I have a hate-love relationship with Klout. The social influence measurement tool has helped me figure out who I pay attention to in social media, and has given me some advice on who I should pay attention to. It has also, to my mind, proven far too subject to manipulation by sleazy multi-level marketing types (ironically, focused on Internet marketing and selling their seminars and books for making a lot of money on the Interwebs). In the past I have contended that anyone with a really high Klout score is one of the three — overexposed, a shameless self-promoter, or a sham artist. So I actually advised people to pay more attention to middling Klout scores (in a semi-tongue-in-cheek fashion).
Klout was gracious enough to put me on the beta list for their new version and I spent the better part of Tuesday trying it out. The revision is still in beta but should go out to the entire Klout userbase shortly. My impressions were largely positive, although I still reserve judgement on the utility of Klout. First, full disclosure. My good friend Andras is an engineer at the company and he pulled me into the trial (thanks, Andras). We have had some long conversations about my true feelings about Klout. And Andras took some of my feedback and applied it to the new iteration. So to start with, here is what I like.
The new Klout dispensed with “True Reach”, “Amplification”, and “Network Impact” subscores. They replaced it with a feature called “Moments” that “…show you how your posts from the past 90 days have impacted your Klout Score.” Moments are arranged in a vertical stream. The format makes it easy to see I enjoyed the stream of “Moments” as a better way to understand who interacted with what piece of content, and, naturally, the Klout score of that person. The UI is very clean and intuitive and it is far less complex to understand what types of posts are more likely to impact your Klout score (no surprise, original pictures have a huge impact). Kudos to the Klout UX team for pulling this off — it seems simple but as someone who has spent a lot of time hashing over UX minutiae, I know it is all but easy to do.
The Klout team also delivered a new “dashboard” look that breaks down the sources of influence (51% Twitter, 14% LinkedIn, 35% Facebook, for example) and allows me to drill down into the impact of the different sections of my social graph to better understand how and why people are interacting with my social content. That dashboard also includes a chart history of my score and also breakdowns of key metrics (lists, retweets, mentions, etc) on the various social networks. In an intro to the new interface, Klout explains that the changes also work on the back-end. The score now includes more sources, takes into account influence indicators such as Wikipedia mentions or LinkedIn titles, and taps other sources of social influence that may sit outside the traditional social graph metrics.
One thing I have asked for, and would greatly appreciate is even more transparency into the algorithms Klout uses to determine the weight of interactions and behaviors in its ratings, as well a detailed explanation of why they made those choices. I would also have enjoyed significantly more detail on the Klout score. As in, geographic influences on the score, time of date impact of posts, and other detailed information. Putting my hyperlocal hat on, I’d like to see impact from even more social graph inputs such as FoodSpotting, for example, and Yelp — also tagged to geography (yes, I know that some Yelpers are anonymous, although they now discourage it). I’d love to see a way that Klout could include self-service offers for high scorers from local merchants — a communications channel with those who score highly.
One thing I did not like in the new Klout interface is a carryover from the old one — the incessant requests to ask my friends to sign up. This is an anti-feature, in my book, a request that quickly alienates new users. Mass email requests are de classe and I believe that, increasingly, these types of broad introduction requests fall on deaf (and potentially annoyed) ears. What is still missing, to me, is coherent ways to use Klout aside from assigning a ranking to Klout members. For example, I would love to understand the Klout ratings of, say, people in a group of LinkedIn connections that I am considering tapping to reach a second-degree of separation sales or marketing target inside an organization.
At its core, too, Klout has great potential to become an analytics tool that shows us surprising things about our interactions with those in our social graph. I’d like to see that detailed more broadly, and with better tools to allow us to drill down into how our social graph interacts with our posts and, similarly, how we fit into the social graphs of others. I am not going to change my general attitude towards Klout. It is a tool that can be useful but can also be abused, in my opinion. I think the new version of Klout makes it harder for self-promoters to abuse Klout. And I like the movement towards more transparency and more simplicity in the data presentation and aggregation. I have a feeling the Klout guys are planning another rev already. And I’d like to see that when it rolls out.
Alex Salkever is an executive at a cloud computing company and a former technology editor of BusinessWeek.com. The views expressed in his column are his own and not those of his employer. His Personal Fight column appears every Wednesday on Street Fight.