Good counts on Facebook are hard to find. Bad counts – all those stupendous numbers on total users, number of posts, network size, yada-yada – are plentiful, but getting good counts on persuasion impact are few and far between. Which is why this is so useful.
A couple of B-school researchers partnered with an app designer to study Word of Mouth characteristics on Facebook that led to peer adoption of a new app. The app in question focused on movies and the film industry in general and allowed people to post and share comments on the topic with their Facebook friends who also had the app. The researchers ran a paid advertising campaign to attract about 10,000 Facebook users to download the app. The researchers working in conjunction with the app developer systematically varied the attributes of the app to affect the kind of WOM these users could use to inveigle their network friends to download the app.
You see the Counting value of this approach. The researchers build a sample of app users, vary the attributes of the app to create an independent variable with different kinds of WOM, then collect the stream of user actions with the app in their network to see how many friends adopt the app and under what conditions. It’s a really nice controlled experiment in the natural setting with tons of data flowing out of it.
If you are crazy for data and those regression equations with Greek letters, get the paper and read it. This team is pretty smart and you could learn a lot of nuance. For this post, however, I want to detail the large scale findings. That where all my persuasion interest lies.
Exactly 9687 people responded to an advertising campaign about the movie app and downloaded it. That campaign cost $6,000 to execute. This is the group from which we can now study WOM and its effect on getting new app adopters from the larger social networks of the original group of 9687 app users. Now, read the following carefully.
The experiment was conducted over a 44-day period during which 9687 users adopted the application with 405 users randomly assigned to the baseline control group, 4600 users randomly assigned to the passive-broadcast treatment group, and 4682 users randomly assigned to the active-personalized treatment group. Users in these groups collectively had 1.4M distinct peers in their local social networks and sent a total of 70,140 viral messages to their peers, resulting in 992 peer adoptions – 682 of which were in direct response to viral messages.
See the key numbers. Nearly 10,000 users have 1.4 million Facebook friends (about 140 friends per user which is a common network size on Facebook). Those nearly 10,000 users sent 70,000 viral messages (the WOM, right?). That produced about 1,000 app adoptions. So, if you spend $6,000 dollars on this tactic, you’ll generate WOM that earns about 1,000 positive outcomes, about $6 per adoption.
Stated another way, $6,000 will buy access to a potential population of 1.4 million people in 10,000 social networks and get about 70,000 WOM actions that will result in about 1,000 good outcomes (downloading a free application on Facebook).
Everyone can draw their own implications from this basic finding. Downloading a free Facebook app is the TACT and that’s exactly what I consider WATtapping, that Low WATT social media twitching that requires little thought, effort, or value, but can still be collected and counted. Social Media like Facebook and Twitter are great twitch collector networks, but some people think that a twitch is more than a twitch and that if you combines the zillions of twitches you’ve got . . . something.
So, you’re running a health campaign to encourage people to drink water or juice rather than Full Strength Soda Pop. What kind of WOM behavior change do you think you’ll get with your Stop The Pop App on Facebook?
You’re a marketing exec with GM trying to shed that Government Motors moniker and get back in the Free Market again with more car sales, say like with that New New Thing, the Chevy Volt, the electric car that will save GM and revolutionize transportation as we know it. You create that Get A Jolt From Volt Facebook app and use this ad campaign tactic to get WOM. How many more Volts will you sell with Facebook WOM?
Hey, you’re running for re-election with Hope and Change 2.0, so you build that Hope And Change 2.0 Obama App and follow this research to unleash it on Facebook. How many votes do you get on Election Day with this kind of WOM?
It costs you $6,000 to build a group of 10,000 Facebook users so you can give away a free app to 1,000 people through WOM. You can have potential access to 1.4 million people in all those networks and get 1,000 hits from those 1.4 million for a success rate of 0.000714285714%.
I have trouble finding good news in this research for Facebook as a persuasion platform. I can see fabulous news in just the research itself. The researchers have some great data that permits interesting testing of communication and persuasion in social networks. Hubba-hubba! But getting the Other Guy to Change in real time in the real world with Facebook?
The best persuasion interpretation of this study is that I might get 10% more app users if I run an ad campaign on Facebook. Remember the ad campaign got 10,000 app downloads and then the WOM from that group got 1,000 of their friends. Of course, this same kind of effect occurs with any form of persuasion. Persuade a group to your TACT with face-to-face communication, with TV or radio, with church sermons; that group then spreads the word through WOM to others in their social networks. There’s nothing in these data to indicate that getting the TACT through Facebook produces more effective or efficient WOM compared to any other communication channel, device, vehicle, or medium. Thus, WOM can make a Small Windowpane difference . . . if you are doing an incredibly easy, thoughtless, and free TACT.
Facebook beguiles. Those huge networks. Those huge networks with well described individuals. Think about the kind of micro targeting of Other Guys you can do and still have all that peer influence, Word Of Mouth, social influence. Good grief the potential is staggering. But then you get into the operation of those social networks and you begin to realize that you are confusing the word for the thing, eating the digital menu. Facebook is a twitch collector and not a social network. People fool themselves when they think Facebook is a social network the same way those Lenten dinners in a church basement are a social network.
Researchers once made a distinction between face-to-face interactions, calling them social, and mediated interactions, calling them, mediated! The persuasion mavens and VC behind Web 2.0 artfully blurred that distinction and dropped the mediation. You do not have a social relationship with your friends, families, colleagues, and those others you cannot now remember, on Facebook. And, all the science we’ve got about social relationships does not apply with the faux connections Facebook builds.
There may be a persuasion science within Facebook and the computer mediated communication networks of Social Media. But, using everything we know about true social influence and even mass mediated influence simply does not work with Web 2.0. My gut feeling is that Social Media is a fabulous fiction for attracting venture capital and a few big IPO kills, but that there is no persuasion there. It requires something too close to the Queen of Tomorrow for gratification.
Sinan Aral and Dylan Walker. (2010). Creating Social Contagion Through Viral Product Design: A Randomized Trial of Peer Influence in Networks.
November 30, 2010
Management Science, Forthcoming
P.S. The central persuasion problem with Social Media is how people think while using Facebook. The information field with a Facebook page (as with virtually all Social Media) is crowded with many different WATTage switches, Arguments, and Cues. Worse still, people enter the page with a particular thinking goal that conflicts with all the available information tasks on the page – you can write or read. Your Father’s Oldsmobile media like broadcast TV carefully controlled the information field and tasks of viewers. Sure, the ads announced themselves so that you could run to the bathroom or simply ignore them, but when you’re spending 8 hours a day in front of this kind of processing field, the ad will eventually get to you. Facebook has no strong way of controlling the information or the viewer’s processing state. Social media are a persuasion car wreck.