Mark Dankberg on consumer broadband, Eutelsat partnership and solving the aero connectivity puzzle
ViaSat’s February announcement that it was investing $1.4 billion into three terabit-persecond ViaSat 3 satellites has shaken the satellite broadband industry.
For the aeronautical and maritime markets especially, ViaSat’s move is just one of many as service providers secure satellite bandwidth and line up airlines and commercial maritime fleets.
Carlsbad, California-based ViaSat is also exporting its U.S. consumer broadband formula to Europe in the form of joint ventures with Eutelsat, whose own Ka-Sat satellite has delivered mediocre results. ViaSat is paying $145 million in cash to Eutelsat as part of the deal.
ViaSat Chief Executive Mark D. Dankberg talked to SpaceNews about his company’s approach to airline connectivity and how he expects to duplicate in Europe — and Asia’s next — what he is doing in the United States with the ViaSat-1 now, and the highercapacity ViaSat-2 to come.
What is the status of ViaSat-2?
We just completed payload integration and testing with Boeing — a big milestone. The way we get enormous amounts of gigabits is with lots of beams. That means a lot of illuminated paths through the satellite, and they all have to be tested.
Now we’re getting ready for thermalvacuum testing. The satellite will be completed well before launch, scheduled for the first quarter of 2017.
You’re creating joint ventures with fleet operator Eutelsat for consumer broadband in Europe using Eutelsat’s Ka-Sat satellite. You see unrealized potential there?
Yes. There’s basically no consumer retail that Eutelsat runs now. We found in the U.S. that mastering retail is the key: understanding how people perceive your offer, how to present it so it’s attractive, and being sure to fulfill promises.
We can offer promotions like zero down, a few months free, higher-tier-service offer and other things to understand how those drive demand.
Isn’t this Retail 101?
It is. But for us it was a learning experience. Most everybody else chose to have wholesale arrangements with dealers. Some of them — like the ones in Italy — are more sophisticated than in other markets and are starting to figure out those formulas.
If all the dealers were really smart, and sufficiently capitalized to make those retail offers, that wouldn’t be bad for Eutelsat. Eutelsat has found that this won’t happen on its own. They need more control. But it doesn’t mean they are abandoning wholesale.
For us, in the U.S., retail shows the way, then we make wholesale offers.
Our main wholesaler is Dish Networks. They’re very smart. People find it funny [Dish is owned by the same company that owns ViaSat competitor Hughes Network Systems], but we interact with Dish in a constructive way. We trade information about, for example, whether to sell in these census tracks because there is a lot of good-quality cable, or at least with this offer at this time.
Eutelsat has also been concerned about channel conflict. This was an issue for us, too.
Wrestling with the Telcos
We went into the U.S. market and talked to the National Rural Telecommunications Cooperative, which are telcos, and asked, “What should we do?” Their answer: “The market needs a 3- or 4-megabit per second service. DSL is not a lot faster but this is not as good as DSL so it has to be cheaper.”
That is a terrible value proposition for us. We said, “Let’s make it a 12-megabit service that’s a bit expensive — but good.” And they’d say, “We don’t want that.”
The reason is that a telco would rather put a customer on a poor DSL service — selling a service on assets they own — rather than wholesale somebody else’s. The variable cost of adding subscribers to their own networks is almost nothing.
So there is a conflict. If we go out there with our own offer, they’ll see the market has changed and they’ll adapt. They may not sell it enthusiastically, but they will offer it. Having the retail offer that we satellite operators create — assuming the market is there — will create an incentive and people will sell it. That component has been missing in Europe. I think it would unlock a lot of value. The financial results of Ka-Sat can be substantially improved.
That would explain the price you paid?
Yes. Suppose you don’t have that market feedback. You have, say, a 12-megabit offer at 30 euros and 10 gigabytes. You say, “This should sell in Europe.” And for whatever reason, you didn’t sell as many as you thought.
The natural reaction is, “Of course it’s not selling, it’s satellite, nobody wants it, the offer isn’t good. So let’s cut the price, offer more gigabytes, and increase the speed.” But maybe we got exactly the response we should have given the calories we put into promotion.
It’s the ARPU, not the subscriber tally
Eutelsat’s Ka-Sat is running out of capacity in high-demand regions, and ViaSat-3 doesn’t arrive until 2020-2021. What to do in the meantime?
Just what we’re doing in the U.S. We have not been adding many subscribers but we’ve been able to increase the satellite’s revenue and yield. Our EBITDA has been going up. We can do that in Europe and get a significant improvement.
Churn is a bad thing because you lose subscribers. But it’s a good thing because it frees up bandwidth. You can bring that freed-up bandwidth to the market to create more value, whether it’s aero or other plans on the subscriber base, without being nasty to subscribers. Churn offers an opportunity for constant renewal of the way you use your subscriber base.
So if Ka-Sat has nearly full beams in two language basins, you can use churn to generate incremental revenue?
To offer better and more consistent service. It has taken awhile for people to get comfortable with our emphasis on quality, not the number of subscribers. But the effect is you’re offering high-priced, higher-value plans. That raises [the Average Revenue Per User] and raises your results.
So the full-beam issue on Ka-Sat isn’t a problem? You can work it out without new capacity?
I’m not saying we don’t need new capacity. I’m saying let’s figure out what we can do, and then determine our capacity projections for these markets for a certain number of years. Then, with Eutelsat, we can decide. We both have said we are evaluating the interim capacity issue.
The Eutelsat play doesn’t include much aero. Can you change that?
It’ll take awhile but we have opportunities. Since they don’t have an offer directly to the airlines, it’s hard for them to sell. They’re dependent on someone having an equipment offer; they can’t go out and make it happen on their own. We can do that. They could work with others, of course, but I think they’re happy to work with [our equipment].
We have El-Al Israel Airlines close to coming to market in Europe. It will set an example just as JetBlue is in the United States. It’s a handful of airplanes, but it can set off a chain reaction.
Finding an airline connectivity business model
The aero market seems full of promise but still uncertain about the business model. Have you identified a winning formula?
There are conflicting claims from credible players in aero: Inmarsat, which has dominated aero connectivity for 30 years; Panasonic, which dominates the in-f light entertainment industry; Gogo, the pioneer in the U.S. — they’re all credible. It’s confusing.
What does a “better” service mean? Gogo would say, “Our service is better if the few people who used it were happy.” On JetBlue you have 150 people on a plane using it, so maybe this is what better means.
We found that video streaming is a notion of better that really strikes a chord. And it wasn’t until the watershed moment — when Virgin America did Netflix — that it became clear.
Global Eagle Entertainment says you can cache it, you don’t need to stream it.
You can. Caching lets you brand the content with Amazon or Netflix, although Netflix does not have downloadable rights, so there’s an issue there.
Our pitch is, “We’re not giving you Netflix or Amazon, we’re giving you the Internet.” You can’t cache the Internet. Even if you tried to cache the top 80 percent of Netflix or Amazon, the resolution you’d get on your screen is not very good.
EMC has been caching top Internet sites for their maritime service. They say it’s working well.
It’s a valid argument. Certainly on a ship you have more space to put big banks of storage on board.
Our strategy is to have the Internet. If you work with us you should take advantage of the fact that you, as an airline, have the Internet and appeal to people in that way. One of the things we’re doing is live sports.
A direct television feed?
Southwest Airlines, which has terrible Internet, has said people really like their TV. They use Row 44 with a broadcast receiver. They have to pay for their channels. But use a Sling or a LiveTV and the airline no longer has to pay for those channels. Airlines have learned that Internet media companies are excited about putting their brand in front of them. With caching they’ll see a low resolution, which could create a negative experience.
The roster of new high-throughput satellites is long. Is there too much capacity despite the growing aero and maritime demand?
Two things drive demand: more passengers using it, and each passenger using more. That means more people streaming, on bigger screens.
The Panasonic model is really big displays built into seatbacks with the resolution of an iPad. I don’t know if the airlines are going to want to install seatback screens with iPad resolution. And if people have their own iPads, then Panasonic is conflicted.
Our view is that we don’t have enough bandwidth, even if we have more than anyone else.
Look at the lawsuit that American Airlines filed [against Gogo, since withdrawn] saying that up to 20 percent of passengers will switch airlines for better Internet or WiFi. That surprised us; nobody in the airline industry had said it before.
We had been inferring something similar from JetBlue’s financial results and their passenger revenue per seat mile. People prefer JetBlue. Inflight streaming has increased their pricing power.
But the airlines are still very confused about that. Ultimately the financial results will tell the story.