The 90/10 Split
In presentations I give on word of mouth marketing for WOMMA, this slide always causes debate. It shares research findings from the Keller Fay Group on where conversations about companies take place. People are shocked to see the online percentage so low.
People are further shocked to learn, according to more Keller Fay statistics, about 1.3% of word of mouth conversations about brands happen on social media websites. Given the reported widespread usage of Facebook, Twitter, blogs, etc, how could that number be so low?
At the recent WOMMA Creating Talkable Brands conference, I heard Brad Fay, coo of Keller Fay, address this issue head-on. Fay explained, “It’s not that the online conversation is so small. It’s the other is so big.”
The big number Brad refers to is … 3,000,000,000. (Ahem, we’re talking three billion here.) That’s the number of word of mouth conversations Keller Fay estimates take place on a typical day in America. If 1.3% of word of mouth conversations about companies happen on social media websites, then that’s 43-million daily conversations. A VERY sizeable number.
In an email exchange with Ed Keller, ceo of Keller Fay, I asked him to support his company’s offline/online word of mouth findings.
“At Keller Fay we are measuring word of mouth conversation, not readership of consumer generate content. Lots of people might read information on social networking sites, but contribute infrequently. Especially when it comes to brands. There is recent research cited in the attached from digital agency Razorfish that says about 2% of online Americans post something online “daily” relating to brands.
People read statistics about the huge growth in social media and the numbers are impressive. There are fewer stats generally reported about the ~3 billion brand impressions created each day via offline WOM. The social media stats do not come close to that number. So the disconnect is that the offline number is so large, but not generally reported, so there is no point of reference from which to compare the social media stats.”
Ed Keller shares more explanation in this editorial piece. WOM Marketers … pay special attention where Ed says,
“… social media alone is not going to be the pathway to word of mouth success for most brands. It is a pillar, an avenue for conversation. The role of social media for consumers is, for now at least, more about connections with other people than connections with brands.”


Although this is a huge number, it covers the general spectrum. When you narrow it down to a specific category, such as restaurants, I’d say the ratio actually changes to 99% OFFLINE versus 1% ONLINE.
impressive data! thanks a lot for sharing it John!
Most of the time we forget that real WOM is bigger than we could watch online. Maybe this happens with products, causes or initiatives that involves a not so techie audience, or maybe the Buzz responsible still needs to know how to pack it to to maket it portable (online), too.
And, IMHO, this would lead us to the innevitable conclusion that online monitoring tools, are not the only measurement tools of WOM that we have to deploy to.
cheers!
@RolandoPeralta
Agreed Rolando. The original social media, face-to-face and voice-to-voice conversations, will always be the bigger playing field. Tougher to measure. Tougher to quantify. Tougher to share on a spreadsheet to show its financial worth. And in many ways, tougher to spark. All which makes online social media the more attractive path for many marketers.
Joel … I don’t have the WOM data specific to the restaurant industry, but your estimate is probably close. Like Brad Fay says, we should see the 1% as being a small number. In the billions of WOM conversations that happen daily, 1% accounts for millions of conversations.
The debate over offline vs. online WOM misses the point. The issue is that marketers need to start by determining who they’re trying to reach and influence and what the strategic objectives are for the program. People first, strategy second, communications channels (offline, Twitter, Facebook, etc.)third.