In an era in which social discussion about TV is readily accessible through a quick hashtag search, much of the information showing connections between TV and Twitter or Facebook may seem less-than-scientific. New data from the Council for Research Excellence measuring the effects of promotions, social-media posts and word-of-mouth is anything but. And some of the results might seem a little counter-intuitive, especially to social-media boosters.
Among the results: Based on self-reporting by those studied, Offline communications and Facebook had more influence on likelihood to watch something when participants were not watching TV. But, as many multi-screen addicts know, Twitter and text messaging had a bigger impact while study respondents were watching TV. "When viewing, one is already engaged and this leads to both higher likelihood of viewing and communications," notes the study, which has not been released publicly.
The organization, comprising Nielsen's media clients, oversaw thousands of hours of academic data analysis work to better understand how new social platforms affect what we watch on the tube.
According to CRE, which paid for the study, it was "one of the highest-budgeted studies" done under the organization's auspices. CRE and Nielsen declined to provide the exact cost of the study.
For its second social TV study, deemed "Talking Social TV 2," CRE measured the average additional likelihood someone would view a show after encountering one additional message about the show in the form of offline word-of-mouth, promotions, social media or a digital contact. It also measured the increase in audience if everyone in the audience received one more encounter with a particular medium.
Results show that among repeaters, or people who watch the same show regularly, offline word-of-mouth drove the highest lift -- around 2% -- compared to around 1% lift from social media and even less from a one-to-one digital communication. Among infrequent viewers, offline word-of-mouth generated the most lift among infrequent viewers, but only a 1% gain. Promotions, digital and social media communications each resulted in under 0.5% gain.
The report also looks at the potential effect of such communications on various types of network and cable programs. For example, it predicted shows including CBS's "How I Met Your Mother" and NBC's "The Voice" each would benefit from reach campaigns in terms of producing higher ratings, more so than frequency and mixed campaigns. Reach-focused word-of-mouth, according to researchers' models, would raise ratings as much as 50% for "How I Met Your Mother" and nearly 45% for "The Voice."
"These numbers look big because they are," said Mitchell Lovett, associate professor of marketing
at the University of Rochester, Simon Business School, one of the head researchers for the study. He explained that a reach campaign, in this case, refers to a simulation in which all people who did not previously receive a particular type of communication about a TV show would receive that type of messaging.
CRE evaluated TV viewing diary entries of 1,665 respondents ages15-54 submitted via a custom-made mobile app over a three week period. With help from Nielsen, CRE compared that self-reported data to actual viewing of around 150 people from that group based on their Nielsen People Meter reports. The result was more than 78,000 diary entries about nearly 1,600 TV shows -- a substantial database of viewing information.
"There's a ton of data that comes in through the app," said Beth Rockwood, SVP market resources and ad sales research at Discovery and chair of CRE's Social Media Committee, who helped lead the study. "It's definitely a big data project." Added Ms. Rockwood, "It was the first time in my knowledge that we have used a mobile diary to capture behavior over an extended period of time, and then also to compare the viewing behavior to the Nielsen People Meter data…. This type of diary is not done often since it is difficult to properly manage and it's also expensive."
Nielsen Life360, the company's digital ethnography division, built the mobile app that respondents used to create mobile viewing diaries, said Ms. Rockwood. Of those, around 150 also belonged to the Nielsen Peoplemeter panel, providing the researchers a way to gauge validity of self-reported diary accounts that sometimes are biased or include incorrect or incomplete information.
Normally Nielsen does not allow its Peoplemeter participants to be involved in additional studies, but in this case, said Ms. Rockwood, the small group involved with the social TV study were leaving the Peoplemeter panel. One goal of the study was to distinguish between how Facebook influences TV viewing vs. how Twitter does. "We wanted to try to decode that a little bit," said Ms. Rockwood.
Along with CRE staff, the organization relied on assistance from academic researchers at University of Rochester and Hebrew University in Israel, according to Ms. Rockwood, who said her academic counterparts spent about 2,000 hours on the project.