"The Power of Like 2: How Social Media Works," looks at paid advertising on Facebook as well as earned media exposure-- meaning mentions of the brand made by Facebook users in status updates and the like. It is based on the experiences of large brands such as Best Buy (BBY.N), Starbucks (SBUX.O) and Target (TGT.N).
The report follows up on a July 2011 paper, "The Power of Like: How Brands Reach and Influence Fans Through Social Media Marketing."
It swipes back at recent research questioning the effectiveness of Facebook messages. A Reuters/Ipsos poll published last week showed four out of five Facebook users haven't bought a product or service as a result of advertising or comments on Facebook.
Most brand exposures on Facebook occur through users' news feeds, comScore said, rather than visits to dedicated brand pages on Facebook.
Fans - consumers who click a button that they like a certain brand or product - tend to outspend others for that particular brand, comScore said, citing examples such as Amazon (AMZN.O), Best Buy, and Target. Purchase data comes via information from loyalty clubs, credit card companies, and third-party collectors, with the permission of the study participant.
In the case of Target, Facebook and comScore studied two groups. One group, made up of fans of Target and their friends, saw "earned" messages about Target - updates about Target that run in news feeds and the like.
The second group was made up of Facebook users who weren't fans of Target and saw no messages. Both groups had identical purchase behavior at Target prior to the study.
After the four-week study, the fans who saw the messages were 19 percent more likely to buy goods at Target than the group that didn't see the messages, and their friends were 27 percent more likely. A comScore spokesman said he didn't know how much messaging the groups were exposed to.
To measure the impact of paid advertising, ComScore conducted a similar study involving a national retailer. It looked at groups of Facebook users who were exposed to a paid online Facebook campaign about that brand, and a test group that was not. Again, the two groups had identical purchase behavior before the study.
By the fourth week of the study, the group that saw the messages was 16 percent more likely to buy goods at the retailer than the group that did not see the messages.
Separately, Facebook said it had conducted research on about 60 campaigns to measure their return on investment, or how many dollars in sales were generated by every dollar spent on Facebook advertising.
About 70 percent of campaigns showed a return of three times or more on the money spent for the advertising, a spokeswoman said. About half of campaigns showed a return of five times or better.
Evaluating the effectiveness of advertising has proved challenging for Madison Avenue, no matter the media, brands have long said. They find it hard to gauge how many people saw a particular ad, and connecting the message with purchases is even more difficult.
Facebook is a comScore client. Along with many other large brands, it hires comScore to measure advertising effectiveness.