Last week's Social Media Insider Summit in Lake Tahoe saw a series of great presentations addressing some of the cutting-edge issues in social media marketing -- including of course measurement and analytics, always a favorite topic for media types and especially important for making social media transparent and accountable, which will in turn allow it to attract more ad dollars. But one presentation, by Nielsen research manager for measurement science Nina Stratt Lerner, was interesting for addressing social media as a source of information about consumer sentiment -- specifically buzz about TV shows, and its relation to actual program ratings over time, which should allow TV broadcasters (and advertisers) to make predictions about the popularity of shows.
The Nielsen case study tracked the volume of online buzz over the course of the TV season for four shows: "Jersey Shore," "The Bachelor," "Modern Family" and "Gossip Girl." As one might expect, for most of the shows there is a preliminary spike around the premiere, followed by slowly (and unevenly) mounting buzz in mid-season, and then a big spike around the finale. Where is all this buzz occurring? Interestingly, Nielsen found that discussion boards and blogs were far and away the biggest sources of buzz about TV shows, compared to Facebook (including Groups). In fact, the volume of buzz on Facebook including Groups basically appeared to flatline across the TV season, while blogs and discussion boards showed increasing buzz along the lines described above.
Nielsen also found that online buzz tends to correlate with actual ratings more strongly among younger demos than older demos: for the 50+ cohort, the correlation is low across the board (males, females, and all persons), while buzz in the 35-49 set is somewhat correlated, and the highest correlation between buzz and actual ratings is seen among 12-17-year-olds. I'm guessing that the overall volume of buzz is also higher among the younger demos (although I suppose there could be millions of 50-something women breathlessly posting about Gossip Girl).
Lerner's presentation also included an interesting graph showing correlation between buzz and actual TV ratings for broadcast versus cable TV. For some reason (I'm curious why this might be) the correlation between buzz and ratings is quite a bit higher for broadcast TV than it is for cable. Is it simply that broadcast TV shows tend to have a wider distribution and larger audience, including a larger -- and therefore more representative -- crowd of social media users? I'd be interested to hear any thoughts readers have on this.
(As I write this I see that Nielsen just launched a new social media measurement and analytics service, BuzzMetrics Exchange, which the company is billing as "a full service social media monitoring and engagement platform," designed to allow clients to directly engage with consumers who are talking about their brands online.)
Correction and Clarification: This column originally attributed the research to Nielsen, but in fact it was performed by NM Incite, a joint venture of Nielsen and McKinsey. Additionally, although only four TV shows were discussed in the presentation, the full study actually covered 250 shows. Finally, it should be noted that the portion of the study addressing Facebook buzz versus blogs and discussion forums only looked at public-facing Facebook fan pages, which may explain the lower volume found there.