Streaming Over Satellite: Companies Make Their Way Gently

Theresa Foley

A year ago, the excitement about streaming–a new Internet-based business that promised to eat up lots of satellite and terrestrial capacity and provide the basis for dozens of new companies offering services in space and on the ground–reached a frenzy state. Analysts, executives, entrepreneurs, investors and, yes, even journalists, anticipated a tidal wave of demand for a new breed of satellite services to deliver video streams around the world for Internet users. Competition for investment, partnerships and customers was fierce, and anecdotal evidence and marketing studies led many to jump on the streaming bandwagon before it was too late.

But the streaming tsunami never arrived. By spring 2001, the streaming shore had become littered with folded companies, laid off employees and shape-shifted business plans. Is satellite streaming a dismal failure, or is it just taking far longer to arrive than anyone anticipated?

Now that everyone has sobered up, it’s time to take a realistic look at some of the streaming sector changes and progress of the last year. The positive news is that two crucial areas for the overall Internet streaming business have improved in 2001. First, the number of broadband connections in the United States that will enable streaming are up significantly, and second, a few business models to earn money from streaming have taken shape, along with the technologies to allow collection of payments.

In the neighborhood of seven million broadband connections were in place in North America between the DSL and cable companies by mid-2001. Cahner’s In-Stat counted 6.8 million U.S. broadband users at year-end 2000 and estimated that would grow to 19 million by year-end 2002. Steady growth for broadband installations is predicted to continue. Another research house, Multimedia Research Group of Sunnyvale, CA, estimated that worldwide, 15 million users will have broadband Internet connection in 2001, with that number to grow to 30 million in 2004. An additional 24 million have office hookups with broadband, the group says.

On the enterprise side, businesses are still engaged in a lot of trials of streaming video and audio–mainly to support training and internal communications but also moving into marketing, public relations, investor relations and product launch uses. But more businesses are adopting streaming today and this recent growth trend will go on for another three years. In April, Jupiter Media Metrix reported that enterprise spending on streaming in 2000 was $140 million and will grow to $2.8 billion in 2005. Streaming Media Inc., a company that closely tracks streaming developments for its conference and publishing business, estimated that corporate streaming would be worth $318 million in North America in 2001.

Cahners predicted in August that corporate use of streaming would produce $5 billion in annual revenues by 2005 as the slowed global economy drove business to use streaming to replace travel and ease the management difficulties posed by reduced staff levels. The report states that 21,000 corporate sites would be ready for streaming this year, and by 2005, more than 225,000 sites would be paying for streaming services. Hughes and other satellite-based providers are at the “top of the food chain for providing streaming media and content delivery services to market enterprise and medium-business customers,” according to Cahners.

Northern Sky Research, the consulting firm founded by former Pioneer Consulting Analyst Chris Baugh, also reported on satellite content distribution companies this summer. Baugh found that by year’s end, satellite content delivery networks (CDNs), which do streaming among their other services, will take in $145-175 million, of which streaming accounts for 40 to 50 percent, and this would grow to $2-4.1 billion in 2006.

“Demand is there, whether it is consumers tuning into Webcasts or enterprises pushing financial information,” says Baugh. In the near term, enterprise use of streaming will make up 90 to 95 percent of the market, he believes. Financial services have arrived first, with medical content, educational content and corporate Intranet markets coming along behind.

Business models to create revenue opportunities around streaming also are being put into action. In another report, Streaming Media analyzed the return on investment that companies are seeing from streaming initiatives with a survey of 111 corporate users of streaming who were spending roughly $1 million each this year on the technology. It concluded that while most firms just started streaming in the last 18 months and are still in trials, early adopters are seeing quantifiable economic returns along with the intangible benefits. Early uses of streaming are pretty basic, implying that a more mature business will hold huge opportunities for new services.

“Few companies have begun using the more sophisticated technologies that enhance the value, and tame the complexity, of managing digital media,” Streaming Media reported in its return on investment report. Vendors have an opportunity to offer asset management, rights management and tracking and measuring services, advanced tools that are currently missing from the mix.

Streaming Media found that streaming is costing the companies an average of $31.35 per hour. The average cost to serve a person a stream was $25.64, based on the total number of 3.4 million stream-hours to more than 4.1 million people reached by the companies surveyed with their streaming events. These metrics allow a corporation to measure the value of streaming against other investments such as buying new computers or traditional types of training for their workforce. The survey of the 111 corporate streaming users provided a few other trends from the past year: 45 percent more companies streaming in 2001 than 2000; 86 percent increase in corporate streaming spending; 35 percent increase in people served; 230 percent increase in hours delivered; 465 percent increase in bytes transferred.

On the consumer side, the first paying customers for streamed content–beyond the pornography market–have been signed up. In July, Real Networks, one of three companies providing the software needed to play streamed content on desktops, announced that it had signed up 300,000 paying customers for its RealPlayer Goldpass subscription service for streamed material in the year since the service was launched. The service costs $10 a month. Goldpass has programming from the NBA, major league baseball and from CBS, among many other content offerings. Real Networks says that of 50 million U.S. homes using media players on their computer, its software was on more than 29 million of the computers, plus another 11 million business-owned computers.

The National Basketball Association is another pioneer in putting streamed video onto many of its 52 Web sites, as is the House of Blues, an owner of concert halls and other music-related businesses in the United States.

Baugh says new streaming businesses in Europe, such as Europe On Line and Eutelsat’s Open Sky service, were just being launched and provide another optimistic sign.

With the financial services sector representing much of the early use of Internet streaming, the use of streaming by Corporate Communications Broadcast Network (CCBN) of Boston, a leader in the financial Webcasting sector, illustrates both the opportunities and the problems facing satellite providers. CCBN hosts the investor relations portion of Web sites for 2,500 public companies (of 5,500 publicly traded U.S. companies) and carries regular Webcasts for most of them, with the help of content distribution network partners Akamai and iBEAM. CCBN also has portal partnerships with retail Web sites like America Online (AOL), Yahoo! Finance and The Motley Fool, giving it further outlets for its content and making it one of the leading sponsors of financial information on the Internet.

CCBN reported 200 percent growth in the volume of financial Webcasts it carried from July 2000 to July 2001. Together, the terrestrial content distribution network company Akamai and CCBN delivered nearly 6,000 Webcasts in the first six months of 2001 to a live audience of more than 1.25 million. The number of listeners was 10 times the previous year. On-demand listening also went way up, with 3.5 million listeners in the first six months of 2001 compared to only 100,000 listeners of archived material in the previous year’s first half.

The CCBN numbers show the strong growth in the number of Webcast events and users but so far CCBN and its clients make little use of the satellite services because nearly all the Webcasts are either pure audio or audio with slides or Powerpoint presentations. The bandwidth required generally does not mandate satellite distribution. If companies adopt a video format for their financial Webcasts, demand for satellite services will rise; but, to date, few corporations have even tried video for their investor relation activities, largely because of the high costs involved.

CCBN launched a video Webcasting service in May that is designed to be interactive with the audience. Shawn Dornan, CCBN manager of streaming media applications, says that as of August, the video service was not in big demand. “The interest is there but it’s still pretty early in the life cycle for critical mass, mainly because it’s cost prohibitive. It could cost $15,000-$20,000 to do video, versus $1,500-$2,500 for audio,” Dornan says, citing average costs for a one to two hour event.

Dornan says satellites play a critical role in getting the signal from the financial Webcast from its source to the broadcast center when a customer wants video with a Webcast. But as long as the customers stick with relatively low bandwidth audio formats, satellites are not likely to be a crucial link in the chain.

Interactivity will also expand the Webcasting business, but the problem there is that technology for interactive Webcasting doesn’t scale up easily to large audiences, Dornan says. The products generally max out at 2,000 to 2,500 users. “For most companies, this represents plenty of seats in the room, but for the larger companies to really start to take this to the next level, the Ciscos, the AOLs, the Qualcomms, to get tens of thousands of participants, the technology providers will have to get greater levels of scalability.”

That’s Entertainment

As streaming ambles slowly toward being a mainstream service, the list of satellite companies pursuing the business has shrunk and the lead positions are changing. Just who was in front in summer 2001 was debatable. Early leaders Cidera and iBEAM’s investment troubles and layoffs this year have slowed their momentum considerably, leaving Loral Cyberstar and Panamsat Net 36 as the more stable bets for satellite streaming services. The two have relatively deep pockets, even though both of their owners also had less than stellar financial performances this year. In any case, Loral and Panamsat placed their bets on entirely different sectors of the streaming market, and thus are not likely to collide as competitors any time soon.

“A year ago I was accused of being too late,” says Rob Bednarek, Panamsat’s executive vice president, recalling the perception that iBEAM, Akamai and Digital Island had the market for content distribution around the Internet locked up. “In contrast to enormous investments others have made, that have been wasted, I think we are perfectly positioned for the take off of streaming.”

Net 36 is aimed directly at the entertainment and media sector, the same customers to whom Panamsat sells its broadcast distribution services. Net 36 was Panamsat’s offer to use satellites and a PAS-owned edge server infrastructure to move video content closer to Internet users, announced in spring 2000. It took PAS a $60-million investment and more than a year, to June 2001, to announce its first Net 36 customers. EWTN Global Catholic Network, Irondale, AL, the largest religious programmer in the world and an existing PAS satellite user, will use a Panamsat Galaxy satellite to distribute its religious programming to Internet viewers after being encoded for IP delivery at a PAS center in Atlanta.

Net 36 has deployed its servers into networks in major U.S. markets owned by service providers like BellSouth and Qwest. PAS estimates that this network will reach nine million U.S. broadband homes by year end through these partnerships. Bednarek prefers to describe Net 36’s size in terms of the number of 300-kbps streams that can be supported simultaneously, which are 45,000 concurrent streams from the edge.

“It has come a lot slower than anticipated, but there is a lot of activity going on,” Bednarek says. “Particularly the larger media companies are looking at streaming as an extension of core video distribution work.”

By mid-summer, Net 36 also signed contracts with divisions of Viacom, AOL Time Warner, News Corp., MPWeb, a European sports content aggregator, and Video Italia, a European music video provider, but had released no details on any of those customers, who are ramping up their services with Net 36. Bednarek says more than 66 percent of the top 40 streaming Web sites for broadband belong to current video programmers. CNN is the top streaming site. Part of Net 36’s challenge has been to help the programmers find ways to charge for content. In June, three features were added–digital rights management and the ability to offer pay-per-view and subscription services. With its customers now able to charge for a stream, Panamsat also should be able to collect for its services, although the streaming rate cards will have a much different basis than what Panamsat uses for broadcast distribution. Net 36 customers will pay on a bits delivered basis, following the practice used to buy and sell both Internet and commodity bandwidth. Without having large volume, signed lease deals for future commitments of capacity, Net 36 is likely to take several years before a strong, visible revenue flow is established.

With the shakeout that has occurred in the telecom services sector, Net 36 also has faced cutthroat price competition for the type of services Net 36 has offered, Bednarek says. Especially as other satellite companies and terrestrial Internet service firms have run into bankruptcy situations.

“To build market share, people have priced services below cost. But a lot of those guys aren’t around anymore… A company can’t sustain pricing its service under cost for very long. One of most significant things that has happened over last six months is a settling down of the market. Net 36 and [the surviving] competitors are now pricing at cost or above cost to make profit,” Bednarek says.

The philosophy of how many broadband users are required before Internet streaming can deliver a critical mass of viewers also has evolved as Net 36 took shape. Earlier, top PAS executives believed that 10 million homes passed–half the number required for a cable company to profitably serve a mid-sized market–was needed to give Net 36 sufficient market reach for content owners to sell advertising around the viewer base. Now the 10 million home figure is less relevant, Bednarek says, as programmers have come up with alternative revenue models to an advertising-centered plan.

Potential users like CNN and ESPN are trying out pay-per-view models for their Web sites. “The big guys are figuring out that they need streaming. It’s part of their plans and companies are figuring out how to get paid for it,” Bednarek says.

He points to a recent CNN Headline News innovation. Headline News has begun to run a small screen video window with banners on its regular broadcast program. Bednarek says the window inside the regular broadcast resembles what a viewer might watch on a computer screen, and he expects to see CNN start streaming the small video in an Internet-ready format. He believes CNN will have almost no incremental cost if it decides to move the small screen video onto the Internet, and thus the service won’t need as many potential viewers as a cable system to make economic sense to the newscaster.

With pay-per-view up and running on Net 36 as of August, Bednarek was optimistic that the Net 36 customers are soon to start collecting for their Internet streams. The pay-per- view business model is well understood on the Internet, he says, generating $500 million in revenue last year. While $470 million of the total comes from pornography, he notes, the other segments are growing. Net 36 has no policy on carrying pornography and PAS will refrain from making any judgments on the content of any streamed services, just as it does for its regular broadcast services.

An Enterprising Opportunity

Loral Cyberstar, meanwhile, will ignore the consumer streaming business and the video programmers in order to focus its energies on enterprise networks and its existing customer base, which it has spent a decade cultivating.

Cyberstar’s streaming initiative, branded as Clearstream, went into service in May. Clearstream describes a family of IP-based services that feature streaming in several forms and is targeted at the enterprise-streaming user, says Dave Puente, Cyberstar vice president and general manager for business media services.

Corporate users can distribute their rich media, including streams, in one of several ways, including live, on demand, over the Internet or within their private networks, using Clearstream’s services and satellite dishes on their premises. One of the Clearstream services, called Clearstream Overnet, is designed to expand the capacity of a private corporate network with an overlay of satellite capacity. Clearstream customers will be able to receive high quality video on their premises via a satellite dish that then connects the stream to desktop computers in the office or other devices. Cyberstar’s Clearstream services enable content to be cached on the enterprise network and not through a network of servers deployed around the Internet, as is central to most satellite CDNs.

As of mid-summer, the only two announced customers for Clearstream’s streaming services were RAW, a London-based global financial services provider that is using the service to distribute information to institutional investors, and the Christian Church, which used Clearstream to send a religious program to members in 500 locations last spring.

With 100 satellite network customers already being served by Cyberstar, the plan is to convince many of those to convert to IP and Clearstream as they move their business TV operations from conference rooms to the desktop. The company brought in $130 million in revenue in 2000 based on its existing network services.

“We believe this is a very hot area in the market in terms of companies trying to figure out how to take existing infrastructure and enabling it to do more applications cost effectively. We believe satellite’s ability to do broadcasting and multicasting is creating the opportunity for us. What is lacking in the marketplace is people trying to figure out how to integrate IP technology with satellite technology,” says Puente. As an integrator with lengthy experience in tying satellite networks to terrestrial ones, Cyberstar hopes to have an advantage with big enterprise users.

Clearstream’s success will be measured not in terms of how many streams it can transport but by how many existing customers adopt its services and how well Cyberstar does in convincing new customers to deploy it, Puente says. But this success is predicated partly on how much these customers have invested in local area networks and workplace computers. Even though most computers today can handle video, Puente says, many of them don’t have the necessary soundcards, headphones or other peripheral equipment to adopt desktop streaming immediately. Information on how prevalent are business desktops that are ready to use streaming is difficult to come by; in terms of sheer numbers, however, Cyberstar estimates that three to five million desktops are connected to its network today through existing customers.

Puente looks for a jump in enterprise spending on streaming networks in early 2002 when he expects corporations to begin investing again in infrastructure as the tech sector slowdown of 2001 begins to turn around. Outside the tech sector Puente says Cyberstar’s customers in finance, health and other sectors have not stopped investing in their networks.

The Entrepeneurial Spirit

Panamsat and Loral apparently are planning on riding out the streaming chaos, but what of the more entrepreneurial satellite streaming projects? iBEAM and Cidera have been at the forefront for several years, and as of late August, both were still in business and claimed to be moving ahead.

iBEAM found a white knight to step in with an investment in late June when broadband services company Williams Communications led a group making a $40 million investment to get a 60 percent stake in the provider of satellite-based streaming services. iBEAM would have run out of operating cash in third quarter 2001 without the Williams’ infusion.

Officials from the companies declined to discuss their operations or plans for this article. iBEAM, founded in 1998, offers a range of streaming communications. iBEAM claims to deliver more than 100 million audio and video streams each month over an intelligent network of servers connected by satellites and located in more than 210 networks around the world. The iBEAM network was to be integrated with Williams’ 33,000 mile fiber optic network.

Earlier this year, iBEAM officials said the company’s network was able to transmit one million simultaneous streams to end users. iBEAM’s architecture uses leased satellite capacity to multicast streams to 190 sites, located mainly at U.S. ISP premises, where 1,000 iBEAM Maxcaster servers are deployed. Besides the ISPs, iBEAM’s equipment is in massive Internet data centers owned by Internet hosters like Exodus and Abovenet. iBEAM also is delivering streams to broadband service providers such as Excite@home and BellSouth, who then redistribute them to their end user customers. Like Net 36, iBEAM’s primary customers are the content owners, big media companies like MTV Interactive and MSNBC.

Cidera, meanwhile, has sufficient investment to operate well into 2002 and announced a new multicasting service in July. Cidera has been delivering Internet content into 500 networks and 10,000 business locations in the U.S. and Europe for several years. In July, Cidera launched a multicasting service for large-audience streaming that delivers streams by satellite to ISPs. This service uses an alternative approach to the concept of putting servers out on the edge of the network. Bob Jones, Cidera vice president of sales, says the service requires no capital expenditures or servers, but instead, the ISPs need to be multicast enabled, so they can receive a multicast transmission of a stream, as opposed to a unicast.

Jones says Cidera will soon announce a big content owner customer and that many of the 500 ISP clients of Cidera’s also had agreed to use the multicasting service, although no customer names were available.

Cidera Multicast’s advantage is that the cost will remain constant no matter how many viewers watch a stream, in contrast to other content distribution services that charge for the amount of bandwidth consumed.

A main reason streaming hasn’t caught on is the required investment in servers, Jones says. “With this multicasting service, Cidera is well positioned to stream through its network by eliminating the need for servers. We are developing a new business model for content distribution,” he says.

Cidera is allied with Akamai, Network Appliance, Cisco and CacheFlow, all of which are involved in Internet CDNs. These partners aggregate content and work with Cidera to push the streams, whether live or on demand, to numerous locations around the world for further distribution to end users over the local networks that provide Internet connections to the public. In 2000, Cidera built its network from 60 to 500 points of presence, and in 2001, the company has been putting its business model into practice.

Fad Or Future?

Who has disappeared in the last year? Edgix and Orblynx folded, and others like Net 36 cut their previous spending projections. Interpacket was unable to complete a planned IPO and instead was acquired by Verestar, a big teleport and global satellite services firm. Cidera laid off half its 300 employees by early in the year and restructured to move its profitability date up as part of an agreement with investors who put another $75 million into the company.

Bigger operators like Intelsat are showing no interest in leaping into the fray, and other big operators with strong Internet backbone connectivity service divisions appear to be waiting it out before embarking on new streaming service initiatives. Intelsat considered a streaming media service but decided against it. “Today, we can’t see how it’s going to make any money for us,” says Graeme Kennelly, Intelsat’s business line manager. Intelsat carries streaming in its regular mix of Internet traffic in the backbone network connectivity service but will not set up a dedicated streaming operation. “Looking at the market today, it’s going to take longer and will be a lot less than the early reports indicated.”

Chandy Nilakantan, Skystream Networks chief technology officer, commented, “Six months ago there was a great amount of interest and terrific momentum. There were a lot of companies in streaming on the technology and the service sides. When many of these companies disappeared, people began to wonder whether streaming on the Internet is a fad or whether it will be part of our future.”

Skystream sells hardware and software that enables service providers to build infrastructure for streaming and delivery of other multimedia rich content, like MPEG-video, directly to consumers or to ISPs. Its zBand content delivery software platform is aimed at satellite and broadband service providers who need to manage their services, including streaming to consumers, businesses and other ISPs. Speedcast, a Hong Kong-based broadband services provider that is offering edge caching and content delivery services to ISPs and enterprises, is a Skystream zBand customer, using the equipment to deliver streaming content to consumers over existing satellite networks. NBC used the gear in a broadband streaming trial to send video to Internet users via ISPs, but the broadcaster had not decided to adopt it commercially as of August.

Nilakantan believes streaming is here to stay especially for enterprise customers where it can be leveraged for smaller audiences that require specialized content like corporate training. For consumer streaming, the problem has been a lack of sophisticated content aggregators who can understand the audience’s needs and pull together a compelling package of content that meets them, Nilakantan says.

Proof Of Life

Streaming has a few more years of development ahead before the twin problems of sufficient broadband connections and revenue models to collect payments for content are totally resolved. In the meantime, satellite streaming service ventures will face lean times and tough sales calls until they can prove to content owners and investors that their technology is key to a more efficient global streaming network.

Theresa Foley is Via Satellite’s senior contributing editor.