>My best theory is that there are no accurate metrics on >audience or conversion with, say, TV advertising so >advertisers are basically buying into the lie tha

>My best theory is that there are no accurate metrics on >audience or conversion with, say, TV advertising so >advertisers are basically buying into the lie that networks >sell them. There also are no accurate metrics on audience or conversion with TV advertising. It is just that everyone in the industry has grown accustomed to using the Neilsen numbers, no matter how inaccurate, as "the one and only", and they don't yet have any comfort with the equivalent on a net based distribution system. Consider the typical Neilsen ratings number. It estimates (because it is a sample set) that X number televisions were tuned to broadcast Y at time Z. And from that advertisers estimate that Q number of viewers watched the ads. But in reality, all the Neilsen numbers report is that a certain number of Neilsen monitored TV's are tuned to broadcast Y at time Z. During the ads, they do not know what happened. The viewer might have hit the can (a common occurrence), or gone to get some munchies from the kitchen (another common occurrence) or checked HN while the ads were running. But because all the "users" of the numbers have grown accustomed to estimating from an estimate they believe they have something useful. In reality, the emperor has no clothes. But they ignore reality.