Wednesday, August 15, 2007

Executive Preview for "Measuring the Ripple" Now Available!

I am pleased to announce my latest research on word of mouth marketing communication. This is from research that has been over a year in the making and is the result of an important academic-industry collaboration.

Due to publication restrictions I am unable to release the full report, which will actually be a co-authored piece to appear in the Word of Mouth Marketing Association's Measuring Word of Mouth, Volume 3 research book in November 2007, so I wanted to make available the following executive preview.

You can read the executive preview here in this post, but I recommend that you download the PDF version at my download page as it is more comprehensive. The last academic-industry research report I released on my blog, the "To Tell Or Not To Tell?" report on the role of disclosure in organized word of mouth marketing programs, generated a great deal of commentary from readers so I invite the same level of thoughtful discourse regarding this piece.

Finally, this line of research has been especially important to my work in the word of mouth space, not only for subsequent academic publication on the topic, but also because it has led to the formation of a new company, called ChatThreads, an independent word of mouth research company that derives actionable insights from how word of mouth communication spreads among individuals and social networks (but more on that later).

Without further ado, I give you the...

Executive Preview for “Measuring the Ripple: Creating the G2X Relay Rate and an Industry Standard Methodology to Measure the Spread of Word of Mouth Conversations and Marketing-Relevant Outcomes” [Download this as a PDF]

From the childhood game of telephone, and our personal experience with rumor and gossip, we are all familiar with the idea that word of mouth (WOM) spreads from person to person. We have also heard that WOM spreads in multiples from one generation, or degree of separation, to the next – one person tells two people who tell four who tell eight, and on and on it goes. And then there are those instances where the buzz about a product, service, or idea seems to explode exponentially. When you add in the fact that peer-to-peer conversations and recommendations are among the most trusted forms of communication, it is no surprise that many marketers find word of mouth initiatives a compelling proposition. But childhood games, folk wisdom, and alluring anecdotes aside, how do we really know how far marketing-relevant conversations spread, and the impact of those conversations? And what data exist to help marketers set norms and benchmarks?

To answer these questions, Northeastern University collaborated with a leading word of mouth media platform, BzzAgent, Inc., to develop an industry-standard methodology that reliably tracks the spread and impact of peer-to-peer conversations over time and across both offline and online channels. The methodology has been developed and refined for marketing campaigns across a range of product and service categories – from beverages to books, food to pharmaceuticals, and personal care to paints – allowing for the development of category-specific and third-party validated norms and benchmarks.

In the context of this methodology, a participant in a WOM marketing program will be described as a Generation Zero or “G0” Program Participant. The first set of conversational partners with whom this program participant speaks will be described as a G1 conversational partner. The people with whom they speak will be described as G2 conversational partners and so on, all the way out to GX. Thus, the provisional name for this methodology is the “Generational Relay to X” WOM Tracking Methodology, or “G2X.” See the Glossary for more details.

Key findings from studies using this methodology include:

o Previous research underestimated the relay rate (the number of people told at each generation, or degree of separation, from the initial source) by not taking into account how conversations continue to spread over time. Existing research by Northeastern University established a Generation 2 (or G2) relay rate of 1.65 for the number of new people told after a person had a brand-related conversation with a Program Participant in a marketing campaign. By investigating if people continue to have conversations about the product or service over an additional six weeks, the relay rate for campaigns represented in this research can be revised to 4.14 (that is, a person who talked with someone participating in a WOM marketing program told 4.14 other people about the product, on average, during a six week period of time).

o While a majority of the participants in the study (78%) had at least one additional conversation during the course of the six-week extended period, researchers observed a noticeable decay in the number of people those participants told over time. The decay rate was as follows: in the first two days after the initial conversation with the person participating in the marketing program, the “conversational partner” reported telling 1.43 other people, on average. During the next time period, one week after the initial report, conversational partners reported telling an additional 1.38 people about the brand. Three weeks from the first report, 0.76 new people were told about the brand. Six weeks after the initial report, 0.57 new people were told. Twenty-two percent (22%) of the people reported no additional conversations after the initial conversation with the person participating in the WOM marketing program.

o Researchers discovered that it is possible to reliably measure the reach and impact of WOM marketing initiatives across a range of product and service categories, which is an important first step in creating industry standard norms. Relay rates, and other key performance indicators, like awareness, trial, and purchase intentions and behaviors, can be measured on a campaign-specific basis, as well as across campaigns, so that managers can determine campaign effectiveness relative to benchmark data.

o Similarly, the reach and impact of WOM marketing programs can be independently measured and validated by a third-party. This opens the door for industry-wide third-party measurement that is seen in more mature media industries like radio, TV, and print.

o By using dyadic (two-person) or multi-party measurement designs researchers demonstrated how it is possible to validate each person’s self-report about conversational details and the impact of those conversations, thus overcoming a traditional limitation of self-report data. For example, in the case of a campaign for an electronic personal care product, WOM program participants reported that slightly more people participated in a conversation than their conversational partner peers (i.e., 1.83 versus 1.55 other people, a difference of 0.28). However, this difference was not statistically significant. Since results vary from campaign to campaign, managers can use this methodology to determine if there are statistically significant discrepancies in self-reported data (whether there is over- or under-reporting), and revise relay rates or other key performance indicator values accordingly.

o In addition to quantitative reach, the methodology developed can also provide rich, qualitative insight into how people engage in brand-related conversations. By looking at the locations where conversations occur, the activities in which people are engaged while talking, the medium used (face-to-face, phone, e-mail, IM, blog, message board, etc.), and references to other media forms (like ads, editorials, and in-store promotions), managers can gain actionable insights into developing and refining subsequent marketing campaigns.

Results from the first set of studies are scheduled to appear as a co-authored paper with BzzAgent, Inc. in the Word of Mouth Marketing Association’s Measuring Word of Mouth, Volume 3 research book, to be released in mid-November 2007. The full manuscript will provide detailed information regarding the methodology and demonstrate how it was utilized to independently validate the effectiveness of an actual WOM marketing campaign.

Funding for this research was provided as part of a Sponsored Research Agreement (SRA-0667) between Northeastern University and BzzAgent, Inc.

Download a PDF version of this executive preview.


Monday, August 13, 2007

Continued Discussion on Net Promoter Score

In my long absence from posting (over a month!) there have been some great discussions about the Net Promoter Score. Be sure to see some of the comments to my last post where Amy Madsen and Deb Eastman from Satmetrix have weighed in, Tim Keiningham from Ipsos-Loyalty (and author of some recent academic articles critiquing the Net Promoter Score), and Justin Kirby, who has blogging and conducting podcast interviews about the debate.

Additionally a new article on the topic, also written by Tim Keiningham and colleagues, has appeared in Managing Service Quality entitled The value of different customer satisfaction and loyalty metrics in predicting customer retention, recommendation, and share-of-wallet. This article is more of a micro-level, or customer-level, analysis and worth reading.