Social Networks and Word of Mouth (WoM) – a perfect marriage

Some history

A network is defined as a set of nodes and links. In the late 18th century Euler invented the field of network theory. For the following century and a half it remained in the realm of the abstract in mathematics departments. In the second half of the 20th century social scientist started studying social connectivity. Milgram’s famous six-degrees-of-separation experiment revealed how highly interconnected humans are. You can reach anyone else in the world following on average six links. Nevertheless, this average hides the fact that some people are “social hubs” and can link to anyone else in as few as 3 links and others are highly disconnected and need more than 100 links to reach anyone far away from their cluster. In the 90’s scientists started looking at digital networks. First they turned into the world wide web which is a highly tractable network where websites are the nodes and their links are the network’s links (a fact that Google uses in its search algorithm). More recently scientist turned their attention to digital social networks where people are the nodes and the friend connections are the links, Facebook being the largest social network today. In parallel, scientists have also studied the evolution of epidemics and spread of contagious diseases.

The connection to WoM

All this lines of research coupled with the internet revolution give rise to the opportunity to start studying Word of Mouth marketing in social networks. Network theory plus epidemics theory applied to WoM can provide huge insights on how messages (promotions, brands, product reviews, etc) spread in a social network (in the same way a contagious disease spreads). The first thing researches need is a map of the network (Facebook has it). Second, they need a way to track messages to analyze the “contagion effect”. Finally, researchers should follow the “touched” users to understand how they act on the message (e.g. do they talk about the product?, do they visit the brand’s website?, do they buy the product?) With this information marketers could leverage WoM much more effectively. In fact, current WoM services are based on artificial buzz generation (e.g. hired agents dropping messages) and sign-in communities that participate in WoM initiatives (e.g. SheSpeaks). The problem with the former is that the buzz is artificial and does not resemble the organic spread of messages. The problem with the latter is that it more a pool of people willing to participate in marketing activities and in that sense are more like a digital focus group instead of a real incubator of worth of mouth.

The opportunity

Given that the tools exist (network theory, social interaction theory, epidemics theory, and complex systems theory) and the current services for WoM do not capitalize on the full potential, there is an opportunity to study WoM on social networks like facebook and start offering a better service. For example, if you know which users are the “social hubs” for different products/services, so companies can know where to drop the message and what message to drop where.

"What's happening" in the cabinet of Japan

What do President Obama and Japan's Prime Minister Yukio Hatoyama have in common? They are both powerful national leaders who "tweet".

Hatoyama is a unique figure in Japan's politial history. First, he was elected Prime Minister of Japan last August leading Democratic Party of Japan (DPJ), whose victory ended more than 50 years of nearly unbroken rule by the conservative Liberal Democratic Party (LDP). Second, Hatoyama is known as "the alien" from the outer space because of his unreadable character and unprecedented engineering background with a PhD from Stanford .

The news that Hatoyama joined Twitter to talk about his life as Prime Minister was a sensation in Japan, especially among the young generation, because during the old rule of LDP, politics was always done secretly whether it is in Nagata-cho or in first-class Japanese restaurants near Akasaka. Then people thought, "No wonder, it's that alien guy."

However, it is not the reason why he joined Twitter. Rather, this was based on their careful strategy to gain continued support from Japanese voters, who tend to be impatient with their leaders, to stay in power long enough to implement the objectives he had promised to the nation.

According to media, the critical factor for DPJ's victory last year was the support from voters older than 40's. These voters used to support the conservative LDP but converted after being fed up with the scandal in which more than 50,000,000 pension records were lost permanently due to LDP's poor management. On the other hand, younger voters in the 20's and 30's favored the LDP because they are too young to worry about their pension, which they might not receive anyway.

Thus, Hatoyama and DPJ's marketing team started this innovative marketing approach targeting those younger voters who supported the conservatives during the election.

One month after joining Twitter, he has about 260,000 followers today. If this strategy succeeds, the alien will add another recognition to his resume.

You can find him on Twitter @hatoyamayukio

(Unfortunately, currently Japanese only)

Calculating the value of online relationships

Understanding online relationships..

Stephen Baker in BusinessWeek, "Learning, and Profiting, from Online Friendships."
http://www.businessweek.com/magazine/content/09_22/b4133032573293.htm

The article discusses the efforts at the moment that marketers and researchers are undertaking to aggregate the massive amounts of data created by the vast social networking opportunities that exist. Duncan J. Watts, a Columbia University sociologist and the head of a research unit at Yahoo!, says that "this flood of data for social scientists could be as transformative as Galileo's telescope was for the physical sciences. It gives us a new understanding of our world and ourselves."

The amount of data regarding online relationships that is available provides an opportunity for advertisers to develop a better understanding of their targeted consumers and those consumers' motivations. And as information is certainly not scarce now, attention most definitely is, and knowledge regarding relationships and motivation just gets more and more valuable as the competition for attention intensifies.

But online relationship networks have changed the understanding of interaction. The article discusses how friends are more likely to be interested in the same sort of products, attracted by the same type of ads. But what constitutes a friend now? Although I have about 250 Facebook friends, would I be interested in wearing the same type of clothes as all of them, or do I like the same movies, books, television shows, etc.? Sociologists, economists and market researchers are using all of the data generated by social network activity to try and understand of my 250 'Friends', whose opinions and tastes are most like my own?

Friendship Data

Cameron A. Marlow, a research scientist at Facebook, is working on a study to determine how close we are to our online friends. He and his team look at how often people communicate with Facebook friends, click on their links, news and photos, as well as how the communication flows (is it reciprocal?). They have determined the following about an average Facebook user (average user is considered by them to be one with 500 friends):

• S/he actively follows news on only 40 of them
• S/he communicates with 20
• S/he keeps in close touch with about 10
• And those with smaller networks (less than 500 friends) follow even fewer

As we've discussed in class, the jury is still out on Facebook ads. Although the traffic is there, obviously (as evidenced through the study findings above) people aren't paying a ton of attention to most of their friends. But if scientists like Marlow can discover how these relationships work, then the ability to target those small, intimate networks amongst the vast pool of friends could be very valuable to advertisers.

The article also discusses the internal value of social networks for corporations in increasing the flow of ideas and harnessing the collective intellegence of its employees on a large scale. IBM Research has one such experimental internal online social network, Beehive. The network (at 60,000 currently) connects employees with similar interests and expertise and suggests them as friends. In addition to more social interactions (posting photos), the friends also use Beehive to discuss patents and critique software code.

It will be interesting to see how these studies impact advertising and internal corporate communications, particularly once the data is used to attempt to better understand these relationships in an actionable way so that the hypothesis are tested and produce some real, quantifiable results.