A bit over a week ago I listened in on John Mack's Pharma Marketing Talk podcast with Kru Researchabout their experiments with engaging the twittering public in pharma marketing projects.

The results, highlighted in John's post, are quite notable and say a lot about the psychology of perception, persuasion and marketing. I think they also highlight the elephant in the room of social media marketing. What constitutes authenticity, how it could be stretched and what does it mean for marketers and audience.

So what was Kru's specific test and what does it mean?

The idea was to measure how easy is to get Twitter users to respond to a follow request, depending on the type of a profile used by the marketer. The beauty of this test is that there are only a few parameters you can vary in your Twitter profile: the picture, the short description and perhaps your profile background.

John distills Kru Research test plan down to basics:

In this experiment, Krū Research created several fictitious Twitter accounts:

1. A regular person with no association to insomnia or a drug; this was our control
2. A person who mentions they have insomnia in their profile; this was considered a patient-peer unaffiliated with pharma
3. A person with insomnia who is representing an unbranded insomnia website; this was our paid patient opinion leader profile
4. An insomnia brand

The fictitious brand was named "Restira" and the Twitter brand account, which no longer exists, was @RestiraRx.

The percentage of Twitter users who followed back the marketer's account decreased steadily in the following order:

• Insomniac & website: 14%
• Insomniac: 13%
• Regular person: 7%
• Pharma brand: 5%

Of course patients come across as more trusted than brands 

This should not be a surprise. Personal stories are always easiest to relate to. They do not look like impersonal corporations. They do not come across as having a marketing purpose. They do not appear to have a profit motive. But they could also be entirely fictional, which apparently does not diminish their effectiveness.

The real eye-opener should be not that a patient is considered more trusted than a brand. The real point here is that the patient's story could be completely fictional and this is unlikely to be noticed and questioned. Indeed can you tell on Twitter whether people are who they say they are? There is no way unless you have a pre-existing relationship. Yet, this lack of authentication still leaves a personal story performing better than an explicitly branded message.

Now is this dynamic specific to Twitter or any place on the Internet?

There is an old saying that on the Internet nobody knows you are a dog. Kru's Twitter experiment merely highlights what savvy Internet marketers have already known. Testimonials sell and in most cases the visitor never bothers to verify them. A first-hand testimonial of course is the best. But credibility can be passed in other ways. A fictitious patient character, could do the job just as well if it is not intended to deceive, but merely illustrate. On the other hand a promotion that makes false claims, once exposed is certain to cause backlash down the road.

But wait, there is more! What if such profiles and/or testimonials are created by a third party, not directly controlled by the brand? Say Internet commenters, reviewers or testimonial submitters? What if they are not faked, but incentivized? What if a brand would donate money or in-kind support to a patient support group that would do this kind of marketing? Will you be able to tell what motivates someone to say what they say and do what they do? I doubt it.

One thing seems certain to me. The line between appropriate and questionable use of testimonials is gray and will keep getting grayer.