Troubleshooting VOD’s 'Failures’


Video on demand is supposed to be one of cable’s best offerings for reducing churn and preventing subscribers from switching to satellite.

And judging from the anecdotal evidence delivered at industry forums by Comcast Corp., it’s appearing to work. Digital churn rates in systems with VOD can drop anywhere from 10% to 30%. And that’s good news.

But what happens when the opposite occurs? What happens when VOD is introduced and the experience is so poor that people actually leave the service?

It’s not something you’ll get MSOs to talk much about, and the numbers may actually be small. But there is some evidence that everyone isn’t always getting what they expect.

Everstream Inc. CEO Stephen McHale said his company has built a data-collection reporting system that MSOs use to monitor VOD usage, as well as the quality of the consumer experience.

Everstream collects information from three different perspectives, said McHale — the subscriber, the content and the quality of the application. As such, Everstream interfaces with billing systems, servers, transport systems, quadrature amplitude modulation devices and headends.

It allows MSOs to troubleshoot engineering and trafficking problems, and monitor VOD usage activity across the movies-on-demand, SVOD and free-on-demand platforms.

McHale recently took a snapshot of activity across a representative sample of its 27 market deployments. (Its MSO client list includes Time Warner Cable and Adelphia Communications Corp.)

McHale looked at four areas:

  • Subscriber activity, defined as the percentage of VOD subscribers that were active in a given period of time, typically 30 days;
  • Take rate, defined as the average number of times a subscriber ordered an on-demand movie or an SVOD selection in a month;
  • Average revenue per subscriber, per month for movies-on-demand only; and
  • The dormancy rate, defined as the percentage of subscribers who had been active VOD users, but had done nothing for 30 days.

McHale then did an overlay of those stats using a three-tiered quality metric — accounting for highly stable, somewhat stable and unstable markets.

A highly stable market was defined as producing an error rate of less than 1%. Somewhat stable markets had error rates below 10%, while in unstable markets — and yes, there were some — error rates exceeded 10%.

Monthly subscriber activity in highly stable markets was 57%, but that quickly fell to 15% and 13.6% in somewhat and unstable markets, respectively.

Take rates went from 2.99 accesses per month in highly stable markets to 1.58 and 0.77, respectively, in the less-stable ones.

Average revenue for movies on demand slid from $1.94 in the highly stable markets to $1.19 down to 89 cents in unstable markets. And the dormancy rate in highly stable markets was 17.6%, compared to 33.5% in somewhat stable systems and 40.9% in unstable ones.

The dormancy rate is important, said McHale, because operators are finding that to be a key indicator for digital churn. The less often subscribers use the service, he said, the more likely they will churn.

It’s clear from the numbers that quality-of-service is huge to consumers. If it doesn’t work, they won’t use it.

Perhaps what’s most alarming is that it doesn’t take much failure for cable operators to fail. Revenue drops sharply when failure rates reach the single digits.

The industry bandies about the five nines — i.e. 99.999% reliability — for voice service, and it’s legitimate. But cable may have underestimated the quality levels necessary for advanced video products.

Consumers may have become used to intermittent cell-phone service, but it seems that TV is different.

When consumers changed channels, it worked every time.

Now that a much-wider swath of America has VOD available to it, and millions upon millions of viewers are trying it, cable needs to get it right the first time — every time.