to get a sense of why i think this could be the case let’s first look at how nielsen media research currently operates (as james surowiecki explains in his book, the wisdom of crowds) and then at how hey! nielsen is making their job easier, better, and cheaper for them.

most people with even a passing interest in televised media and ratings knows about nielsen media research and their tv ratings. what most people don’t know is that these ratings are actually calculated based on the habits of an insignificantly small number of people. for example, based on the last count, nielsen’s ratings are formulated based on 5,000 american homes, where they have wired the television sets with electronic monitoring devices called ‘people meters’. whenever someone from these families wants to watch tv, he or she has to punch in a unique code to make sure that nielsen knows who is watching what show and at what time.

since the company uses only 5,000 families for data that has consequences for over 300 million people across the country, nielsen has to work hard to ensure that the 5,000 families are a good average demographic representation of the whole country. even when the company manages to do that, it is hard to make sure that every family logs every show it watches, and the data that gets across inevitably gets sloppy and somewhat less reliable over time.

though ‘people meter families’ provide somewhat sound data, the major networks have adopted a tradition of ‘sweeps’ that makes matters much worse. 4 times a year, nielsen will send paper diaries to 2.5 million americans and ask them to record what programming they watch, for a week, and then mail the diary back to the company. this creates several problems:

1. sweeps ratings rely heavily on people recording data correctly. unlike ‘people meters’ there is no way to ensure that the data is accurate.
2. the sweeps system has an incredibly low response rate (of about 30%) which creates a ‘cooperator bias’ (i.e. people who respond are likely to watch all the same shows and people who don’t respond all watch the same shows which don’t get recorded).
3. people are lazy, especially if they have nothing to gain. even for most people that respond to the surveys, the are a chore, which means filling them out is probably not done systematically (as people watch the shows) rather done all at one time (possibly after the week of watching tv is up), causing even more inaccuracies in reporting.

all this data is important because this advertisers are interested in who is watching what, and when, because they want to reach the most targeted audience. it’s in everyone’s interest to have more reliable data, but the problem is that to get a substantial installation of ‘people meters’ would require an investment somewhere in 9-figures. enter hey! nielsen.

hey! nielsen: now with more ‘people meters’

hey! nielsen is nielsen media research’s foray into social media. for an overview, have a look at my review of the site. i said back then, ‘with hey! nielsen, nielsen media research has hit the social media nail on its head,’ and this rings even truer in this context.

we know you love tv, movies, music, the web, and the personalities behind it all, just a little too much. so here’s your chance to let your voice be heard, your face been seen, and your opinions be a force for change. you, plus the exclusive access and power of nielsen means that your opinions not only reach millions of people via the web, but may also reach the media moguls who decide what goes on the air and on the web.

with that said, and requiring only a strong opinion and an email address, everyone can become a ‘people meter’ with almost zero marginal cost for nielsen media research and no barriers to entry for the user. while you can read other people’s opinions without registering, you have to register to have your say. and when you do register, they record your location (zip codes are mandatory for u.s. and canada), gender, and age right away. furthermore, and so there is no doubt, let’s make sure that they do plan to use this site to aggregate opinions and information for marketing research analysis.

from their privacy policy:

we also automatically collect certain non-personally identifiable information including ip address, browser type, requested and referral urls, time spent on certain portions of the site and other forms of site usage data from members and visitors to the site…we may use this information along with the other information we collect for general market research purposes, including preparing reports and analysis on the opinions, behavior and characteristics of the members and users of the hey! nielsen site.

as a leading provider of information and services for the media and entertainment industries, one of the goals of the nielsen company is to help others understand consumer behavior and opinions across all media and entertainment platforms, including the internet. one way we do this is by conducting market research and analysis on the opinions, interests, demographics and behavior of the users and members of the hey! nielsen site. we use this information, along with information we may obtain from other sources (such as other market research firms), to prepare various market research reports and analyses for our clients and prospective clients relating to the various content and other information made available in connection with the site.

there are several reasons why i think this is a great move for nielsen and one that shows that at least one media company knows what’s going on, on the internet.

1. it’s dirt cheap: brainstorming, creating, maintaining, and marketing the site, and aggregating the data from the site is far more cost-effective and efficient than attempting to install ‘people meters’ by the same volume or continuing with sweeps month.
2. there’s much more information: users have the option to read, react, and respond to information the other users post and can rate events, shows, and personalities. not only is the degree of interactivity much higher, but even if a visitor doesn’t interact with a page, just visiting a page is measures a degree of interest.
3. it’s completely hands-off: people-powered content aggregation and content rankings ensure that the ecosystem keeps working without outside intervention (unlike the sweeps). furthermore, data is automatically aggregated and sorted so that the community does all the work and nielsen can reap the benefits of the users’ interactions without much work and without having to push the users.
4. it projects the future: profiles and calendars help nielsen keep track of what people are interested in currently, but they also show what what upcoming events people are looking forward to.

overall, hey! nielsen is not just cheaper, but it is more efficient, more accurate, and accumulates much more information than is possible using existing means.

this post is a part of my journey through james surowiecki’s the wisdom of crowds.

Technorati Tags: nielsen media research, nielsen, hey! nielsen, entertainment, television, ratings, rankings, advertising, james surowiecki, the wisdom of crowds