Thursday, 25 November 2010

Klout. Nail. Coffin. Who cares?

[Showing true and visible Influence, "The swirl at the wingtip traces the aircraft's wake vortex, which exerts a powerful influence on the flow field behind the plane"]

This is my last post on @Klout - period

Tom Webster published his latest post on Klout. Tom is a man of mild and modest words, and managed to attract Joe Fernandez, Klout's CEO, to his previous post on Klout.

Here's some of what Tom said this time, next to telling his Klout score has been stable for the past few weeks:

  1. My “Classification” has bounced around enormously. In the past three weeks, according to the “Influence Matrix,” I’ve been labeled a “Specialist,” “Networker,” “Thought Leader,” “Explorer,” “Taste Maker” and now back (down?) to “Explorer.” Given the relative stability of both my score and my social activity over the past few weeks, this variation is inexplicable.
  2. My “Influencer Of” and “Influenced By” persons haven’t changed since I started looking at Klout several months ago. Most of the former, by the way, are almost nonsensical. One of them I don’t think has ever retweeted or replied to me, or if they did it had to have been only once.
  3. Similarly, my “Topics” haven’t changed over the past few months, either. The list includes things I haven’t tweeted about in many months, as well as topics that I don’t think I’ve ever tweeted about (Tribes?) Irritatingly absent from the list is Research.
  4. Other underlying metrics, like “True Reach” and “Network” have also been static – and I mean static – for many weeks. My Mutual Follows number has not budged, though it clearly changes every day.

Numbering bullets and emphasis there all mine. Number 1 should be relatively static, not crazy volatile like this. Number 2 should change, and not stay put. Number 3 should change every few times a day even. Number 4 is a clear confirmation of what I described in my previous post(s): Klout stats are flat-liners and statistically impossible

Tom ends his post with a plea:

I’m all for companies engaging in experimental marketing tactics. But those experiments shouldn’t make customers their guinea pigs. Differential treatment based upon loyalty or prior purchases? Fine. Differential treatment based upon a number that remains inscrutable? I’m not there yet.

Emphasis all mine. Joe Fernandez commented, and doesn't hold back this time.

Yes, it's early days for metrics like the Klout score but I think saying that companies shouldn't experiment is short sighted. You and everyone else trying to make a living from the social web should be happy that there are companies that are willing to experiment. This is how innovation happens and keeps this space moving forward. In the end it's pretty simple, if our numbers don't prove out and help companies drive some value in terms or ROI or happier customers than we'll be replaced with something else.

Emphasis all mine there.
Come again? The "Standard for Influence" is in its early days? An experiment?
We should be happy that Klout is experimenting with us, and labelling us, and assigning us a score? And not being able to cancel our account?
If the numbers don't prove out - I proved over and over and over again that they don't, they just don't prove out at all; but like Joe says, Klout's just waiting until someone else does it better

Joe's next remark is equally not-too-well-thought-off, I think:
I also think you miss the mark on thinking influence doesn't belong with loyalty and prior purchases. Those two factors are huge and should be rewarded but network value in the social connected world we live in is also critical. Think of how Southwest might have better handled the Kevin Smith had they been better informed

Emphasis all mine.
So if Kevin had been a social illiterate Northwest could and should have just treated him like crap? That's just great

Joe also tries to mitigate the situation again, but suffers from great memory loss there:

When we moved to daily scoring last week the effect on the classification system was more pronounced then we had expected. The team is working hard on this and it will be fixed very soon.

Joe's highly likely referring to Klout's blog post of November 17th, which merely announces the daily score, unfortunately also stating that

On your score analysis page you can see the new daily score data going back to October 21st.

So it's not last week that Klout entirely 180-ed their individual scoring approach and moved to daily scoring, it's 5 weeks ago. Not really a minor detail, certainly not one the CEO should overlook

Tom reacts to Joe with a comment that sums it all up, really, pointing out that differential treatment based upon a given Klout score obliges Klout to have accurate and up-to-date information, and a mechanism to fix anything if it's broken. Klout has none out of those three, AFAIC

Successful business depends on a delicate blend of Marketing, Product and Service. Marketing to attract consumers to the product, the product to self-support itself, and service to take care of the exceptions where consumers apparently have gotten dissatisfied with the product. If either one is too small or too large, you're losing or throwing away money and reputation - but it all evolves around the product

Klout's marketing techniques are fabulous. Very 1.0 and a bit 2.0 at times, largely targeted at a 1.0 market, but fantastically successful. Even my grandma claims that Klout is the standard for influence.

Klout's product is bad. It bends when Twitter squeezes the API limits like at the World Championships, but doesn't break and just gives you part of the score, letting e.g. Brian Solis score 53. It introduces new ways of slicing and dicing you, but badly tests that, resulting in a total score drop of 20 in one day, and regaining 11 in the next 12 days - it's like little goblins operate the insides of the Klout Machina. Last but not least, it makes up scores that don't exist - or should I call that "statistically extrapolating"?

Klout's service is horrific. Or rather, non-existing. In June, Klout wrote me
We do not have a dedicated support team
Even if your product is perfect, you must be able to support it. And Klout's product is far from that.
Like a kid putting its hands in front of its eyes, thinking that is a perfect way to hide, Klout has used 1.0 enterprisey communication ways to handle questions and complaints: see no evil, hear no evil, speak no evil. I've seen this behaviour often at servicedesks that have a "close-call mentality": they are scored and rewarded on the number and speed of closed calls, so that's what they aim for: to close calls asap. Help the customer, repair (the damaged reputation of) the product? Really - don't be silly

The time has come for Klout to make up its mind: 1) sell out and get the heck out before the Klout bubble bursts, or 2) re-establish the balance between marketing, product and service

I wish them bests of luck, or should I say "All the best". Happy Thanksgiving - don't be a turkey

0 reacties:

Post a comment

Thank you for sharing your thoughts! Copy your comment before signing in...