Join 3,382 readers in helping fund MetaFilter (Hide)


Measuring Twitter Influence
June 30, 2010 9:44 AM   Subscribe

Klout.com attempts to measure influence on Twitter. Recognizing that follower count does not measure influence, Klout considers 25 factors to determine how engaged you are with your network, how widely your tweets are read, and how likely your tweets are to drive action (retweets, replies, clicks).

Some feel that Klout's game-like achievements turn Twitter into a game, but over 250 companies are using Klout's API to utilize the data, including third-party Twitter clients like CoTweet and HootSuite.

Here's a sample of the more detailed analysis you get if you sign in with your Twitter account.

They have a pretty interesting blog too:
* Why Follower Count Doesn't Measure Twitter Performance
* When Fake Twitter Accounts Beat Out Their Real Counterparts
* 10 Most Influential Tweeters in Toronto
* 10 Most Influential World Cup 2010 Players on Twitter
posted by Faust Gray (48 comments total) 7 users marked this as a favorite

 
Apparently I have a "score" of 25. Not sure what that's out of (25? 100? 1000?), so I'm not sure yet whether to be proud or ashamed.
posted by ThePinkSuperhero at 9:46 AM on June 30, 2010


Weird, I'm a 51, but I can barely tell what that is supposed to mean.
posted by mathowie at 9:50 AM on June 30, 2010


and how likely your tweets are to drive action (retweets, replies, clicks).

That's an interesting definition for action.
posted by Chuckles at 9:51 AM on June 30, 2010 [2 favorites]


Ah ha, wait!

The Klout Score is the measurement of your overall online influence. The scores range from 0 to 100 with higher scores representing a wider and stronger sphere of influence

So, ashamed it is!
posted by ThePinkSuperhero at 9:53 AM on June 30, 2010


Thank goodness someone has finally refuted the canard that Twitter users are self-indulgent navel-gazers.
posted by Horace Rumpole at 9:54 AM on June 30, 2010 [1 favorite]


The scores are from 1 to 100. You get a pretty detailed analysis if you sign in with your Twitter account. (Just click "sign in" in the upper right, while logged in to Twitter.)

That's an interesting definition for action.

It's a highly measurable definition, anyway.
posted by Faust Gray at 9:55 AM on June 30, 2010


Most influential topics

Panda Express
Pizza Hut
Desjardins Group
Zachary Quinto
Jimmy John
ZQ


What? OK, so I used to go to lunch at Panda Express and Jimmy John's fairly frequently, but I haven't stepped inside a Pizza Hut in over a decade. Desjardins Group is a bank that's in no way associated with me. They got me nailed on Zachary Quinto, though.
posted by desjardins at 9:57 AM on June 30, 2010


It appears to sign in, I have to connect on Twitter AND Facebook AND create a Klout account? No thanks.
posted by ThePinkSuperhero at 9:59 AM on June 30, 2010


In order to sign up it forces you to connect to your Facebook account. Um, no.
posted by desjardins at 9:59 AM on June 30, 2010


Some feel that Klout's game-like achievements turn Twitter into a game.

No more than Foursquare turned going places into a ga...hang on I've just been ousted as Mayor of the Canyon Road Chick-Fil-A.
posted by griphus at 10:00 AM on June 30, 2010


(It didn't make me connect Facebook, or create an account. Just approve connection with Twitter. That was 2 weeks ago.)
posted by Faust Gray at 10:02 AM on June 30, 2010


(Just confirmed with another account - Facebook connect is not necessary. I approved the Twitter connection, typed my full name and a fake email address and clicked submit, and got in.)
posted by Faust Gray at 10:07 AM on June 30, 2010


The only stat I care about it that www.twitblock.org says I have blocked nearly 7,000 people! That's right, I don't want followers!

I use Twitter Karma and thetwitcleaner quite extensively. I boot any link spammers, any bots, and any inactive accounts. I force people I don't like (or inactive people) to unfollow me.

I give curtesy follow-backs if the person is real, but I block 50% of the people that follow me, ignore 25%, and follow maybe 25%. Of those that I follow half will be unfollowed within a month or so.

If says my klout is a 34.
posted by cjorgensen at 10:11 AM on June 30, 2010 [2 favorites]


Heh, my account gets a score of 5. I'm not sure what I could have done to get a lower score than that, other than not signing up in the first place.
posted by burnmp3s at 10:11 AM on June 30, 2010


Wow, thetwitcleaner report is fascinating.
posted by ThePinkSuperhero at 10:14 AM on June 30, 2010


Ashamed? I'm and 8!

As an aside, signing up/in was up there with Facebook in terms of horrid user experience.
posted by slogger at 10:15 AM on June 30, 2010


Heh burnmp3s - same as me - but I haven't been working Twitter - LinkedIn is my preferred social network technology.
posted by jkaczor at 10:20 AM on June 30, 2010


Anyway I have a higher score than ColdChef so I'm happy.
posted by desjardins at 10:24 AM on June 30, 2010


I expected to see the word "influencer" here. I was not disapointed.

I have a score of 24. One thing: when I first loaded this, the Influence Matrix listed me as a "Socializer," but when I reloaded the page it listed me as a "Feeder." A few bugs in the system, I guess.
posted by brundlefly at 10:24 AM on June 30, 2010


Nature documentary voice-over: "For cjorgensen, Twitter is a harsh and unforgiving place, where the struggle to survive shapes every tweet....."

Cut to: the twitching, broken bodies of the unfollowed heaped at cjorgensen's feet

Voice-over: "...there is no pity or mercy here. No coddling of 'what I had for lunch' or 'check out my etsy shop' tweeters. No, cjorgensen's is a lonely road through the wasteland of social media. The journey of one man. To hear only the real, and to tweet only the true.

And damn the consequences..."

Cue music, fade out, credits roll
posted by BitterOldPunk at 10:25 AM on June 30, 2010 [7 favorites]


The Klout Score calculation is based on a rolling 30 day range from your last calculation date.

Well, that explains my low score. Hah! I've been mostly taking time off from twitter. I'd thought I'd had a much higher score on this at one point.

I really don't care about many of the usual metrics:

How to Get Fewer Followers on Twitter and Why This is a Good Thing

The Secret to Fewer Followers

Both links are guest posts I wrote for the twitter fail blog.

> Wow, thetwitcleaner report is fascinating.

It catches a lot of cruft. I don't know anyone that curates their followers like I do. Most people don't care who is following, but I've got a theory that spammers follow the following/followers or other spammers. Once I let one in I get bunches. Also, most people could care less about inactive accounts following, but to me this just inflates your numbers (the ones I just got done saying don't matter).

I, because I overthink beans, once went through 100,000 followers of @aplusk Of these 20,000 were inactive (meaning no posts in the last 90 days). @opra was even worse.

This means that nearly 1 million of Ashton's followers no longer tweet. That's amazing.

The whole idea of the Library of Congress acquiring the entire tweet catalog is silly. Many of the accounts are removed and recreated. There is no way to verify who was behind what account when.

Anyway, I spend too much time on twitter, thinking about twitter, or wondering why @delmoi hates it so much.
posted by cjorgensen at 10:29 AM on June 30, 2010


It's worth mentioning that you don't even have to sign in just to get your score. Just type your Twitter name into the box on the main page.

Which is good, because I don't know about you, but I suffer from an acute case of Create A Login Exhaustion.

I'm also very leery of going to website A and clicking through to sign into website B. Particularly when I know that signing in is unnecessary for report-gathering, thanks to Twitter's open API.
posted by ErikaB at 10:31 AM on June 30, 2010


I wanted to register skepticism at Klout's numbers, too. I have to question the accuracy of any system that scores Neil Gaiman's Twitter strength at only 71 out of 100.
posted by ErikaB at 10:36 AM on June 30, 2010


Since I have two twitter accounts (me and my site), I can compare klout stats in a different way. It does funny things, like my site is the influencer of..... then a picture of one of my mates that retweeted something my site posted two months ago. My site account has a lot of retweets, and there are some that retweet stuff the siteaccount says four times a week - yet they are not on this list. But on my personal account I'm the influencer of.... and then a list of people that I know don't follow me. Which is even odder.
posted by dabitch at 10:36 AM on June 30, 2010


It says I'm influencing Cortex.

CORTEX CAN YOU FEEL MY SUBTLE YET ALL-PERVASIVE INFLUENCE @_@
posted by fairytale of los angeles at 10:40 AM on June 30, 2010 [3 favorites]


Is there really no overall leaderboard?
posted by smackfu at 10:45 AM on June 30, 2010


cjorgensen has made some very valid points here. I am ruthless about followers myself and whom I follow.

As for Klout, I think it is meaningless. I tried it out when it first made the rounds, I don't know how their algorithm works but they don't take one key aspect into account. That is, being an influencer by proxy - you may have only 10 followers, for example, but if each of your followers have followers in the thousands and they happen to find some link of yours worth retweeting, it totally increases your reach beyond what you yourself could actually do.

Its this same problem with metric that first infested the erstwhile blogosphere back about half a decade ago when Technorati first made the news with its top 100 lists. You could be popular but where you influential?

Its a conundrum that the social networking/marketing guys have been trying to crack for some time - I was talking to a PhD student interested in this space and I pointed out to him that I'd bet the people who really influenced other's thinking or spread ideas were either hidden meme spreaders or did not want to be found.

There's way too much emphasis on how to "monetize your social networks" and not enough on real people building real connections and enabling real sharing...
posted by infini at 10:47 AM on June 30, 2010 [1 favorite]


I'm a 5 and am apparently "very new to social media."

In fact, I am very old to social media and just don't give a fuck anymore.
posted by swift at 10:47 AM on June 30, 2010 [6 favorites]


My active twitter account is 26, the one I stopped updating months ago is 47 (but it does still have more than double the followers).
posted by oneswellfoop at 10:50 AM on June 30, 2010


Also, under my Topic Summary it lists "Spymaster." Which I don't play.
posted by brundlefly at 10:52 AM on June 30, 2010


So, this is the information age, huh.
posted by boo_radley at 10:56 AM on June 30, 2010


Topic Summary

* social media
* Yoplait
* Mouse
* Breast Cancer
* Cats
* Long Beach


Uh, what?

I am apparently a 27, and (on this page load) a Syndicator.

Oh, wait, now I've taken an interest in Mefi and am a 13.
posted by cortex at 11:22 AM on June 30, 2010


@ [me] is a Thought Leader

You are a thought leader in your industry. Your followers rely on you, not only to share the relevant news, but to give your opinion on the issues. People look to you to help them understand the day's developments. You understand what's important and what your audience values that.

YEAH!!!

umm, or something...

posted by bitter-girl.com at 11:44 AM on June 30, 2010


Syndicator, Spymaster, Thought Leader. Sounds like Team Fortress mod based on telepathy.
posted by Babblesort at 11:50 AM on June 30, 2010


I have blocked nearly 7,000 people! That's right, I don't want followers!

Holy crap, no kidding. I'm on identi.ca and I've spent waaay more effort on blocking spammers and the "now I'm typing command $x!" idiots than following anyone interesting.
posted by DU at 12:00 PM on June 30, 2010


I'm a socializer! So, like, hey where's the afterparty?
posted by thekilgore at 12:09 PM on June 30, 2010


That's right, I don't want followers!

Well, to be fair I went to the Keyser Söze School of Blocking.

When someone horribly egregious came along I not only blocked that person, but also anyone following that person. I had a script that did this automatically. I pointed it at Anderson Cooper and there went 2,000 followers in one shot.

Followers mean nothing. Want 80,000 followers in 40 days? Just follow everyone that is following @whatma

I took my @SMmonkey account to 15,000 folowers in 30 days just to show people it could me done. I even did all that clicking by hand! Manually followed the twitter max every day. Unfollowed those that didn't follow back. I pulled it off. It actually took me like 20 days, but I had a tumor pulled out of my head in there as well. Once I got to 15k I committed twittercide and unfollowed them all.

The best way I've found of finding valuable people to follow is to check out the people that post to twitpic and yfrog. Spammers don't post pics.

Martha Stewart follows me. Make of that what you will.
posted by cjorgensen at 12:19 PM on June 30, 2010 [1 favorite]


These are the same people who would come up with some sort of chart or ranking system to determine who the most popular person in their high school was.
posted by flarbuse at 12:26 PM on June 30, 2010


This is broken, no? When I cycled through the three pages back to the first my Klout score had suddenly halved, I went from strong reach to little reach, and from little amplification to "@gnfti content is almost guaranteed to be highly amplified" but with all zeroes for "total retweets" and all the other stats.
posted by goodnewsfortheinsane at 12:37 PM on June 30, 2010


I also have my doubts it does any significant topic analysis for languages other than English.
posted by goodnewsfortheinsane at 12:38 PM on June 30, 2010


I have significant doubts about it's topic analysis for English, to be honest.
posted by cortex at 12:44 PM on June 30, 2010 [2 favorites]


Heh, my account gets a score of 5. I'm not sure what I could have done to get a lower score than that, other than not signing up in the first place.

Ha, 4! Beat that, twitternonrati!
posted by kmz at 12:51 PM on June 30, 2010


Klout is wonky. i used to be up in the 70s but now am in the 40s. then i look at the other stuff and am still a thought leader but all the topics it listed are completely off and i mean, waaaaaaaay off the mark.

i am a thought leader because people do come to me to break down the issues of the day. why do i know? because they ask me in emails or comments or tweets to do exact that :D

klout feels like another "OMFG IF WE CANT QUANTIFY THIS SHIT WE CANT MAKE MONEY OUT OF IT" faux-tool for corporate america to justify using social media or, for that matter, to talk to people like me. i think the numbers are bullshit although half the context they got right.

btw, cjoergensen, i dare you to unfollow now :)


posted by liza at 2:49 PM on June 30, 2010


yea @goodnewsfortheinsane i tweet in Spanish and Portuguese as well and i've been tweeting about the World Cup furiously. they're obvs not up to date and not up to any language challenges.
posted by liza at 3:00 PM on June 30, 2010


You don't take this Twitter stuff too seriously.

I am pleased that my Klout report included this sentence. (13!)
posted by immlass at 3:54 PM on June 30, 2010


My score is 43, and apparently I've built a personal brand around [my] identity.
posted by defenestration at 3:56 PM on June 30, 2010


Oh, and my most influential topics are "alcoholism" and "LinkedIn."

That is, in fact, a pretty fair summary of professional VFX work.
posted by fairytale of los angeles at 4:33 PM on June 30, 2010 [2 favorites]


Wow, Klout certainly is wonky.

I started a junk account and tried the cjorgensen method. After about 15 minutes of clicking, my junk account has 120 followers. 120 spammy, worthless followers.

Klout gives this junk account a 5 score, and says "Your audience relies on you for a steady flow of information about your industry or topic. Your audience is hooked on your updates and secretly can't live without them."

Without any real tweets, this account is somehow influential on the topics of design, art, beauty, dating, and jobs.
posted by Faust Gray at 10:53 AM on July 1, 2010


« Older Motorcycle modification means something entirely d...  |  Vuvuzela Chorale from Brahm's ... Newer »


This thread has been archived and is closed to new comments