I saw an article earlier this year that said that some smaller triple-play providers have decided to get out of the cable business. Specifically the article mentioned Ringgold Telephone Company in Georgia and BTC Broadband in Oklahoma. The article said that small companies have abandoned over 53,000 customers over the last five years, with most of this being recent.
I’m not surprised by this. I have a lot of small clients in the cable business and I don’t think any of them are making money with the cable product. There are a myriad of outlays involved such as programming, capital, technical and customer service staff and software like middleware and encryption And all of these costs are climbing with programming increasing much faster than inflation. And there is pressure to keep up with the never-ending new features that come along every year like TV everywhere or massive DVR recorders. I have a hard time seeing any cable company that doesn’t have thousands of customers covering these costs.
But small cable providers are often in a bind because they operate in rural areas and compete head-to-head with a larger cable company. They feel that if they don’t offer cable that they might not survive. But it is getting harder and harder for a company who doesn’t have stiff competition to justify carrying a product line that doesn’t support itself.
I’ve written several blogs talking about how software defined networking is going to change the telecom industry. It is now possible to create one cable TV head-end, one cell site headend or one voice switch that can serve millions of customers. This makes me ask the question: why isn’t somebody offering cable TV from the cloud.
There are big companies that already are doing headend consolidation for their own customers. For instance, it’s reported that AT&T supports all of its cable customers from two headends. A company like AT&T could use those headends to provide wholesale cable connections to any service provider that can find a data pipe to connect to AT&T – be that a rural telephone company, a college campus or the owner of large apartment complexes.
This wholesale business model would swap the cost of owning and operating a headend for transport. A company buying wholesale cable would not need a headend, which can still cost well over a million dollars, nor technical staff to run it. In place of headend investment and expense they would pay for the bandwidth to connect to the wholesale headend.
As the price of transport continues to drop this idea becomes more and more practical. Many of my clients are already buying gigabit data backbones for less than what they paid a few years ago for 100 Mbps connections. The only drawback for some service providers is that they live too far of the primary fiber networks to be able to buy cheap bandwidth, but the wholesale model could work for anybody else with access to reasonably priced bandwidth.
The wholesale concept could be taken even further. One of the more expensive costs of providing cable service these days is settop boxes. A normal settop box costs over $100, one with a big DVR can cost over $300 and the average house needs two or three boxes. The cost of cloud memory storage has gotten so cheap that it’s now time to move the DVR function into the cloud. Rather than put an expensive box into somebody’s house to record TV shows it makes more sense to store video in the cloud where a terabit of storage now costs pennies. Putting cable in the cloud also offers interesting possibilities for customers. I’ve heard that in Europe that some of the cable providers give customers the ability to look backwards a week for all programming and watch anything that has been previously broadcast. This means that they store a rolling week of content in memory and provide DVR service of a sort to all customers.
The ideal cloud-based cable headend would offer line-ups made up of any mix of the channels that it carries. It would offer built in cloud DVR storage and the middleware to use it. I think that within a decade of hitting the market that such a product would eliminate the need for small headends in the country. This would shift video to become a service rather than a facility-based product.
There would still be details to work out, as there is in any wholesale product. Which party would comply with regulations? Who would get the programming contracts? But these are fairly mundane details that can be negotiated or offered in various options.
It is my hope that some company that already owns one of the big headends sees the wisdom in such a business plan. Over a decade, anybody who does this right could probably add millions of cable lines to their headend, improving their own profitability and spreading their costs over more customers. AT&T, are you listening?
That headend in the cloud is called Netflix, Hulu, apple, amazon Prime, etc. Most ISPs report over half of their traffic is from these “providers”. The pure video content via coax cable, as these providers have learned, is out of date. KISS: Time to deliver a simple IP-only plan.
Unfortunately the numbers do not yet support that, although that is where we are probably eventually headed. The vast majority of TV watching is still of content that is delivered over the cable networks., While on-line viewing is growing rapidly, it is still only a tiny fraction of total TV viewing. Until the content providers of all of the shows that people really want to watch are available live and online then TV stays very relevant.
If you are in a market and competing with one of the large cable companies and you don’t have a cable product, then the large company bundle is still going to attract the majority of households. People want to watch their sports and news and the hundreds of shows that are only available through cable.
If all TV content was shifted to the web we would probably need a tenfold increase in the capacity of the web infrastructure to handle it. Netflix is big, but the traditional cable networks are still far bigger.
This was done in 1997 when TCI launched their Headend in the Sky (HITS) to distribute digital TV programming and national control of digital set-top boxes (STB).
But there still is no equivalent to this today where a cable operator can get all of the needed content through the Internet cloud and it still requires satellite dishes to get content out of the sky.
I know this is an old thread, but I was wondering about what this looks like 5 years later. Do virtual MVPDs (particularly ones not owned by the big cable players) still require a physical headend? Essentially the same architecture, just swapping out coax for delivery over broadband?
Anybody that pulls signal from satellites still needs a headend. All of the big cable companies operate a number of them. The headend accepts satellite signals and then translates it to whatever format is needed to transmit over customer cables. Since the cable product has little or no profit in it, we don’t see smaller ISPs building new headends, but buying signal from existing ones.