Long Tails and Media Attention

Lex Miron of CIBC has produced an extremely interesting, analysis of the role of the Long Tail in today’s interactive media universe (Part 1 & Part 2). What he was trying to determine whether investment value is greater in the most popular sites or in the mass of less popular sites which he defines as the long tail.

Using Nielsen/NetRatings data to deduce time spent on each site he ranks them developing a power law function.  To determine value he uses time on a site as an analog for advertising value – and uses this to determine the overall value of the long tail vs the top 100. The study covers the period between Jan 06 and 07. 

The results are quite predictable, the top ten sites accounted for 76% of all the time on media sites while the top 100 sites accounted for 91% of the time. What is also interesting is that except for YouTube the top ten sites were the same at the beginning of the study period and at the end. During the study period the number of sites Neilson tracked grew by18% while and the overall time on these sites grew by 16% – with most of the growth in time on site occurring in the tail.

Some of the conclusions have more to do with Nielsen’s data set than the tail its self (see Chris Andersons comments the study here) – though the general conclusion is that sites in the tail have a less tenuous hold on their audience that those in the head as can be expected (more about this later). That said the tail is where the best investment opportunities lie.

What’s frustrating about the study is that it leaves so much unknown that would be useful both from an investing perspective – and to understand how the top sites are so sticky.

For instance is the time on top sites driven by high profile content or diversity and breadth of content? As well how much is time is driven by social media features like commenting, sharing and mash-ups and how much is content consumption alone.

In this regard it will be interesting to see how features such asYouTube’s TestTube  (Active Sharing, Audio Swap and Streams 2.0) affect time on site as all require users to commit significant amounts of time to gain the value from these services.

Two little quibbles. 

First the concept of Long Tail has to do with the value in the diversity of content not necessarily the ranking outlets for it. Much of what troubles media is a result of the drop in attention they get because people seek diversity elsewhere – not because there are more media outlets (ie it is the interest in diverse content that makes the sites viable). 

Second is that in defining the long tail in terms of time people spend on specific media outlets, instead of focusing on the diversity of content and the role of content vs interaction in that we miss an opportunity of understand the factors that contribute to high profile sites.

That said this is an important first step in understanding the opportunity and value available in interactive media.