TV EverywhereTraditional and basic metrics that are losing their power to describe financial performance or end user behavior are among the sure signs a market is changing. So it is that the television industry is within three months of creating new definitions of what a “TV household” is, for purposes of measuring viewership, and hence developing ratings that directly affect the value of advertising on TV shows.

A committee at Nielsen started work in 2012 to figure out what a TV viewing household is, these days, in part because the number of such households is falling.

The number of TV households fell from 115.9 million in the 2010 to 2011 TV season, to 114.7 million the following season. That was the first drop of TV households in 20 years.

An additional 500,000 households disappeared before the 2012 to 2013 season, leaving a total of 114.2 million television homes. That worries many in the ad-supported TV ecosystem because such declines automatically mean fewer people consuming a product that creates the basis for advertising revenue.

But advertisers also care because attention is shifting, and advertisers want to be where the attention is being directed. In a broad sense, that means measuring the video people actually are watching, no matter which device is used, even if such consumers do not in fact own TVs or watch over the air TV or use a subscription service.

Among the proposals being studied is a change of definition to reflect viewing of Internet-delivered video to a tablet as reflecting a TV household, even if that household owns no televisions. The other issue is whether a household where Internet-delivered video is viewed also constitutes a TV household.

The drop in TV households probably has a number of causes, though. There do seem to be more homes that simply do not buy television sets or watch TV.

The tougher economy has sliced the number of new home starts, and most new homes do still become additional TV households. Also for economic reasons, fewer younger people are starting their own households.

Additionally, a growing number of people choose to watch video only on their smart phones or tablets or PCs.

Some might view the proposed changes with a bit of skepticism. The wide embrace of multitasking, for example, suggests that even when TV household members have the television on, it is not being “watched” in the same way. People are interacting with their smart phones, tablets and PCs while “watching.”

At another level, one might say that when an industry has to change fundamental “definitions” to measure its financial and operational success, the business itself is changing. Recent history suggests that such definitional changes only reflect changing revenue drivers.

There was a time when it made sense to count and measure “access lines” and “revenue per access line,” as it once made sense to count “basic TV subscribers” and “basic TV revenue per user” as the fundamental measures of enterprise success.

That began to change in the mid-1990s, and became a serious problem after 2000, when telco revenue streams began to reflect serious erosion of voice lines and the new impact of high speed access and television subscriptions.

The cable industry encountered the same problem as it began selling voice subscriptions and broadband access.

And both industries moved to multiple-service packages that made “revenue generating units” a more meaningful metric than “lines” of any sort.

Technology also played a role. When alternate access providers and then competitive local exchange carriers started selling bandwidth services, the notion of a line became an imprecise measure. Instead, some service providers began reporting new and different statistics, such as “access line equivalents.”

So Nielsen’s new definitions strongly suggest the TV business is changing, in ways that fundamentally will change its revenue trends and prospects. As always, those changes will have a differential impact on ecosystem participants. Some will win; some will lose.