The set-top box: Costs and benefits

May 24, 2002 12:00 PM, Steve Hawley

In my April column, I made an appeal for feedback from any service providers that have deployed TV, wanting to hear of their successes, challenges and unexpected situations. Several of the messages I received included comments about, among other things, striking the right balance between cost-per-subscriber, selecting the "best" technology and maximizing revenue. (By the way, please write me if you haven't already - tell me your stories!)

Taking that feedback as my cue, this month's column represents the first of an occasional series that will examine the issues and tradeoffs that service providers will experience as they move from being traditional voice and data carriers toward becoming full service providers. Nobody said it would be easy to implement and launch a twenty-first century media delivery ecosystem that serves the needs of today's customers. As you have seen, the many interdependencies involved in this communications, entertainment and information services delivery ecosystem make it difficult to know where to start.

This time, we'll talk about the primary piece of customer premises equipment that makes TV "happen" for the subscriber: the TV set-top box. Cable and satellite subscribers take set-top boxes for granted, but few know what really goes on inside the unit, other than having vague assumptions that the box is some kind of signal de-scrambling device, needed because the cable operator is worried about theft of programming. That's true, but it's just the tip of the set-top iceberg.

First of all, consider that a set-top box is essentially a computer. Lifting the hood, you'll find a motherboard and power supply, a microprocessor, input-output, signal processing, firmware, memory, an operating system, middleware and, sometimes, applications software. Pretty much the same stuff that's in your office PC, in my Mac laptop and in the current generation of game consoles. A natural reaction to this is, "Gee, it must be expensive." Yes, set-top boxes have the potential to be very expensive, depending upon what they can do and how many units the service provider plans to deploy. And that gets to a core issue surrounding the set-top box: the tradeoff between cost-per-subscriber and the benefits afforded by enabling technology.

The question becomes: "Realistically, how much revenue does this technology enable me to capture?" Any TV set-top box enables the service provider to offer regular MPEG-2-based digital TV, "live" network, specialty and premium TV programming. This is likely to be the first, and sometimes the only, "converged media service" that most service providers will offer. A low-cost set-top box can be used if TV is the only goal--regardless of whether the service provider offers 129 channels or only thirty. The number of channels in the system happens to represent several other sets of technology choices and capital investments that are made in other parts of the TV delivery ecosystem, but it has little impact on the choice of set-top box. A TV-only set-top box can cost the service provider as little as a hundred dollars, or more than three hundred, depending largely upon quantity, on-board network connectivity and expandability.

If the service provider plans to offer just digital TV using a $180 set-top box, and charges $29.95 per month to the subscriber, it will take six months to recapture the cost of that subscriber's set-top box. A predictable six-month payback is pretty good, given that telcos can have twenty-year amortization schedules for some capital expenditures! 

A corollary to our enabling technology question is, "How quickly does the service repay the incremental investment?" (Or, "How much can I charge for this service, minus how much extra do I have to pay to enable it?") The answer is based upon how many other services the service provider might eventually offer. Let's look at two additional services that have little impact on the cost of the set-top box. Video-on-demand is typically implemented at the service-provider's end, not requiring any storage on the set-top box. Another is home networking: Some set-top boxes act as home network residential gateways. If TV service yields thirty bucks a month to the service provider, ten to fifteen for VOD and another ten per month for home networking service, and a home-networking-enabled set-top box is just 10% more expensive, it sounds like it could be a good deal, and easy to justify.

Obviously, this excludes other capital costs that must be amortized, such as the cost of the "head end" (the multimillion-dollar facility that receives "live" TV programming and prepares it to be moved over the service provider's own network to subscribers), the cost of video servers for VOD, and the cost of the programming, but still serves as a basic measure.

Other ventures might be more speculative. To date, personal video recording (PVR, or "TiVo functionality," just as "Xeroxing" equates to photocopying) has been a relatively small business and has come under increasing scrutiny by those who would protect copyrighted content. Yet it may involve local storage - a hard-drive that raises the cost of the set-top box. Another speculative idea is to offer a set-top box that includes a DVD player. On the plus side, a mature video-on-demand service might provide an option for the subscriber to purchase the video via a menu on the TV screen, yielding some margin to the service provider for each DVD sold, plus some added convenience to the subscriber (not to mention a space savings for the subscriber, no longer needing a separate DVD player). To support both of these services, a "fully loaded" set-top box that includes a hard drive and DVD player can run as high as $800 or a thousand dollars per unit.

Another issue is a service provider's concern over the opportunity cost of advancing technologies? On my personal computer, I can run software that's two or three "revs" behind the current release - just as I can still watch TV via my three-year-old set-top box. But if I want to take advantage of the latest functionality, sooner or later I'll need to upgrade the hardware. Similarly, if the service provider offsets the cost of a low-end set-top box in six months, and the time between set-top box hardware upgrades is two and a half years, then there's still two years of profit (or, actually, zero additional cost) on the chosen set-top box. Spending more up front on the set-top box may not make much of a difference if it enables a service that's planned for eighteen months in the future.

Of course, you'll recognize that there are other set-top related decisions. Is the set-top box compatible with the TV middleware I've chosen? Does it support the kind of content protection and security that's required by my video-on-demand movie supplier? Also, there are applications I haven't mentioned, such as targeted advertising, certain implementations of which involve local storage in the set-top box.

In summary, the set-top box decision is one example of how service providers will need to give serious thought to all the tradeoffs between financial commitments, technology choices, the specific components of the system, which services to introduce when, where the content will come from and so on. It can be pretty scary for a traditional telco to see its local cable TV operator in the rear-view mirror, knowing that these decisions and tasks lie ahead and not knowing quite where to start. 

Perhaps choosing the set-top box first is like putting the cart before the horse. A service provider will only begin to get the upper hand on a TV deployment by thinking about "requirements." As in, "What do I want my service to look like, what are the business goals and will it beat the competition?" A clear set of requirements will go far toward helping the service provider sort out all these issues. We will talk about requirements next time.

P.S. Speaking of promises kept: Last month, I promised that I'd be profiling an additional service provider this time.  Please watch for it in the near future.

Steve Hawley is principal consulting analyst of Advanced Media Strategies. He may be reached via his Web site, http://www.tvstrategies.com.