There I was, happy as a clam with my Klout ranking of "Thought Leader" and a Klout score of 62 when suddenly my Klout score plunged to 1. On a one to hundred scale that put my social networking influence slightly below the neighborhood cat. It turns out I wasn't the only one. All of Klout's millions of users had had their scores reset to 1.
Now you think "What nonsense! Who cares what your Klout score is!" I'd agree... until a few months ago when I was in the early stages of a book deal and the publisher insisted on knowing what my Klout score was.
So what is a Klout score? I quickly found out that, according to Klout, that the Klout Score is the measurement of your overall online influence. The scores range from 1 to 100 with higher scores representing a wider and stronger sphere of influence. Klout uses over 35 variables on Facebook and Twitter to measure True Reach, Amplification Probability, and Network Score."
I soon discovered that many companies, and not just publishers, were using Klout scores to determine if they want to work with social networking "experts," public relations people, marketers, and, oh yes, writers and journalists.
With all the talk about social networking influence, people wanted a quick and easy way to tell if someone, or some organization, was really influential in social networking or just talking a good game. Klout, somewhat like Google's PageRank, gives people a quick and easy way to determine who counts and who doesn't.
Now you can certainly argue that neither Klout nor PageRank accurately shows how important or influential a person or a site is, but in the work-a-day world people use them all the time. Klout's not the only one in the social networking measurement business. PeerIndex and Empire Avenue offer similar services.
I woke up this morning to an avalanche of emails from people assuring me that they have been creating great content and engaging with their network and that there was no way their Klout scores should have dropped so sharply. Something was definitely wrong and it quickly became clear that all of our scores had fallen with most people landing at 1.
Every day between 4-5am PST we load updated scores for over 75 million social media profiles. Due to the growing dataset size, scoring took longer than expected and our pipeline started to load today's scores prematurely. Since that data was in an indeterminate state for the last step of our pipeline, what generally would cause a red flag, instead, defaulted everyone to a score of 1. By 6:30am PST the team was working on the issue and by 8:00am PST scores were reloaded and returned to normal.
While our API [application programming interface] and site have averaged more than 98% uptime for this year, this issue caught us off-guard and we apologize for any inconvenience this may have caused our users or API partners. Today we will be re-instrumenting our data load process to make sure we are immediately notified and that the system rolls back any drastic score changes. We will also be formalizing our communication flow to make sure that messages immediately are posted on the site and that our API partners are notified of these issues.
I was glad to see that Klout both fixed the problem quickly and admitted to their problem publicly. Too many companies-cough, Sony, cough--mishandle their public problems.
Does Klout really work? Is it an accurate measurement? I don't know. All I know is that businesses that might hire me take it seriously so I have to take it seriously as well. And, based on the number of people I had writing both to me and Klout this morning about their Klout scores, I'm not the only one taking it seriously.