Lucy Bernholz has a great post called "Metrics Are Good, Unless They Are Bad" which talks about the problems we encounter when we're trying to measure hard to measure stuff - like social media, social return, and social enterprise. And while Lucy firmly believes that it is possible to measure, sometimes people measure the wrong thing.
Her case in point is the recent Washington Post article suggesting that Facebook Causes had driven very little money to charity and was a failure. Several of us disagreed with the thesis money was the only one metric for success. As I blogged in response to the article, Causes is best for creating connections and awareness. If you didn't read the thread over at Allison Fine's blog, you can find it here.
Lucy also muses about what the right metrics are for social media efforts. (I've been obsessing about this topic for a while, you can read my latest musings here and here). Lucy brings up some points about metrics for Twitter. Whatever the tool we're using, the right metrics are those that can help understand engagement and relationships.
This last week I was at NTC 09 in San Francisco leading a session on Mapping Metrics To Strategy. However, while checking my Twitter stream I realized that the Queen of Measurement, KD Paine, was at a conference a few blocks away in the lobby bar. So, headed over to say hello and to hear about her masterclass "ROI of Relationships."
Over the weekend, SobCon was taking place in Chicago and I wish I could have been there - (next year, Lis, I promise) KD Paine was there and follow her Tweets and KDPaine pointed me to her slide deck.
There's a lot of great stuff in this deck. A couple of things that I really like:
- She talks about conquering our measurement fears. It's not just math phobia. It's about confronting what doesn't work and improving it. It gets at that failing formally point that Clay Shirky was talking about NTC Keynote.
- She puts the what to measure question in context from hits, to views, to engagement. And, the whole point is about improving relationships.
- She offers a frameworks output, outake, outcome - looks sort of like a logic model to me.
- She gives a seven step framework for measuring the ROI of relationships.
- She gives us some suggestions for tools
- Most importantly is understanding what your engagement index is.
There is also a detailed taxonomy for types of social media conversations. I happened to see a wonderful remix of this from Buzz Canuck called "The 27 Types of Twitter Conversations."
One of my favorite metrics geeks, Jim Sterne, who has very advice on "How To Measure Engagement On Twitter." He suggests setting goals and identifying audience before selecting the metrics. His approach is not to select a metric that shows impact, but to pick the right metrics to help you improve what you're doing.
He writes about an analytics tool developed by Eric Peterson that if you apply thoughtfully can help you improve the results of your Twitter strategy. A slightly different approach than demonstrating impact, Twitalyzer calculates your influence based on your signal-to-noise ratio, generosity, velocity and clout, and it also allows you to calculate a score for any other Twitter user you wish to track. You're tracking relative increases and decreases to your influence over time and helps you refine Twitter strategies.
So, how might this work?
(1) Identify goal and audience
(2) Identify Twitter strategy
(3) Use Twitalyzer for a baseline
(4) Implement strategy
(5) Use Twitalyzer to measure again
(6) Reflect on the changes