Saturday, April 01, 2006

Glossing Over Word-of-Mouth Measurement & Metrics?

I recently received an e-mail from Marianne Spain, a student at Franklin University, about my recent presentation on word-of-mouth and consumer-generated media. She felt that my presentation glossed over the measurement piece of word-of-mouth marketing and wanted to know if I could provide her with additional information. Below is her e-mail andmy reply:

Dr. Carl

I found your presentation at while doing research for a paper at Franklin University. This topic intrigues me and your insights are spot-on. I'm enjoying learning all I can about this topic and you and WOMMA are great sources of information.

If you don't mind, I have a question. You indicate that the key challenges to implementing a CGM/WOM campaign are "Ethics & Societal Effects and Credible Metrics." You thoroughly explain the ethical aspects and fully discuss disclosure, but the presentation glosses over the measurement piece. Other than stating that it's difficult to put the metrics in place, do you have any other data about measures? Most marketers won't support a program that can't prove it's worth and my paper really needs that anchor!

I would appreciate any information you can impart. I eagerly look forward to your reply.


Marianne Spain
--- --- --- ---
Hi Marianne,

Thanks for your interest in my research. I discussed a little more of the measurement piece in my actual talk but it didn't make it into the slides (and plus I ran out of time towards the end). There are a number of different outcome metrics that companies are using, and many are based on the WOMMA Terminology Framework (disclosure: advisory board member). The metrics used depend a lot on what type of WOM programs you are talking about. For example, if you take an influencer and product seeding program that tries to get a product into the hands of specially screened people there are some commonly used metrics: purchase behavior, inquiry rates (when people go to a website, for example, to learn more information), and pass-along rates (how many people are told). Attitudes towards the brand can also be tracked, brand awareness, and likelihood to recommend are other common ones too. Matchstick describes how they track performance for their influencer and product seeding programs using the Terminology Framework.

The Net Promoter Scale/Score, which is a measure of likelihood to recommend a brand to a friend, is increasingly being used by a number of companies (WOM marketing firms and otherwise). For how WOM marketing firms are using this metric I suggest that you visit BzzAgent's blog to see how they are using the NPS for their campaign tracking in a pre-test, post-test design (what was the likelihood to recommend before the campaign and what was the likelihood to recommend after the campaign). Bazaarvoice is another company using the NPS too. The NPS has been correleated with revenue growth, both in the US and UK (some critiques of this research: 1, 2).

Ways of measuring ROI include control/test groups. BzzAgent has a white paper that tracks different markets in order to determine ROI (for example, City A that has the WOM program and City B that doesn't have it; if sales are higher in City A, and all other things are equal, then this is taken as a sign that the WOM program was effective). However this control/test group design has been critiqued by George Silverman since if it's a really successful WOM program the control group might get "contaminated" by the buzz from the test group.

Brains on Fire helped organize a grassroots marketing program where they were empowering kids in an effort to curb teen smoking, so the ROI outcome metric here was teen smoking rates in South Carolina (Rage Against the Haze program).

You could also monitor online chatter or buzz using various companies' tools (Cymfony, Nielsen BuzzMetrics, etc.) to see if there are increased levels of conversations, the polarity (positive or negative) of the conversations, who is making the comments, etc. The Keller Fay Group is tracking offline and non-publicly available online conversations on a nationwide basis with their Talk Track methodology (disclosure: consulting relationship).

There are more metrics and ways to measure WOM as well but I hope this provides a little more information for you. WOMMA also has put out a book called "Measuring Word-of-Mouth, Volume 1" that has a lot of interesting case studies. The WOMMA Research & Metrics Council will soon be putting out a call for Volume 2 (disclosure: co-chair of this council) and for its next metrics conference.