Heard it from a friend

Editor’s note: George Silverman is president of Market Navigation, Inc., an Orangeburg, N.Y., research firm.

In a past article in the December 1996 qualitative research issue of Quirk’s, I made the case for researching word of mouth, and wrote about how to do it. I’d like to build on that article and add some of what additional years of experience have taught us about the importance of word of mouth and methods of researching it.

Word of mouth is becoming an extremely hot topic. You are sure to be asked about this subject by clients and colleagues. Three recent books on the topic are business bestsellers and I heartily recommend them: Tipping Point, by Malcolm Gladwell, Ideavirus, by Seth Godin, and Anatomy of Buzz, by Emanuel Rosen.

They all give amazing examples of the power of word of mouth, but have nothing in them about how to research word of mouth and little in them about how to systematically construct a word-of-mouth campaign. (For more on that topic, consult my book, The Secrets of Word-of-Mouth Marketing: How to Trigger Exponential Sales Through Runaway Word of Mouth, published by AMACOM.)

How powerful is word of mouth?

All three books consistently underestimate the power of word of mouth. Just how powerful is it? It is thousands of times more powerful than all of the rest of marketing combined, and I’ll prove it using estimations. Let’s not quarrel over the exact numbers. It is orders of magnitude that are important here: the average consumer receives several hundred commercial communications per day (studies show 200-800), but only acts on approximately one every week or two. And these actions are usually weak actions, not to buy the product but to find out more information. So the average person receives several thousand commercial communications before taking an action.

Contrast this with the fact that when a person receives a recommendation from a friend, colleague, or trusted advisor, that person acts on approximately one out of three. And that action is often trial of the product. That is a ratio of three to several thousand, or one to about one to 3,000.

Another way of looking at this is that one person telling a few friends can be more powerful than a 1,000-person sales force working for a day. Or, a few people telling a few more people can be as powerful as an entire advertising campaign. Every marketer knows that when those people, in turn, tell other people, a chain reaction can easily overwhelm and reverse an entire expensive marketing campaign.

Barking up the wrong tree

All of this means that most marketing and marketing research is barking up the wrong tree most of the time. People are trying to understand the influence of advertising and sales, and trying to make them more effective, when they would do 1,000 times better putting the same efforts into understanding and influencing word of mouth. In fact, it is an illusion that advertising and salespeople are selling products. What is usually happening is that these media arouse interest, then people talk with their friends about the products or services, and then purchase. In other words, word of mouth is the intervening variable or trigger to trial, purchase, and adoption.

The implications for marketing and market research are enormous. We would better serve our internal and external clients by inventing ways to research and influence word of mouth.

How to research word of mouth

I’ve described how to research word of mouth qualitatively in the article referenced above, but let me briefly describe the methodology. The fundamental idea here is to have people actually engaging in word of mouth in front of you. And there are several ways to do this. The most effective we have found is to run focus groups in which you have invited enthusiastic users of the product to talk with interested skeptics. The users are asked to convince the skeptics and the skeptics are asked to raise all their questions, qualms, and objections. Usually, the enthusiasts will convince the skeptics. You then discuss with the skeptics what convinced them, or what moved them further toward a positive decision. In this way, you develop powerfully persuasive stories that can then be spread through other means.

Notice that this is in direct contradiction to conventional ways of running focus groups. Usually, users are separated from prospects to avoid “contamination and influence.” Focus groups are severely criticized because one person can sway the whole group, so steps are taken to avoid this “problem.” I realized about 30 years ago that this is usually not a problem at all. We want to understand influence. If the group is swayed, that is an important finding. If the group is swayed toward your client’s product, you have found out how peoples’ opinions can be changed. You may have found the key to persuading people to use the product. If the group is swayed against the product, you have discovered a vulnerability. Presumably, the competition is also running focus groups and discovering the same thing, to be used against your client’s product.

Opinion change in focus groups is not to be avoided, it is to be encouraged to emerge, to be observed, and then studied in depth. The opportunity to understand the underlying dynamics of opinion change is one of the strengths of qualitative research. Opinion change is often measured in quantitative research, but its underlying dynamics are rarely understood by quantitative research.

New techniques

Since describing the above technique, we came up with another very powerful strategy. One of our clients had a service that tracked the source of new customers. They didn’t just ask, “How did you hear about us?” If the person said that they were referred by a friend, they would actually get the name of the friend. In that way, we could tell which people were actually spheres of influence. We invited those people (people who had actually gotten more than three of their friends to sign up!) and some of their friends to be in the same focus groups. We would typically have two to three verified persuaders, and about seven to 10 of their friends. We openly discussed what the influencer said that piqued the interest of the referees, what information they found most persuasive, in what sequence, and what actually triggered their trial. We explored in depth what got the referrers to try to convince their friends. We actually used that to build a customer referral system that was so powerful it actually stressed the order capacity of the service.

Believe me, these weren’t the things that the advertising agency came up with. They weren’t the things the salespeople were saying. What motivated people to want to convince their friends, and what actually convinced their friends was a totally different set of benefits, expressed in a different way, and in a different sequence than was obvious.

Another way to research word of mouth is to bring together an expert with people selected from the population you are trying to influence. So, you might bring together a medical expert with a group of doctors. The medical expert, in effect, teaches a seminar. However, you just don’t let him or her lecture. You make it a rule that the information will only be forthcoming in response to questions from the people in the group. In that way, you force out the issues that are most important and you find out through probing what was most persuasive.

With a little imagination, you won’t have any trouble figuring out your own approaches to fit almost any circumstance. I’ll give you some idea stimulators: dyads, triads, mini-groups, super-groups, online groups, telephone groups (in which you split the groups, then re-merge them). The possibilities are endless once you get past the belief that different respondents have to be in separate cells.

So far, I have been writing about understanding the content and motivation of word of mouth. Clients will ask you to also measure the extent to which their products or services are receiving positive word of mouth. To be honest, I don’t have a clue about how to measure word of mouth quantitatively. I do, however, know many ways not to do it. Usually, a question is asked about how the customer heard about the product. Usually, about 50 to 75 percent of people claim that it was word of mouth that triggered their purchase. This only tells us that word of mouth is important, something we already know. It does not tell us which products are receiving what kind of word of mouth. If anyone out there has a way to do this, contact us because we certainly know clients who would pay a great deal of money for a solution to this problem!