ramp up satellite capacity. That's for sure. The other good thing about the 2Ku antenna, we talked about it being future proof. We think that it is very conducive to future technologies like LEOs, et cetera. As those kinds of technologies come online, we'll be well-positioned to serve our airlines with lower latency, lower cost, et cetera.
You don't see any technology impediments? It's just a question of you spending more money. It's just a money question, and if the money's there, then you can offer the free Wi-Fi at scale?
Yeah, that's basically right. We are making some software improvements, in terms of portals and things like that. That all has to change when you go to free. You don't need to get a person's credit card number anymore and things like that. There's some software development involved, but that's relatively minimal.
Okay. On the business jet side, you've talked a lot about the ADS-B creating a constraint. Would you expect ATG antenna shipments to be up next year relative to this depressed 2019 number, or is it too early to say right now?
Well, I think we expect there to be some crowding of the aftermarket channel through second quarter.
We've looked at the numbers of jets that are still not installed with ADS-B. We expect some of those will never be installed. They'll just be retired. We do expect a set of those to still come in next year to get installed, and that we think in talking to the dealers and MROs, that they expect to still be installing a fair amount of ADS-B equipment through Q1 and Q2. We think that'll still impact our equipment sales going into the year, but we expect it'll pick up again in the second half.
Yeah, in total for the year, Louie, we do expect the equipment revenue to kind of be in a trough this year relative to 2018 and then pick back up again in 2020.
Gotcha. A more broader question. Gogo shares traded down sharply by over 30% from like $6 over the past two months going into this earnings report. You guys probably noticed. We think there was concern that your record first quarter EBITDA performance was a fluke, and with your leverage, it would magnify weak second quarter results. Instead, you doubled the EBITDA consensus expectations. Can you talk about, after the strong first half of the year, how your improved visibility may or may not allow you to invest into what I would refer to as discretionary projects such as the 5G and flat panel phased arrays with Phasor, and it's been speculated you're working on a tail-mounted business jet antenna with Gilat. Do you feel comfortable in your liquidity to do these discretionary projects now, despite the fact that the stock market seems to be doubting you?
Yeah. Well, the stock market hates us. We know that. Our projections include our strategic investments like 5G. Phased array antennas, yes, we made an investment in Phasor, which we believe in ESAs at some point. We're not talking about the timing of those. We have not officially announced a tail mount antenna for the BA market, but we have a vendor who seems to have announced that prematurely. We hope that means that they will deliver the product prematurely. Yes, those are all contemplated in our current cash flow guidance, and I'll turn it over to Barry to get in a little more detail on the numbers.
Yeah. We're keeping the pedal to the metal on 5G and those other initiatives. Louie, we, as I had said, are continuing to make those investments. We think it's important to maintaining BA's market position and ultimate financial performance. We are still pressing forward on those. We're very pleased to see the first half obviously come in with the level of EBITDA that it is, but we're also just being cautious as we provide guidance for next year. There are some known things we talked about. These good guys in the first half of the year on the revenue side that translate into the EBITDA benefit, those will not persist. We are planning on ramping some of these expenses, like we talked about, in Satcom and in OpEx to cover these strategic investments.
Having said that, we think the guidance that we have of $105 million-$115 million is solid EBITDA performance, 55% growth again at the midpoint. We see that momentum continuing into 2020. I think that's the important point here, and we realize that we've had to dig ourselves out of a hole in the last 15 months or so on the heels of the icing crisis. As we talked about on the last call, we've knocked those off one by one, and we're in a much different place now. We think have the foundation laid, and it's showing up in the financial performance that gives us confidence about really being able to deliver on the kinds of increases in EBITDA that we've talked about.
Sounds good. Thanks, guys.
Thank you, Louie.
Thank you. Our next question comes from the line of Greg Burns with Northland Securities. Your line is now open.
Good morning, guys. Thanks for taking my questions and congratulations on the quarter. First, in the past, you've talked about using beamforming technologies to more efficiently utilize capacity, really allowing you to reduce satellite leasing costs by quite a bit. Could you provide some additional color on the timing of when we might see those occur, given expenses or Satcom expenses are expected to grow in the near term?
Well, obviously, there's a couple types of beamforming technologies. We're more and more moving to high throughput satellites, which are Ku satellites with spot beams as opposed to wide beams. That is helping our efficiency now. We've also talked about software-defined beams and more dynamically programmable satellites. Those are out there, I would say, in the 2023 and beyond. That's technology that is pretty well designed, but it needs to obviously be manufactured and launched. There are some technological hurdles that need to be overcome in terms of modem technology and other things for that to become meaningful. The third type of beamforming technology we talk about is in our 5G product, in the ATG world, where that will use beamforming technology. The antennas will point more directly at an aircraft as opposed to broadcasting a large funnel like our current ATG product does.
That will roll out in 2021.
I think, Greg, to your question on the increase in Satcom expense, yes, we of course, are projecting increases in Satcom expense, but that's driven by significant increases in usage as we add more planes and take rates increase that we've described. The level of growth in that expense is muted by the kinds of things we're talking about. High throughput satellites, raw bandwidth costs coming down, better utilization through the kinds of network management capabilities we have.
Right. Okay, that's helpful. Second, as we think about ARPA in the Rest of World segment, you've talked about how the percentage of aircraft online that are in new fleets is expected to grow from maybe now till year-end. You've talked about new fleets being a little bit more reluctant to market their IFC until the entire fleet is installed. Does that mean we should expect ARPA and the Rest of World to slide lower through year-end?
Well, yes, you have certainly the concept right, which is that you have this kind of blended ARPA that is a result of the difference that's more than 2x between the seasoned aircraft and the new aircraft. If you look at that over time, over the last couple of years, that gross ARPA has seen that impact. What's happening more recently is that, if you look at this year versus last year, for example, we're seeing improvement in blended ARPA overall, but we see that kind of staying relatively flat-ish as you go forward. We see it coming up in the back half of this year, but I wouldn't count on big improvements in that blended rate of ARPA until we get these newer planes installed.
Fleets.
Fleets, excuse me. Newer fleets installed and seasoned for the new airlines. You really see the benefit of that.
Got it. That makes sense. Last quick one from me. Why do we see take rates decline sequentially in CA-NA, and on the Rest of World side?
That's seasonal, and relatively typical.
Okay, fair enough. Thank you.
Thank you. Our next question comes from the line of Lance Vitanza with Cowen. Your line is now open.
Hi. Thanks, guys. A nice quarter. Just to pick up where we left off with the last question on the ARPA being flat sequentially is, as the new planes are coming online, and it's offsetting the growth in the planes that are already seasoned in, what would you say, though, is the underlying ARPA trend here, if we think about on a like-for-like basis for those seasoned planes? Is it up low single digits, mid-single digits, high single digits?
For the seasoned planes?
Yeah.
It's up in the second quarter. As I mentioned, it was 16.1%. In the year-ago quarter, it was 14.5%, so it was up since then. We've seen that trend grow.
I'm sorry, in the Rest of World Segment? Those were the numbers that
Correct. Wasn't that what you were asking about, Lance?
Yeah. No, that's great.
Were you asking about take rates?
No. That's right. Thank you. Speaking of take rates, so given the trend toward airline-directed and free, I'm not sure take rates are enough to really measure the progress, and I'm wondering how you think about that. Seems like we should be thinking also about, I don't know, duration of usage or megs per flight or something like that. Do you have any thoughts there? What other metrics are you looking at to gauge usage, and what can you share with us?
Yeah, I agree with you that take rates compared to today are going to be kind of meaningless. They're going to be a lot higher. The way we look at this in very simple terms is volume is going to, we think, explode with free. Unit costs will come down. Pricing rather will come down, obviously. They'll be more than offset by the growth in volume and produce solid revenue for growth for us and value for us.
Okay. On the Satcom cost side, and I apologize if I missed this too, but I heard, very obviously, that unit costs are falling. What would you say is going on with the rate of that decline? Are cost declines accelerating or decelerating today? Given your outlook for increased global capacity, what do you expect over the next few years?
Yeah. We're doing deals right now at continually declining costs on the Satcom side. We see them in the forward, as we look sort of at the forward market, if you will, falling pretty dramatically over the next couple of years. There's a lot of supply coming on in the 2023, 2024 timeframe, which we think will further depress pricing. I think that the long-term trend in Satcom is going to be continually declining in pricing. At our side, we'll be focused much more on continuing to make efficient use of megahertz appropriately and bringing our capacity utilization up. All those trends should be, on a unit basis, driving Satcom costs down.
Okay. Thanks very much, guys.
Thanks, Lance. Okay, that's.
All right. Well, thank you very much for attending our Q2 2019 earnings call, and I'd like to leave you with a couple thoughts before we go. First, we have a very strong cash flow generating business in BA. Not only does it have a unique competitive advantage by virtue of our spectrum ownership, but it also has ample runway for growth because the BA market is relatively unpenetrated. Second, the rest of world is growing. It's also an extremely large and unpenetrated market. With our global 2Ku platform, our progress on Linefit, and our strong backlog, we're well-positioned to win our share of that attractive market. Third, CA North America will bottom out in the second half of this year as the impact of American Airlines deinstalls and their conversion to the airline-directed model will finally be behind us.
Fourth, we've strengthened our balance sheet and given ourselves strategic flexibility by refinancing our $162 million convertible stub and our $690 million senior notes and pushing those maturities out until 2024. Finally, by virtue of our industry-leading market share and our asset-light operating model, we're well-positioned to take advantage of the opportunities that I just described. We look forward to demonstrating this to you in future quarters, and thank you for your time this morning for joining us for our quarterly phone call. Thanks again.
Thanks very much.
Ladies and gentlemen, thank you for participating in today's conference. This does conclude today's program. You may all disconnect. Everyone, have a great day.