Social media monitoring is a frustrating business. Some days it seems there is an army of spammers whose only purpose in life is to create garbage entries to gum up tools like Radian6. In previous posts I’ve already railed about Autoblogs and Autocomments so my distaste is on record. Today I want to look more closely at some of the fringe elements, blogs that seem to exist between legitimate content and full-on autoblog garbage.
Here’s an example of a typical blog run by an e-Patient, Six until Me. This blog is super-relevant. Not only is it full of insights by the author, but there are insightful comments throughout.
On the other end of the spectrum is this obvious delinquent. This site is cobbled together from bits and pieces of other sites and makes no sense whatsoever. The sentences have the semblance of grammatical correctness, and are not just keyword-stuffed forum entries, but they actually do not convey any meaning.
These entries are the most devious as they can easily fool search engines into thinking they are legitimate and rank highly for their chosen keywords.
Which brings us to the posts I’m interested in today, the ones that are either truly works of horrible quality or spam but which are hard to distinguish. For example, the blog Pharmacy Made Simple seems to be nothing more than a large number of articles stolen from WebMD, MedPage Today, Drugs.com, the FDA, Medicine Net, and other places. The sources are cited, but there is no inherent value in the posts and they are typically only snippets of the originals without links back to the sources.
So the question remains… why is this person, who purports to be a pharmacist, spending time copying these articles over? There are no ads on the site to benefit from any traffic generated nor are there any obvious sponsors. There are many brand names mentioned, but only in context of the copied articles.
This type of material is some of the hardest to purge from social listening services. It can even find its way into selected examples because the analysts are rarely experienced healthcare marketers and they are looking at hundreds of posts a day.
Spam Quotient by Channel
If you’ve read this far you’re probably pretty disheartened about social media listening and the amount of garbage you can expect to find. All is not lost, however. Not all channels have the same amount of spam in them and there are techniques that can be used to combat this disease.
Channels with the Least Spam
LinkedIn: let’s face it, this platform is essentially resumes. While we may joke about “padding” on resumes in the real world, this site has the highest authenticity of the common social platforms.
Facebook: when used as intended, Facebook becomes a central feature of its users’ lives. This deep integration means that most Facebook accounts are legitimate and spam volumes are very low.
Channels with the Most Spam
Blogs: as a technology, blogs lend themselves to being used as keyword magnets on “autoblogs”. There are even WordPress plugins for this express purpose. This channel has by far the most spam.
Twitter: because it is essentially anonymous and is used to point toward blogs, both legitimate and spam-focused, there can be a moderate amount of spam found here. This channel can be difficult to analyze because many users “let it all hang out” and will use extremely poor grammar, sometimes to the point where it is indistinguishable from spam.
The rest of the channels fall somewhere in between. Currently, the only solution for this problem is focused attention on social media reports by analysts who are attuned to the important elements of the brand’s messaging and environment:
- Legitimate brand mentions
- Common usage patterns
- Patient motivations
- Relationships between patients and caregivers
- Sensitivity of the condition
What techniques does your firm use when reviewing social listening reports? All ideas welcome!