Gogo Inc (GOGO) 2018 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q3 2018 Gogo Inc. Earnings Conference Call. (Operator Instructions) As a reminder, this conference call is being recorded. I would now like to introduce your host for today's conference, Mr. Will Davis, Vice President of Investor Relations. Sir, you may begin.

Thank you, and good morning, everyone. Welcome to Gogo's Third Quarter 2018 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 caption Risk Factors in our annual report on Form 10-K and 10-Q and other documents we filed with the SEC. In addition, please note that the date of this conference call is November 6, 2018. 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 this call, we'll present both GAAP and non-GAAP financial measures. We included a reconciliation and explanation of adjustments and other considerations of our non-GAAP measures to the most comparable GAAP measures in our third 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's remarks, we'll host a Q&A section -- session with the financial community only. It is now my great pleasure to turn the call over to Oakleigh.

Thanks, Will. Good morning, everyone. We had another strong quarter with all 3 of our business segments coming in ahead of our expectations on every financial metric. And we made good progress on our Gogo 2020 plan as evidenced by our strong cost control in the quarter. Our consolidated EBITDA performance was especially strong at $21 million versus our expectations of around $6 million, with roughly $11 million of the outperformance coming from our CA-NA business and the balance almost evenly split between CA Rest of World and Business Aviation.

On the revenue side, the biggest drivers of outperformance were third-party payer programs. And on the cost side, we had lower cost of service driven by Satcom savings and lower OpEx driven by personnel costs, lower performance related penalties, and $3 million of timing and ED&D spent. Maybe most importantly, our cash and cash equivalents ended the quarter $8 million ahead of plan.

I want to thank the Gogo team for working hard to make these results happen and for working even harder to make sure we meet or exceed our customers' expectations, and build Gogo into a valuable partner and profitable company in the future. We've been hard at work improving our Gogo 2020 plan and that's the long-term business plan we discussed at the conclusion of our integrated business planning process in July. As we stated at that time, we were very conservative in revenue projections and Satcom spent projections in that plan. Since our second quarter call, we've adjusted the plan to: a, reflect more realistic revenue projections based on our current revenue trends; b, reflect more realistic Satcom spending based on current offers we're receiving from our Satcom vendors; and c, reflect the current state of our airline subsidy renegotiations. The enhanced plan called Gogo 2020 plus significantly improves the performance of our CA business and maintains a positive cash balance throughout the planning period, thereby significantly improving our ability to fund the business going forward.

I'll start my comments today by diving into the quarterly highlights by segment. Then, then I'll discuss: one, our progress dealing with the deicing issues we faced last winter; two, some industry and business model issues including a discussion of Ka versus Ku-band satellite capacity, the airline directed business model versus turnkey and the impact of Delta wanting to offer free WiFi on their aircraft, which we think is an exciting development; And third and last, I'll close with a few Gogo planning topics including recent developments in our next-gen ATG development program and more detail on the 2020 plus enhancements to our business plan I just talked about a moment ago. After that, I'll turn it over to Barry to discuss the numbers.

Two things we are not going to spend a lot of time talking about today are strategic options or refinancing of our convertible notes due 2020. As I said on the Q2 call, we continue working on those matters with great urgency and will provide more details once decisions are made.

So let me start with the BA division, which continues to outperform with 20% of revenue growth year-over-year, driven largely by equipment revenue, which was up 40% year-over-year due to the strong growth of our AVANCE platform. We shipped a record 296 ATG units in the third quarter, though I caution we expect shipments to decline a bit in Q4 as the OEM and dealer channels have largely ordered what they need through year-end. These equipment sales portend good things for service revenue next year, as these products get installed and activated in the field. We also introduced a new Gogo DASH product in the quarter. This product is designed to allow fleet operators and flight departments to use a mobile app or a web-based portal to monitor the in-cabin health of our AVANCE system in flight, to make sure that passengers are having a positive experience or to take remedial action if they're not. DASH is an example of the types of product extensions and third-party applications our AVANCE platform will enable us to launch and host that I discussed in our last quarterly call and that should allow us to accelerate ARPU growth over time. AVANCE L5 is delivering 3x the bandwidth of our traditional ATG product and is available with a streaming data plan or a browsing data plan with about 35% of customers choosing the streaming plan. Finally, we have more than 50 aircraft models now with full equipment STCs for exhaust and another 198 aircraft models covered with WiFi only STCs. There are 2 factors I think investors should consider when thinking about BA next year. First, our strong equipment sales this year portend good things for service revenue next year. Second, we also plan to spend more money on network upgrades, which will hit OpEx and CapEx, and which I'll discuss in more detail when I get to my next-gen ATG comments in a few minutes.

Now let me turn to CA North America. We're really pleased with the resiliency of our CA-NA service revenue. If you back out American Airlines, service revenue for all other airlines was up 18% for the quarter over prior year. We expect that to slow next quarter because of an unusually strong comp in Q4 2017 but to pick up again Q1 next year. That 18% Ex-American service revenue increase was driven by a 5% increase in average aircraft online and by a big increase in take rates. When you exclude American, take rates hit 14%, up 87% from 7.5% in the prior year, driven by our multi-tier pricing model, including new lower-cost offerings such as airline sponsored free messaging and our T-Mobile partnership. Though multi-tier brings average revenue per session down, it drives overall revenue up by reaching more segments of the cabin with pricing and service levels appropriate for each segment. And we expect this to start driving a gradual return to ARPU growth over the next couple of quarters.

When you include American, our service revenue was down about 1% from prior year from the combined effect of de-instillations and lower session pricing under the American Airlines AD arrangement. It's important to note why we exclude American when analyzing our revenue growth. And as we've discussed previously, our ATG product was installed in most of the American mainline fleets. And in 2016, we lost the bidding process for upgrading 550 of those aircraft to a satellite product. We view this as a onetime anomaly because all of our other ATG customers have elected to upgrade their mainline fleets from ATG to our 2Ku satellite product instead of competitors' products.

Operationally, it was a strong quarter. We had our first Gogo Commercial Aviation OEM line-fit delivery, a beautiful A220 for Delta Airlines, which had both our 2Ku IFC and our new wireless seatback entertainment system installed on the assembly line. The entertainment system, Gogo Vision Touch, is a really cool product because it brings seatback entertainment to the narrow body in regional jet markets at a much more affordable price and with much less weight than traditional seatback IFE products.

On the 2Ku front, our system availability was a solid 97% for every month in the quarter, which drove significant increases in customer satisfaction and net promoter scores as measured by Gogo and by our Airline Partners. I should note that Ed Bastian, the CEO of Delta Air Lines, even mentioned at the Skift conference that Gogo had become a contributor to Delta's high NPS.

On the product front. In Q3, our new agile development teams started delivering the user experience improvements I discussed in our last call, including projects that remove friction in the sign-in process. One such product, which involved implementing captive portal for Apple devices, has been demonstrated to increase take rates significantly. And based on testing that we did post launch, we estimate will grow revenue by more than $7 million a year.

As we look forward to next year in CA-NA, we expect pressure on revenue from the American Airline de-installs to continue, but we also expect significant segment profit improvement, as many 2Ku STC costs and deicing costs will be behind us and as our Gogo 2020 costs and quality improvement initiatives reduce our cost structure further.

Now let me turn to CA Rest of World or ROW, our growth engine. We're in the early days of building our ROW business and very proud of our results to date. In Q3 2013, we started with 0% market share of installed aircraft, and at the end of Q3 this year we've grown that to 17.1%. For the quarter, service revenue was up nicely and we saw operating leverage on our satellite and other operating costs. We're very excited about some of the new world class customers we're installing in ROW this year, including British Airways, Air France-KLM, Air Canada International, Cathay Pacific, Virgin Australia and Iberia, which together represent more than 500 future revenue-generating aircraft. The key to our success in ROW will be to get these great new world-class airlines to perform at the same levels as our existing ROW airlines. In Q3, airlines that were installed prior to 2017 had a roughly $200,000 ARPA versus airlines which started installing in 2017 or later, which had an $82,000 ARPA. In the quarter, we activated 54 ROW aircraft 2Ku and we expect to install more than 200 for the year leaving us with more than 600 satellite aircraft online in ROW going into next year and with another roughly 500 in backlog.

Looking ahead to next year, for ROW, we expect strong revenue growth and we expect that growth to outpace growth in service costs as we scale to meet the demand of new airlines. And we expect operating expenses, excluding service costs, to be roughly flat. Total CA installs. On the downside, across all of CA, we are behind on this year's installation plan, which has a timing impact on service revenue. We're lowering our install guidance to 450 to 500 aircraft for the year. Install delays have occurred for a variety of reasons, including airline scheduling issues, teething pains in our international kitting and shipping, errors by airline engineering vendors, OEM program delays and an Airbus ACE issue concerning placement of an emergency location transponder behind our radome, which we expect Airbus will resolve by early next year.

On the positive side, we had 3 new inductions in Q3, including the Go 737-700, the Air France Boeing 777-200 and the American A319-100. We're very excited about our first induction of Q4 because it's our first service Boeing 787 and Air Canada B787-900 being installed by Boeing global services in Victorville, California. This is very important to us as we have roughly 60 more ROW 787 installs scheduled for the next 2 years. Despite our delays this year, we'll have achieved close to 500 installations a year for the third year in a row, which is a pace that no other player in our industry has matched.

And now I'd like to update you on our deicing progress. As we mentioned in our last call, we have 3 levels of modifications to fix deicing. The first has 2 components, a modification of the radome sealed with fuselage and the addition of deflector plates inside the decompression and drainage holes of the radome. We've tested this solution extensively and feel that if installed properly, it will eliminate in the high 90% range of all deicing events. The second modification is a bottom cover on the antenna itself. Importantly, we'll only install these if either the first approach does not work or at the customer's request. And finally, we may roll out a top cover as a third solution should we need more modifications or a further testing of the top covers proves them to be more effective than other solutions. As of yesterday, we'd installed 72% of our North American aircraft with the first modification and anticipate having 95% of the North American fleet installed by Thanksgiving. We'd like to thank our Airline Partners, especially Delta Air Lines, for their collaboration on this important project.

Now I'm going address some industry and business model topics of special interest to investors, starting with the old debate about Ku versus Ka radio frequencies. To me, the proof is in the doing. Today, we serve the busiest airline at the busiest airport in the world. Delta Air Lines at the Hartsfield-Jackson International Airport in Atlanta, Georgia. We serve 700 Ku satellite flights a day in and out of Atlanta, with 97% system availability of user speeds above 15 megabits per second on more than 80% of those flights. In total, we have 3 spot beams and 4 wide beams that cover the Atlanta region. And we have ample opportunity to add more if we want to. It would be difficult to do that with Ka today because of the 10-Ka satellite beams that hit the Atlanta region, 3 are part of closed systems, so you'd have no way to add capacity if you needed it, 2 are primarily serving a home Internet market and 6 are quite old and had very limited capacity. I also want to be clear, we're not opposed to Ka. To us, it's just another radio-frequency. And if someday that band is significantly cheaper and more abundant than Ku, we'll move there. Though we don't think that's going to happen soon. Our engineers have frequent conversations with our vendors and potential vendors about affordable path to 2Ka if we want to go there and we have multiple options if we want to pursue modifying our 2Ku installations to work with Ka.

We think it's important to be agile in a world where technologies change rapidly, and that's why we've adopted an open architecture approach to our product roadmap. Central to that is our modular design so that we can upgrade different components of our system at relatively low cost as technologies evolve. And we maintain constant contact with all technology providers in our ecosystem so that we are on top of technological change, not a victim of it. We've been very successful with this strategy. No company in the IFC industry has migrated to more different technologies than we have. We started with analog ATG cell service in the '90s and then migrated to a Iridium Satellite Service in 2000 and then back to 2 different flavors of digital ATG in 2008 and 2010, and now the Ku. So we've proven our agility. Another question that comes up a lot is about the airline directed model and whether that is a threat to Gogo. It's true that our revenue has been hurt by American's migration to AD, but it's important to understand that all airlines that had a right to move to AD at a predetermined price have done so. I'd also point out that airline directed is not necessarily the best thing for an airline. When they make the switch, the airline goes from getting a check to writing a check, with the added problem now having a new charge to cost per available seat mile or CASM, which is a key metric in the airline industry. In fact, some airlines have started to realize the CASM issue and are talking to us about moving back to turnkey. Another topic that has come up in recent weeks is Delta's desire to provide free service in their fleet. We have other airlines that offer free service and those partnerships have been very successful at growing revenue for both parties. We are fully supportive of Delta's move to free and we're already working with them to make that happen. We think that free will drive take rates and customer satisfaction, which will ultimately be good for both Delta and for Gogo.

Finally, I'd like to address a couple of Gogo planning topics, starting with NextGen ATG. I'm going to spend some time on this because it's become a major concern for some investors who fear that our very cash-generative BA business is vulnerable to attack from a potential competitor who has announced that they are rolling out a similar product.

Before I go any further, I want to make clear that we have not made a formal decision to proceed with next-gen ATG. The point of this discussion is to make it clear that we can deploy very quickly and effectively if we need to. As a result of the Department of Commerce lifting its denial order against ZTE, a major vendor on the project, we're able to start test flights again in September. We have a testbed of 10 production towers between Michigan and Colorado. In a test flight, the system is now performing to the levels predicted by our computer models more than a year ago. Some of the key metrics include peak throughput of more than 100 megabits per second. Average and mean throughput performance of approximately 10x the performance of our current ATG system. And perhaps, most importantly, a 300-kilometer range from the tower to the aircraft. The last point is really critical because it has a very direct impact on the CapEx one needs to build out the network and because it's not easy to do. With our years of experience building ATG networks, we have a proprietary modified air interface protocol that enables us to extend the range of our coverage to about 3x the range of more traditional LTE approaches. We have further advantages, which begin with the fact that we don't need to build out greenfield towers. We will be able to co-locate about 2/3 of our next-gen antennas at existing ATG towers where we already have backhaul and shelters for our base transceiver stations, power generators and the like. We'll also be able to leverage our existing data centers, core system software, network operating center and network staff, and perhaps most importantly, leverage our existing ATG network. This last point is important because 2.4 is unlicensed spectrum. There will be noise interference from other users in densely populated areas. If one does not have a backup system, users will lose their sessions when they encounter heavy interference. In our case, this system will simply flip back to our 850-megahertz license spectrum and maintain the user's session, much as a 4G phone flips back to 3G when no 4G is available. We've been working on this project for several years and most of the startup R&D is behind us. However, we will not commit to roll out until we complete the commercial feasibility phase of our product approval process, which we should complete by the end of the first quarter next year.

Preliminarily, it looks like this should be a nice step-up product for our regional jet customers and for midsize and larger business aircraft that fly over the United States.

Now let me finish with an update on the progress we've made on our Gogo 2020 plan. In our IBP and Q2 calls, we noted that our original 5-year Gogo 2020 plan was primarily an operationally focused plan, and that we had made very conservative projections for revenue growth for Satcom expense and on the level of subsidies we provided to certain airlines. At that time, we also said that we were going to focus on improving the 2020 plan on all 3 of those fronts in order to get -- in order to drive Gogo to positive cash flow sooner than anticipated in the original plan.

On the revenue front, as 2Ku performance has improved, we've seen strong growth in take rates and service revenue, ex America. And in fact, we hit a 12.4% take rate in Q3, which is almost at the 12.6% take rate we projected for 2022 in the original Gogo 2020 plan. We've also made significant progress in driving third-party revenue deals, which further enhances our revenue prospects. So as a result of looking at current run rates for airline, passenger and third-party paid revenue streams, we've raised our internal projections on a per airline basis for future revenues in the 5-year plan.

On the Satcom front, in the original 2020 plan, we used current contracted pricing to project costs for our future megahertz needs. Given the current state of play between us and our Satcom providers and our new Satcom capacity planning process, which was part of our IBP plan, we expect much lower megahertz cost coming online over the next few years and have reduced our projected Satcom spend accordingly.

I want to digress for one moment and address an issue that some analysts have raised about whether migrating planes from AGT -- ATG to Satcom will dramatically reduce our operating margins, and the answer is no. Our Satcom data margins are positive today and will get more positive as utilization grows and a lot of network and opening costs below the data margin line are quite fixed in nature, and therefore, give us the ability to get operating leverage on our Satcom spend. So as we shift to more Satcom, data margin percentages will come down slightly, but not dramatically.

Finally, on the equipment subsidy front, several airlines have entered into negotiations with us about reducing subsidies today in return for lower service fees or other changes in future contract terms. We will update you when and if appropriate on those negotiations. I should note that even before renegotiating any deals, Gogo has made substantial progress in reducing subsidies, as evidenced by our net investment per aircraft declining 25% from Q1 of this year to Q3 of this year. As a result of these changes, we've updated our Gogo 2020 plan to reflect the positive trends I just discussed. We're not going to provide specific long-term guidance today, but will say that the revised plan includes substantial increases in EBITDA in all 5 years and leaves us with a positive cash balance throughout the 5-year period.

Thank you very much, and now I'll turn it over to Barry to cover the numbers.

Thanks, Oak, and good morning, everyone. Let's jump right into the numbers.

Gogo delivered third quarter adjusted EBITDA of over $21 million well ahead of expectations. Our Business Aviation division continued its strong momentum, with revenue of $73.6 million, up 22% year-over-year. And our CA-NA segment delivered 18% service revenue growth, excluding American Airlines.

I'll highlight a few metrics for the quarter before reviewing the performance of our 3 business segments. Adjusted EBITDA of $21 million was up 63% from a year ago, and meaningfully exceeded expectations. This strong performance was primarily due to higher CA service revenue with the recovery of 2Ku performance, lower-than-expected Satcom costs and continued strong momentum in BA. Our costs to remediate deicing were also lower than anticipated and there was a benefit of approximately $3 million in timing of expenses, which we now expect to occur in the fourth quarter. It's also worth noting that our quarterly results include absorbing a $1.5-million severance charge related to headcount reductions in July. Gogo's consolidated revenue reached $217 million, up 26% year-over-year with higher equipment revenue generating nearly 85% of our overall consolidated revenue growth in the third quarter. The adoption of the new revenue recognition standard, ASC 606, was a primary contributor to this increased equipment revenue. Strong BA equipment sales were also a smaller factor on a consolidated basis. The adoption of ASC 606 increased equipment revenue by $31.4 million and reduced service revenue by $4.5 million in the third quarter. Consolidated service revenue grew 5% from a year ago, driven primarily by strength in BA and CA-ROW. Net ARPA in CA-NA increased nearly 7% in the third quarter from the prior year period, reversing the 1% year-over-year decline we experienced last quarter. Excluding American Airlines, CA-NA net ARPA increased 16% year-over-year in the third quarter of 2018.

I'll now move to a discussion of our business segments, starting with Business Aviation. Service revenue grew 14% versus the prior year period, driven by a 10% increase in ATG units online and a nearly 5% increase in average monthly service revenue per ATG unit online.

We achieved a major milestone last month as we surpassed 5,000 total BA aircraft on our AVANCE platform and other ATG systems. BA equipment revenue grew 41% from the year-ago period to over $24 million in the third quarter as demand for our AVANCE L5 and L3 systems remained robust. Importantly, BA's recurring revenue model continues to benefit from both unit and ARPU growth. Adjusted EBITDA for BA in the third quarter reached $35.2 million, up 65% versus the third quarter of 2017, with record segment margin of 48%. We believe that we're well positioned to achieve our long-term growth and profitability objectives based on the strength of our product offerings, continued execution and the headroom that remains in the market. Today, less than 25% of all Business Aviation aircraft are equipped with broadband connections. We would expect the segment margin percentages to be muted somewhat from today's levels if we decide to invest more heavily in the ATG network upgrades as Oak outlined.

Now let's turn to Commercial Aviation, beginning with North America. Total CA-NA revenue grew over 13% from the third quarter of 2017 to $108.5 million with equipment revenue driving the increase. Equipment revenue was $15 million, up from $1.3 million in Q3 '17, which reflects the adoption of ASC 606. Service revenue in CA-NA declined 1% to $93.4 million from the prior year period and an improvement from the 3% year-over-year decline in the second quarter of 2018. Excluding American Airlines, our service revenue grew by 18% from the prior year after growing 14% year-over-year in Q2 of 2018, reflecting some nonrecurring benefits this quarter from third-party payer programs. We're pleased to see the strong underlying demand in CA-NA as evidenced by growing take rates. This reinforces our expectation for CA-NA service revenue to resume growth in 2020 after absorbing the impact of the 550 American Airlines de-installations.

CA-NA segment profit was $8.7 million for the quarter, down from $16 million a year ago, driven by increased satellite cost to support capacity requirements for our expanding 2Ku fleet, the impact of American Airlines and higher operating expenses. CA-NA segment profit was up from $6.8 million sequentially based on an improvement in service gross margin, although some of the associated cost improvements are nonrecurring. CA-NA net ARPA in the third quarter was $114,000, up 7% versus the third quarter of 2017 and an improvement versus a 1% year-over-year decline in net ARPA in the second quarter of 2018. Excluding American Airlines, our net ARPA increased 16% versus the prior year as I mentioned. CA-NA ended the quarter with 2,712 aircraft online, down 97 aircraft or about 3.5% versus the prior year period. Total aircraft equivalents in the third quarter of 2018 were 2,809, down about 2.5% from the prior period. The cause of this decline was the previously announced de-installation of American Airlines aircraft. We expect the pace of these de-installs to accelerate in the fourth quarter, resulting in about 2/3 of the planned 550 de-installs occurring this year and the balance by mid-2019.

Let's now turn our attention to CA-ROW. Revenue reached $35.2 million, up from $16.6 million in the prior year period, with most of the increase tied to higher equipment revenue, reflecting the impact of ASC 606. CA-ROW service revenue increased nearly 13% from the third quarter of 2017 to $17.7 million as aircraft online increased from 352 to 513. ROW net ARPA grew slightly sequentially to $148,000 from $147,000 in the second quarter, but declined as expected, from $205,000 in the year-ago period from dilution of newer fleets. The percentage of ROW aircraft on new airlines, defined as those installed since 2017, tripled from a year ago to 29% and we increased aircraft online by 161 during this quarter.

Segment loss improved both year-over-year and sequentially as we're seeing the benefits of scale in ROW. Two key metrics illustrate this operating leverage. First, ROW service gross margins improved by 18 percentage points in the third quarter from the prior year period. Secondly, total operating expenses excluding depreciation and amortization declined as a percentage of revenue to 47% in the third quarter from 113% in the prior year period. These benefits were offset by increased equipment cost of sales now accounted for under ASC 606. We still expect that 2017 represented our peak investment year in ROW as segment profit continues to improve from the levels of a year ago. Importantly, we are also seeing improvements in ROW on a unit basis. Take rates for both existing ROW fleets and new airlines improved year-over-year. We do expect blended ARPA to be down due to dilution from new fleets. Combined with the improved net ARPA trends we described in CA-NA, we believe we're setting a foundation for net ARPA improvements globally over time.

I'll now summarize our cash position and capital expenditures. We ended the quarter with approximately $191 million in cash and equivalents, down from $264 million sequentially, reflecting our semiannual interest payment of $50 million. Unlevered free cash flow improved by $14 million sequentially from negative $37 million to negative $23 million, driven by lower non-airborne CapEx and airborne cash flows. In the third quarter, our gross CapEx and cash CapEx decreased both year-over-year and sequentially. Gross CapEx decreased $59 million year-over-year and $43 million sequentially to $9.2 million in this quarter. Cash CapEx decreased $51 million year-over-year and $44 million sequentially to $2.2 million.

In the quarter, the cash CapEx decline was driven by lower equipment purchases and a shift in PP&E classified as inventory based on a shift to the airline-directed model.

Inventory within net working capital increased $31 million in the quarter. While we expect fluctuations from quarter to quarter, our CapEx spending remains on a meaningful downward trajectory. This CapEx trend is largely due to our expectation of more airline shifting to the airline-directed model and reducing airline subsidies. These are declining due to the lower contractual amounts embedded in our backlog and our focus on reducing or even eliminating equipment subsidies in future airline deals.

As we've said, we are in active discussions regarding subsidies with some of our current airline partners. Based on these trends, we expect free cash flow to improve by more than $100 million from 2018 to 2019.

I'll now turn to our guidance for 2018 and offer some perspective on our outlook beyond this year. The highlights of our 2018 guidance are as follows: Total revenue in the range of $865 million to $935 million, unchanged from prior guidance. Gross capital expenditures in the range of $150 million to $170 million and cash CapEx in the range of $110 million to $130 million, both unchanged from prior guidance. Increased 2Ku aircraft online is now expected to be in the range of 450 to 500, decrease from prior guidance of the low end of 550 to 650 for the reasons Oak mentioned. Adjusted EBITDA is now expected to be in the range of $45 million to $60 million, an increase from the prior guidance of $35 million to $45 million.

I'd like to provide some additional context for the relatively wide range of adjusted EBITDA guidance this late in the year and the rationale for providing a low end of $45 million, which is below the year-to-date adjusted EBITDA of approximately $52 million. The primary reason for this relates to potential deicing remediation costs. As Oak described in detail, we are implementing what we believe is an effective solution with the sealant and deflectors. In the unlikely case that this solution proves inadequate, we have backup plans ready to be implemented that are well bounded financially. We won't know whether this additional remediation will be required until late in the year. If we are required to implement our maximum remediation plans, which involve installing top and bottom covers on a broad scale, this would result in approximately $10 million in additional cost in the fourth quarter above our base case.

As we look beyond 2018, we believe adjusted EBITDA is poised for meaningful growth. There are 2 fundamental reasons for this conviction. First, 2018 was a bit of a perfect storm for Gogo. And second, our conviction about the positive prospects for our Commercial Aviation business have deepened as we've developed the Gogo 2020 plus plan. Several factors converged to make 2018 a particularly difficult year. First, with the deicing issue, which impacted 2Ku performance and the customer experience during the first half of the year and led to the significant remediation costs we have discussed. Second is the negative financial impact of American Airlines, which we've described in some detail. Thirdly, we have incurred significant expenses this year for programs which lay the foundation for the future, but which will not require the same level of expenditure in future years. These programs include completing 30 STC programs in 2018, heavy investments to achieve line-fit with Boeing and Airbus, and deploying aircraft for a significant number of new airlines launched during the course of the year. With the major investments for these programs largely behind us, and a company-wide focus on operational improvements and financial discipline, we expect our functional spend to be meaningfully reduced from the peak 2018 levels beginning in 2019. In addition, we completed another round of planning since announcing the results of our integrated business planning process on July 13.

As we stated at that time, that plan was purposely conservative. Its primary objectives were to impose discipline on our organization to ensure we sized our cost structure properly and to frugally manage our cash. There were 3 major areas of conservatism in that plan: Revenue, Satcom cost and the level of airborne equipment subsidies. We have reasons for additional optimism in each of these areas as Oak outlined.

Here are some key highlights from the revised forecast. We expect higher revenue growth based on more realistic assumptions. For example, we previously assumed take rates of 12.6% in 2022 and CA take rates were 12.4% in the current quarter. We meaningfully reduced Satcom costs based on current and expected future market pricing, and as Oak described in some detail our focus is reducing or even eliminating equipment subsidies in future airline deals. We are also in discussions regarding subsidies with some of our current Airline Partners. Solid underlying demand trends, coupled with the operating improvements and cost reductions from our IBP process, are putting the business on firmer footing. We continue to expect solid EBITDA growth in 2019, largely due to improvements in our cost structure and continuing revenue growth, excluding the impact of American Airlines. A consequence of our improved operating expectations is a related improvement in the outlook for our cash position. On our July call, we were projecting a worst-case cash deficit of $50 million to $75 million. While we may choose to add some buffer capital at some point, we are now projecting cash to remain in positive territory going forward.

I'd like to conclude by reiterating our priority for addressing our balance sheet. As we said previously, we've been diligently evaluating a range of strategic and financial options. We also said we plan to address our convert maturities before they go current in March 2019. We're on track to do that as we expect to address our current maturities well ahead of that date. We look forward to updating you on our progress as appropriate.

This concludes our prepared remarks. We're now happy to take your questions.

(Operator Instructions) Our first question comes from Louie Dipalma with William Blair.

You previously said that the cash balance could dip in 2021 to negative $50 million to $75 million, and now you indicated that you expect that, it should stay in positive territory. Does this modified outlook assume any positive impact from renegotiated airline contracts?

Yes, Louie, you've got that right that we had projected a negative $50 million to $75 million balance and it's projected to be in cash positive territory now. We have assumed some renegotiations. We're very close to those. We're in active conversations and have -- are making meaningful progress on those. As Oak mentioned, there is certainly a quid pro quo here by us getting more cash upfront, that it would be a trade-off of providing -- having that come out of future revenues, but we're making good progress there. I think the airlines understand that. We have identified deals that are mutually beneficial and feel good about the progress we're making there.

Louie, that reflects the current status of those negotiations is what I would say.

Okay. And in the press release, you suggested strong EBITDA growth for 2019 and I was wondering if you have any sense of how much of the investments this year for STCs, service bulletins, deicing and other programs would you expect to not recur in 2019? In other words, I'm trying to get a sense of whether there are going to be easy comps for 2019 on the cost side?

Yes, Louie, just to frame that question a little bit, as you know from a previous call, we said we expected to reduce OpEx excluding Satcom by nearly 20% in the 10 quarters from mid-2018 through the end of 2020. That represents about a $75-million reduction. A very meaningful portion of those reductions, in fact, more than half, we would expect to come in 2019 as we see some of these key program investments rolling off, and those include STCs and line-fit like I outlined. So we would expect to see a meaningful contribution to increased EBITDA in 2019 from those -- from the lower cost structure.

Great. And Oakleigh, do you have any thoughts on the Inmarsat-Panasonic partnership?

Yes, I've got a lot of thoughts on it. I guess the impact on us is, I think it'll depress Ku pricing somewhat since Panasonic will be pulling out of the Ku market, so that's a good one for us. I think that the way we view this competitively is Panasonic's been very challenged running its network and has quite poor service and they basically are giving up on running a network and buying Inmarsat service. For Inmarsat, Panasonic is a now another reseller. They've got several other resellers as well, so this is not a 2-way exclusive. So it's interesting industry development. We think it's relatively positive for us from a Ku supply perspective, and I think it also creates some confusion in the market around where Panasonic and Inmarsat are each going and that helps us in the sales process.

And our next question comes from Simon Flannery with Morgan Stanley.

You've got, I think, 9 or 8, 98 2Ku planes on -- installed now. Can you just remind us of where you stand on the backlog and how should we think about the install race in 2019? Should we be in that 450 to 500? Or should you be able to accelerate that? And then I think last quarter Oakleigh had talked about the board asking you to evaluate certain approaches on the strategic side. I was wondering if you had any updates on that.

Yes, so Simon, I -- we're not going to get into giving guidance around the exact number of installs for next year, but I'll give you a little bit of color. First of all, we have delayed installs from this year as we've talked about, which will feather into next year. And then we already had a pretty good install backlog going into next year. So I would expect to run at about the same pace, maybe even a little higher next year then we have this year.

We also, I would say, have -- we've got a lot of the regulatory approvals behind us on the lot of things that delayed things this year so that should also help next year's cadence.

In terms of the strategic options, I said at the beginning of the call that we really weren't going to talk about those but I'll just reiterate things as to where they stand. And we noted before, we got a number of strategic inbounds. The board asked management to pursue those and to investigate them further and negotiate them, et cetera. And I would say we're in the middle of a number of strategic conversations and we'll have comment when we have progress that we can talk about.

And what about the -- on the Rest of the World. What about the pipeline of new aircraft awards, new airlines? How -- is that active at this point? Or is it, are we in a bit of a lull?

It's been very active. I think deals have been slow to be awarded this year. I think airlines are concerned about all the vendors to be honest, so they are being very careful about their selections. We remain active in all the airline deals that I was very excited about on the Q2 call. So we've seen more deals come into the pipeline. I think that if we can make the market feel better about us from a financial security perspective even through something strategic or a refinancing of some kind, I think that would actually help break a lot of deals in our favor because 2Ku itself is flying really well, the reliability is very good, the references from other airlines are very good. So I think we've got positive traction on the product side. We just need to reinforce that we're stable from the financial side as a company and I think will get a couple of those coming our way.

And our next question comes from Philip Cusick with JPMorgan.

Oak, recognizing that you don't want to spend much time on refinancing or strategic, can you tell us if there are options that you've ruled out in either area?

Are there options we have ruled out in our other area? I don't think we have explicitly ruled anything out, Barry. Have you --

Well, we -- on the strategic front, as Oak mentioned, we're pursuing those actively. On the financial front, I would say we've evaluated a broad range of alternatives there. As you know, you get a lot of ideas coming your way in a situation like this and we have really taken them seriously. So I would say that as I mentioned in the script, we clearly recognize the importance of addressing the converts before they go mature and that we're very committed to doing that, certainly, on that timetable we're ahead of it. So we've taken it very seriously and are tracking along with some of those alternatives.

And second, can you quantify any of the EBITDA benefit impact from slower installs this year?

I don't think we have that handy, but I would say that just because the install has been delayed doesn't always mean we've have avoided the EBITDA impact of it because we are working on certifications, STCs, et cetera. So Barry, do we have a number around?

No. And the EBITDA, I would say, is less affected. It's more certainly on the CapEx side, so you can kind of play through the math on our average installation cost. And so I won't give you the exact number, but it's a relatively one-for-one impact on cash as you delay those not completely but that is the case, although from the countervailing piece of that is the inventory build. So you saw that we built inventory and in anticipation of those, so I guess that's the way I would describe it a slip on the installation side but muted by inventory that's bought in advance to that.

And your next question comes from Lance Vitanza with Cowen.

I had 2. The first is, could you discuss what you're seeing on the customer satisfaction metric side? I think in the past you talked about looking at those metrics differently or looking at different metrics and I'm wondering if you could update on how that's proceeding? And then a question about, I guess, the underlying trends in the business seem pretty good when you ex out American. I understand why that was an anomaly, but what gives you comfort as you look forward that a year from now we won't be faced with another such anomaly where perhaps you were to lose another airline partner and we start the cycle all over again?

Okay, so 2 questions. Customer satisfaction has grown a lot over the year as we measure it, and a number of our airline partners measured net promoter scores and those have grown dramatically since the second quarter. So we're not going to share exact data with you, but that's the overall trend. The in-cabin experience is new metric we started, and you are right to point that out, that's also improved dramatically with things like the captive portal implementation that I talked about earlier in the call, and we think that's another one of the drivers of customer satisfaction. So the in-cabin measures are really trying to correlate performance of the system in the cabin, for instance, with page load times, browse speeds, et cetera with customer satisfaction and we do see that correlation and we see improvement in both.

In terms of deals and losing another customer, the big -- sort of our clients we'd have with us today is that they have, what I call, shiny new product clauses like American originally had, and if there's a product that comes along that's dramatically better than what we offer, airlines want the ability to be able to switch because it is going to damage their business. So you can read that clause in our Delta contract, which is a public document, you can find it by going into our 10-Q or 10-K and just linking to it and they have to prove that it's a much better product and is going to harm their business if they don't switch. And they and some others would have that kind of clause in their contract. The big thing to remember, though, is that American, the contract for them was done under a lot of duress in the middle of 2016, I believe, when they decided they wouldn't award us the 550 aircraft. You might remember that broke in the middle of a high-yield bond offering behind the -- between the actual settlement that the pricing and the actual close that blew up this high-yield deal and there was a lot of duress. And in that period, the company agreed to a predetermined pricing structure for airline directed with American. And it gave American the right to trigger that whenever they wanted. And so American decided to trigger it on January 1st this year and that's had a pretty negative impact on revenue. No other airline out there today has that unilateral right. So that I view it as an unusual circumstance and one that won't repeat. We do have the obligation with other airlines to negotiate going into airline-directed in good faith and we would obviously do so, though we wouldn't be forced to do something that was extremely negative for us under our current contract terms. So those are why we kind of think that the American's a one-off and we don't expect to see another -- a repeat of this performance going forward.

Our next question comes from Ric Prentiss with Raymond James.

This is [Chase] on for Ric. I knew you guys noted the gap between vintage and new craft ARPA for CA-ROW. How do you think about the timing of ramping those new installs and ultimately converting on gap?

That's not a 1-week or 2-month process. We have got to get those fleets installed fully first, which will sort of drag into next year and then, because airlines don't really like to promote the product much until the fleet is fully installed, so I think that's kind of a 2-year process to try and get them up to that level of performance.

And just to add to that with that, new airlines coming on all the time that it's -- we would expect to see that dilution from them as they do come on, but the good news is the kind of airlines that we have been adding that Oak mentioned, really some of the marquee names in Rest of World, they have the kinds of planes and routes and passengers that we would expect to see attractive take rates from. So I think that's an important consideration when you really look at the quality of those 500 aircraft that are represented by those airlines in terms of our ability to drive higher engagement rate and drive higher ARPA.

And I'm showing no further question in the queue at this time. I'd like to turn the call back over to Will Davis for any closing remarks.

Thank you, everyone, for joining the third quarter 2018 Gogo Conference Call. The call is now completed. Please feel free to contact me for any follow-up questions. Thank you, and have a nice day.

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