Social media is giving television viewers a forum and a voice to debate their favorite shows — but how do networks and advertisers harness the buzz and put a value on it? Wharton operations and information management professor Shawndra Hill is studying how “social TV” could be used to effectively build viewer engagement. By understanding who is using social media to talk about TV, and why, she says, companies can provide consumers with better recommendations for other content they might like, and also figure out how to better market and sell their products in a rapidly changing media landscape. Hill’s research can be found at the Social TV Lab.

An edited version of the transcript appears below.

Knowledge at Wharton: Shawndra, thanks so much for being with us today. We’re here to talk a little bit about your research into the growing use of social media to market shows and products being shown on television. Could you tell us a little bit about what “social TV” is and what got you interested in researching it?

Shawndra Hill: Broadly speaking, social TV is the integration of social media and TV programming. This has happened in a variety of ways: For one, people contribute somewhat organically on social networking sites like Twitter and Facebook. More recently, there are a number of [successful] social TV applications that have been developed by [several] businesses to allow people to basically show how big a fan they are of different TV shows. [Users are able to] do things like “check in” to shows and communicate with one another. Social TV applications have even been developed by the networks themselves. So networks in the U.S., at least, have ways for their viewers to interact with one another on the networks’ websites and in fact are trying to drive them to their own websites to do just that.

Knowledge at Wharton: What do you think is driving the interest on the part of television networks and advertisers to want people to talk about their shows or products using social media?

Hill: People like to talk about television. Both networks and advertisers — only recently, I would say in the past three and a half years — have noticed this. So people are already doing this on a large scale. And whenever advertisers notice that people are doing something on a large scale they get very interested. Then it begs the question, how can [companies] use the fact that people are talking about them on social media to better engage with viewers and customers or potential customers, and then also hopefully make the viewer experience a little bit more meaningful? The networks are more interested in making the viewer experience more meaningful so that it will in turn drive more viewers. But advertisers are really interested in understanding what this interaction means for them with respect to selling their products.

Knowledge at Wharton: You’ve done a number of different studies looking at various aspects of social TV. Could you talk a little bit about some of those studies, what you have looked at and what analysis you have done?

Hill: We have run a number of studies now, including some that are in the preliminary phase. But the two that I think that are most notable are one, where we studied the popular reality singing show The Voice, and the other is the Super Bowl. On the one hand, we’re interested in following a TV show that has used social media as part of its programming. And we also want to better understand what that means with respect to viewer engagement. On the other hand, we followed TV advertisers [and how they used social media] during major national events, including the Super Bowl.

So let’s take The Voice. The Voice, as part of its programming, does a number of things to engage viewers through social media. One is to simply put a hashtag on [the bottom of the screen] during the show. So #thevoice is used as an identifier for the topic The Voice [on Twitter]. Additionally, they run specific tweets on the screen that are generated both by users and by people who appear on the show. Additionally, they have what’s called a “social media room.” I don’t know of any other shows that were doing this, at least when they did it for the first time. In the social media room, they have an emcee who basically interacts with both the contestants on the show and the viewers in the audience.

By using all of these different strategies, The Voice has been able to generate quite a bit of word of mouth on social media. And so we have tried to understand what has triggered the most word of mouth, what seems to be working for the show and what seems to be less effective. We found a number of things. One, and I think most importantly, that actually providing these triggers via either Twitter hashtags or specific tweets leads to more engagement. And also, we find that the content, in terms of diversity, changes when these triggers are posted on the show versus when they are not. We also find that there is a different response to the social media triggers when the show is aired for the second time — for example, when it is aired on the West Coast after it has already been aired on the East Coast.

Knowledge at Wharton: So when it’s “live” versus not live?

Hill: Exactly.

Knowledge at Wharton: You also mentioned a study about the Super Bowl. What did that entail?

Hill: We knew in advance who was going to be advertising on the Super Bowl. And so we followed those advertisers on Twitter and on Facebook and were able to see both what they were contributing during the Super Bowl and also how viewers were talking about the brands during the Super Bowl.

The remarkable thing is that we could see in real time how the users were responding to specific triggers. And so in this case, it was both content that is happening during the show where firms are advertising, but also some of the advertisers were doing things like playing social media games [during the entire broadcast.] We found that for the advertisers that actually included some kind of social media content, but specifically the games in the social media content, their buzz, if you will, or the number of tweets that people were contributing online, were not only greater in number, but also more sustained over the Super Bowl, which suggests that actually incorporating some kind of social media into your content actually leads to more engagement and also to more sustained engagement. And if you believe that that’s important, which I think most advertisers do, then they can think about how to incorporate more social media content into their advertisements.

Knowledge at Wharton: Can you give an example of a “social media game”?

Hill: The types of games varied, but one example actually happened before the Super Bowl. Chevy had a competition where the company asked viewers to contribute their own commercials. One of the commercials shown during the Super Bowl was one contributed by a user. So they had people submitting these commercials and as a result, people were interested in finding out who won.

Another example is something that Coca-Cola did. They had [the company’s signature] polar bears responding to the Super Bowl in real time. Based on what was happening during the Super Bowl, the commercials were [changing] and users were contributing content as a result.

Knowledge at Wharton: One thing that I have noticed about The Voice is that it’s not just viewers who are tweeting. They’ve also got the judges tweeting and the contestants are tweeting. Would you say that is unique? It seems like the show has really made a very concerted effort in terms of keeping the buzz going off-screen on social media.

Hill: Absolutely and it’s really important. They’re using a number of strategies, as you just mentioned, to keep people engaged. And the reason why we think it’s really important is because what we found in our data is that it’s almost [a case of] out of sight, out of mind. When you see a particular contestant, buzz increases. And when that contestant is no longer singing, the buzz goes away. While the show is on, there’s a lot of buzz. When the show is off, not so much.

And so it’s really important for The Voice and shows like it — or TV shows in general — to continue to engage with their viewers if they want their viewers to continue to engage with the show when the show is not on. Because what the shows need to do is remind the viewers that there is new content [available] or that the show is coming on in a few days.

Knowledge at Wharton: You had mentioned earlier that you and your research team have gone into pretty exacting detail while studying these shows — not just watching them once, but several times. Can you tell us more about some of the methods you have used?

Hill: In order to do the research that we’re doing, we are basically pulling data from a number of sources. We are looking at Twitter and Facebook response, obviously, because we’ve talked a lot about that. But in addition, we’re actually recording the shows. And in order to later look at response to specific triggers, we actually need to time-stamp those triggers. We go back through the shows and note exactly when certain events happened so that we can see how that directly ties to the social media response.

What’s nice about The Voice [is that] in addition to having the Twitter and Facebook data and the content of the show that is tagged, we also have sales data: Part of The Voice — how people vote, in part, is by buying songs [performed by the contestants] on iTunes. We were able to look at the sales rank of the iTunes songs associated with The Voice over time as the season was going on. Because we were able to collect that data, we have been able to link the buzz to sales.

There are huge questions out there — not just in relationship to TV, but for social media in general — around whether buzz actually leads to sales or in the context of TV, does it really lead to more viewers. It’s been really nice to have all of this data because we’ve been able to link buzz directly to sales, at least in the case of The Voice.

Knowledge at Wharton: Now in terms of the research, was there anything particularly surprising or something that really caught your attention while you were collecting or analyzing the data?

Hill: In general, it’s surprising just how much people are responding to TV, but also to TV triggers specifically. It suggests, at least right now — things can change over time as users and viewers become used to seeing these TV triggers — that the TV triggers actually do lead to more engagement, whether it be something that’s happening in a TV commercial during the Super Bowl or The Voice putting these hashtags in specific tweets on the screen…. And I think we went in thinking that that would happen — otherwise we wouldn’t have started the research — but it’s surprising just how much that seems to be true.

Knowledge at Wharton: There’s been a lot of talk out there in a number of industries about the importance of putting a value on social media — that is, if you have a million followers, great, but what does that mean for a company’s sales, or its marketing strategy? What do you think is the value of this research in terms of those questions, at least in this particular space?

Hill: Everybody’s asking that question, right? And I won’t say that we have all the answers, but for the first time we have been able to link buzz to sales. So that’s promising. And I hope that as media becomes more integrated and as advertisers begin to launch campaigns where they’re looking at advertising across different screens during television events, that they will be able to better measure the impact of social media buzz. But I’d like to believe that at the very least, linking buzz to sales for The Voice is a first step toward better understanding the relationship between buzz and sales.

Knowledge at Wharton: Where do you see the research going from here? What are your goals for the future in terms of taking this to the next step?

Hill: We have a lot of goals. In addition to the specific case studies that we’re working on, we’re actually working on a number of projects that look at social TV in general. For example, we’re building a social TV recommendation engine that takes advantage of the fact that people are contributing this content about shows. And we use it to calculate the similarity between shows, to make better recommendations about TV shows. In addition, we’re studying the relationship between how popular a show is using Nielsen ratings … and how social a TV show is. Other data that we’ve been able to get are check-ins on some of these social TV apps that viewers are making in public on Twitter, for example….

In the future, what is most exciting is one, doing what you’re talking about — coming up with some better analytics for social TV and social media in general to begin to link that to some business value. And then I think for me, what’s really nice to see is that people are now using social TV for good. One example of that is there was this telethon … called Stand Up For Cancer…. During the broadcast they were, of course, trying to get people to donate for cancer research. But the celebrities on the program were interacting with viewers of the show. In addition, as people were donating, they were talking about that on social media and also saying that they stood up for a specific person.

It was one example where a television show was actually trying to, not even incentivize, but motivate people to talk about this particular event and donate to a special cause on social media. There are other examples like that. PBS just recently had a broadcast….

Knowledge at Wharton: Is that Half the Sky?

Hill: Yes, the show was called Half the Sky, and it talked about oppression of women. And they partnered with one of these social TV apps that I’ve been talking about, GetGlue. And along with GetGlue, they gave the Get Glue app users special badges for watching both episodes. They partnered with a social TV app to get people to view and talk about the show. And I think [these are] nice examples where not only is there the advertising story, where people are trying to come up with analytics to better understand what this means for products and brands, but also show how important it is to think about the fact that regardless of if you develop content to promote word of mouth on social media or not, people are going to be talking about the shows — and so why not? Especially in the case of these good causes, try to think about the best ways to promote that word of mouth so that you build more awareness, even when people are not watching the television show.

We’ve been talking with a couple of TV networks — and not networks in this country, by the way — about running some experiments. As we begin to better understand what seems to be driving sustained word of mouth and more engagement using these TV triggers, we want to begin to experiment with different messages to see whether we can actually figure out exactly what is causing the response.

Knowledge at Wharton: How can your findings be used by marketers or network executives in developing a business strategy?

Hill: I think from the network perspective, they have to be thinking about a number of things. Networks make money in part from advertisers, so they want to have a lot of viewers of their shows. It’s kind of an obvious thing to think about how incorporating social media content will not only lead to buzz, but also to more viewers. Hopefully, the results of our work will better inform networks that are trying to incorporate social TV content.

Additionally, advertising, [which is a] huge part of how networks make money, is changing. Consumer behavior is changing, and if people are glued to their computer or their phone or even their tablet when they are watching television, it kind of changes how they’re interacting potentially with advertisements. I think networks — and they know this — are going to have to pay attention to that fact and, as a result, come up with new advertising models that take advantage of it. And that’s not a bad thing, because potentially what will happen is because people are contributing content and thus revealing more information about themselves and their preferences, you might be able to do better targeted advertising to consumers and therefore, more effective advertising to consumers that are watching TV.

Knowledge at Wharton: So networks could be potentially find ways to turn the fact that people’s attention is divided and that they might be able to fast forward through the ads from a negative into a positive?

Hill: Right. And then, even better, the Holy Grail of course is to drive people to buy things, right? So, at least today, you can’t buy anything by watching television alone. But if you’re online or on your phone, you can click something to purchase or be driven directly to a website or something like that.

Knowledge at Wharton: Finally, what are the implications here for consumers?

Hill: First of all, I think networks care about viewer engagement. And so on the one hand, you’re seeing networks respond to the fact that people are contributing in public for free by creating new ways to enhance the viewer experience. I think viewers will benefit because there will be new applications and new ways to engage with TV shows and even [with] celebrities. What’s nice, too, is that we can begin to use the content that viewers are contributing to help them learn about new content. In some way, we hope to give back to viewers some insights into how TV shows are related and what they may like. And we can only do that because they’re contributing this content.

I think users and viewers stand to benefit tremendously from the insights that the business analytics will give us. Perhaps they’re not as happy about the advertising angle, but that’s positive, too, perhaps because they will get more targeted advertising. And so the advertising at least — if you’re going to have to watch it anyway — will be more relevant.