[opendtv] Re: 20040712 Mark's Monday Memo

  • From: "David Keeble" <david.keeble@xxxxxxxxxx>
  • To: <opendtv@xxxxxxxxxxxxx>
  • Date: Wed, 14 Jul 2004 08:55:37 -0400

Re: the discussion about how many households continue to rely on OTA.

I've made frequent attempts to track this number here in Canada, using
similar data, and have found it to be be difficult for several reasons:

1. The numbers reported often fail to distinguish carefully between
"households", "subscribers" and "customers", and

2. It's impossible to get anything accurate on the "Stealing" phenomenon

3. Who knows what the cabsat overlap actually is?

#1 is the worst problem, and I suspect it happens becuase the people
reporting statistics are the cable and satellite companies, who often have
an agenda they are supporting. Meanwhile, there is no interest group arguing
for OTA.

In terms of penetration, households is the only useful unit of measurement.
But the reports of various parties often include other customers -
hospitals, hotels, restaurants, bars, schools, and other institutions.
There are a lot of these non-household subscribers, and if they are
included, and the total compared against national households, it exaggerates
cabsat penetration. In using cable stats, I've always had to go to the
source to separate the institutional subs, and to make sure I'm not using
"percentage of homes passed", which can be a very misleading figure.

I note that the table quoted (Chart 10 in
http://www.ncta.com/pdf_files/Overview.pdf) uses the term "customers". That
tells you right away that it can't be compared to households.

Chart 11, evidently based on an FCC report, uses the term households but
provides no figures, only a graph. I'm not even sure this is correct - I'm
no expert on the FCC, but I note that Bruce Leichtman reports that the FCC
report cited in Chart 11  does not distinguish between households and
subscribers. (See: his article at
http://www.satelliteguys.us/showthread.php?t=6373. For a general discussion
of these stats, see his previous article at

Also, two subscriptions in one household may be counted as two customers
(one subscriber) and can look like two households.

The overlap between cable and satellite is real and is often estimated here
at 2% of total HH.

When I do my best to make a real apples-to-apples comparison, I am usually
surprised to discover how many households still rely on OTA. Cabsat
penetration has often been exaggerated by as much as 10%.

I also find that when you use viewing stats to measure actual usage - i.e. %
of viewing hours received on an OTA antenna, it's remarkable how much
viewing an individual station or OTA network is still receiving from
antennas. This viewing combines  "uncommected" households, second sets in
cabsast HH that are using antennas, or switching to antenna to get a local
station that isn't on satellite.

None of this is to say that the OTA audience has not declined substantially
over the last ten years - it clearly has - it's just a caution about the use
of the available stats to calculate the number of OTA HH by subtracting the
cabsat numbers from total HH.

David Keeble

You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts: