Klout’s Social-Media Influence API

26 Jun
June 26, 2011

Klout has developed a really cool API that gauges a user’s “Social Influence” by aggregating the user’s activities and interactions within the social realm, The API is currently free access (up to a certain limit of usage), and still in Beta stage.

The Klout’s Social Influence API currently seems to focus on Twitter user activities to measure the social influence score, although it does claim grabbing data from other social hubs such as Facebook, although it does not seem to be part of the current output, maybe something they’re working on in the background at the moment. One cool thing though is that they imply that they track traffic through links on your social posts, I assume this has to do with the open access analytic data offered by some URL shortening services, a very good measure in terms of social media marketing, since it allows one to estimates the ROI of approaching this particular user (as possibly part of a blogger-outreach strategy), and asking them to engage with the brand.

the numbers are nice (as in informative) and consistent with the user’s actual influence, at least within the examples I have tried. Especially since they provide an over-all view as well as a nice aggregate breakdown (stuff like “amplification” or “true reach”), which would probably allow one to “tweak” the Social Media approach based on campaign requirements and KPIs, for example brand awareness campaigns will have different social requirements from conversion based campaigns, etc. For more details on how the algorithm calculates the score, Klout gives out a brief description on their Klout’s Score page.

When you register for the API, you get issued an API key and given a usage policy (maximum calls per second and per day rates), which will be used to track your API usage. The API is very simple and uses a RESTFUL protocol, which is really nice if you want to test the output quickly in a browser, a sample URL looks like this:

http://api.klout.com/1/users/show.xml?key=[api-id]&users=[user-name]

Currently the supported [user-name] value is the registered Twitter username, although I would imagine in the future they would map out all the user accounts, and link your Twitter user with your Facebook name, etc. to give a more comprehensive overview.

An example output of one of the popular names in online marketing, Andrew Girdwood, Twitter profile ID:AndrewGirdwood:

<users>
  <user>
    <twitter_id>14179324</twitter_id>
    <twitter_screen_name>AndrewGirdwood</twitter_screen_name>
    <score>
      <kscore>63.91</kscore>
      <slope>0.05</slope>
      <description>creates content that is spread throughout their network and drives discussions</description>
      <kclass_id>0</kclass_id>
      <kclass>Broadcaster</kclass>
      <kclass_description>You broadcast great content that spreads like wildfire. You are an essential information source in your industry. You have a large and diverse audience that values your content.</kclass_description>
      <kscore_description>creates content that is spread throughout their network and drives discussions</kscore_description>
      <network_score>67.26</network_score>
      <amplification_score>34.44</amplification_score>
      <true_reach>1497</true_reach>
      <delta_1day>-0.08</delta_1day>
      <delta_5day>-0.18</delta_5day>
    </score>
  </user>
</users>

Cool huh?

Do you know of any other Social Media Influence APIs? please share in the comment section.

2 replies
  1. Graham says:

    It’s a nice little API and I especially like the classification and description of a user’s tweeting habits. While not exactly quantifiable alone, it gives a very nice categorisation of users that would be very interesting to tie back into the other stats, e.g. which categorisations tend to have a wider reach, which have the most influential network and so on.

    It’s interesting to note though, that users who have connected FB/LinkedIn will have higher scores than those who haven’t, something which should be accounted for if using the scores to perform a user ranking.

    Reply

Trackbacks & Pingbacks

  1. […] or loses influence by staying away from social media or posting content that dies on its arse.  Links Naji provides a nice breakdown of the Klout score in his blog post about the Klout API, so I won’t cover the same details here.  The good news is that the algorithm used to […]

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply to Klout online influence scores gaining extra sway | picnicerror.net Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>