The ever-excellent Don Bartholomew, MetricsMan, provides an overview of some of the Twitter applications that purport to measure “influence.” He says:
Influence is contextual not absolute. An individual may have the ability to influence certain people in specific subject areas. Authority and trust are important constituent elements of influence. Do they have the authority to speak within a particular area and are their words and deeds trusted? The notion of coming up with an influence score without context is inherently flawed. It might be interesting, but it is not actionable.
Read the post, especially if you use Twitter, but even if you don’t, much of the content can be easily extrapolated to other forms of social media. A fair number of social media “experts” are bottling measurement snake oil these days, and the rigorous concepts Don discusses are the antidote for such chicanery.
Tags: Blog, communication, communication experts, communication methods, discuss, effective communication, measurement, Public Relations, Research, Social Media, Twitter
[...] This post was mentioned on Twitter by Mark Schaefer, Sean Williams. Sean Williams said: Cautionary tales on Twitter metrics from @donbart. Like @kdpaine, I wish I'd written his post. I could only do this: http://bit.ly/5oxvQY [...]
Thanks for sharing this, Sean. I think the concepts Don discusses are useful even beyond social media. I am a big fan of measurement in the communication profession, but we always need to be transparent about exactly what we are and are not measuring. I think this gets back to your prior post about objectives, too.
This has often been a challenge when measuring the effectiveness of media relations efforts. Even those programs that extend beyond awful AVEs typically stop at measuring the quantity and quality of media coverage. Achieving a particular volume and quality of media coverage is almost never the ultimate objective of a media relations campaign. The goal of media relations — as with other forms of communication — is the influence of attitudes, opinions and behaviors that Don mentions in his discussion of influence and Twitter. Measuring the frequency, tone, prominence, etc. of media coverage has its place (and is certainly better than using AVEs), but it should not be confused with measuring actual outcomes in a target audience.
Just because a great story about our company, product or candidate ran in a targeted publication doesn’t necessarily mean that it changed anyone’s mind or behavior. Obviously, we hope it has that effect, but I wouldn’t blame our clients if they expected more compelling evidence.
Jeff
Jeff, thanks for stopping by. I agree that the esteemed DonBart offers very useful stuff indeed. And, we are in violent agreement that “more compelling evidence” is crucial. That’s why unsealing the measurement “black boxes” to the harsh light of critical evaluation is so important. Now, if I develop my own “black box” I may not be as supportive of opening it up to competitors! I might be sore tempted to keep that little fella quiet long enough for me to cash in. But, you bet, we need to strive for moving beyond mere descriptive statistics to real outcome measurement.
Thanks again for your usual thoughtful post, Professor Dafler…