Word of mouth

In 2005, the Philips marketing team was looking for ways to advance the promotion of its Sonicare electric toothbrush. They already knew that sales were fre­quently driven by word of mouth, whether through recommendations from dentists or conversations with family and friends. Those who tried the product for the first time were often struck by the difference it made in their brushing routine, and would subsequently proselytize for the brand. It was with this in mind that Christy Mommsen, then brand manager of Philips’ Sonicare, ana­lyzed Sonicare’s marketing mix and the cost of acquiring new users.

The high cost of television ads, and the recognition that Sonicare already benefited from word-of-mouth rec­ommendations, led to the conclusion that a professional word-of-mouth marketing campaign, supplementing the traditional TV buy, made sense. “It was efficient to reach out to Soni­care users and get them to encourage friends and family,” Mommsen says.

Philips’ decision is part of a growing trend: More companies are budgeting for WOM marketing. In 2006, spend­ing on WOM hit $981 million, up from $722 million the previous year, making it one of the fastest growing nontraditional marketing sectors in the US (PQ Media). When early ver­sions of WOM, or “buzz,” campaigns became prominent about five years ago, they inspired passionate debate over transparency and public trust. But in an ROI-obsessed world, per­haps the strongest argument against WOM wasn’t that it was deceptive, but that it was not measurable.

WOM has become accepted
Today, WOM is an accepted part of the marketing mix and has its own set of ethical guidelines and mea­surement techniques that in many ways are informed by practices used by direct marketers.

For its Sonicare buzz efforts, Philips turned to BzzAgent, one of the first agencies to focus specifically on WOM marketing. Founder and CEO Dave Balter says his company main­tains a heavily segmented network of 400,000 “influencers” — all of them volunteers — who can be tapped for any given campaign. “We capture about 250 data points on each con­sumer, then offer them campaigns based on their profiles,” he says.

The appropriate influencers are shipped the product to test it and for­mulate an opinion. They are then free to spread the word however they want, be it conversation with friends and family, posting blog items or submitting reviews on the Web.

But there is one important rule: “People have to disclose their involvement as volunteers,” says Balter. WOM is not about deceiv­ing your friends, he explains.

Influencers dominate interactions
Each influencer writes a weekly report for BzzAgent documenting their interactions. The first part comprises statistical data such as the date and location of the conver­sation, the relationship of the per­son spoken to, and whether it led to a sale. “It’s all the things a direct marketing program would want to know,” says Balter.

The report then includes a “nar­rative box,” where the influencers can characterize the conversations. Were they passionate or neutral? Was the listener interested? Did he or she want to know more? “What you’re getting is a real-life report of what happened,” Balter explains.

The client then receives a range of data, including how many people were communicated with, what type of people are being reached and a net promoter score — which is a single-question scoring system that aligns recommendations with future sales. This score is calculated by subtract­ing the number of detractors (those unlikely to recommend) from its pro­moters (those likely to recommend). Information on ROI is also provided, such as coupon redemption rates, test and control marketplace returns and intent to purchase.

The Philips campaign started out small, with a test in Atlanta consist­ing of just 500 agents, but the results made Philips happy enough to com­mit to larger and larger campaigns: 30,000 “buzz agents” the following year, and 45,000 the next.

“The key metrics for us were the number of toothbrushes sold per buzz agent, as well as the net pro­moter score,” says Mommsen.

Mommsen says it’s wise to take the detailed, but self-reported, infor­mation WOM provides with a grain of salt, since a buzz agent can choose not to tell the truth. But she notes that WOM offers information that other marketing channels don’t pro­vide. “You get hard data you don’t get with a consumer who has seen your TV ads,” she says.