According to the recently released white paper, “The Value of Managed Word-of-Mouth Programs” (pdf) ... managed word-of-mouth (WOM) programs, like those from BzzAgent, generate more WOM than happens from everyday people. (For a sum-up of the paper’s key findings, check out this post from Ben McConnell, a deacon of customer evangelism do’s and don’ts.)
The study, conducted by Dr. Walter Carl (Assistant Professor at Northeastern University), tracked the WOM activities of 1,000 BzzAgents compared to a convenience sample of everyday, non-Bzz Agent people.
As happens with most white papers, this one reads very much like an advertorial giving credibility to BzzAgent’s word-of-mouth marketing ways. It’s heavily framed to support BzzAgent’s position on generating WOM. For example, the BzzAgent sample set is referred to as ‘WOM Volunteers’ when in reality, the so-called convenience sample is more a WOM volunteer than are the BzzAgents. (BzzAgents aren’t called ‘agents’ for nothing.)
Since these WOM volunteers are BzzAgents, it’s no wonder to me they reported generating more WOM than the convenience sample did. A requirement of being a BzzAgent is to report when and where they generated WOM for the campaign they’re participating in.
Could it be BzzAgents are more skillful at recognizing and reporting WOM activity than their peers in the convenience sample? And if so, wouldn’t that make the data in the study suspect?
I’d also like to learn just how similar, in demographics and psychographics, the two sample sets are. The white paper only outlines the everyday, non-BzzAgent sample as being college educated adults ranging in age from 18 to 29. Since the paper empirically states the BzzAgent sample set is more socially active, I’d like to know if BzzAgents are potentially more socially inclined than are the people in the everyday, convenience sample.
Hopefully these questions will be addressed when the complete study is published next year in the Management Communication Quarterly.
John, I had some of the same reactions as you did, especially regarding BzzAgents generating more WOM. My thought was that they are probably more attuned and aware of WOM than members of the "convenience sample." Most BzzAgents are probably in the habit of making a mental note of WOM in a conversation whether they are reporting it or not.
In any case, it always pays to be a wary consumer of statistics. Just raising questions such as these helps us to think about the research in a more critical (analytical) way.
Posted by: Paughnee | August 02, 2005 at 05:20 PM
Really good points, John!
Posted by: Mark Ramsey | August 03, 2005 at 12:57 PM
Hi John (and Paughnee and Mark),
These are really interesting points. I just posted a detailed reply to your comments at the Word-of-Mouth Communication Study Blog
Thanks!
Walter
Posted by: Walter Carl | August 05, 2005 at 05:41 PM