James Mayes

Blog: Calculating influence

In Social Media, Twitter on January 17, 2011 at 9:36 pm

I’ve read a fair amount of blogs (and associated comments) recently on the subject of influence on Twitter.  What is it, how is it defined, can it be measured, etc.  A quick scan of online reference sites offers a number of definitions. Since this is as much for my benefit as anyone’s, I’m going to select a definition which I feel is both simplistic and roughly appropriate to my intent: “the action or process of producing effects on the actions […] of another”.

What I really want to play around with here is the idea of measuring it.  There are a number of tools out there which try do this, Klout probably being the best-known.These sites don’t generally disclose their algorithms or methods – partially to stop people “gaming the system”, partially for competitive reasons. So to the meat of this. What do I think should be measured? How does the metric look?

Most systems start with follower numbers – but this is no good if the person in question tends not to retweet or share.  It’s also no good if the person follows thousands of people and is highly unlikely to see (or even be interested in) my output.  On the other hand, if someone’s happy to share my material on a regular basis, but has only limited followers of their own, what good will that do me? Finally – whatever number of followers and retweets are involved, it’s still of negligible benefit if the followers in question don’t represent my target audience.

If a platform was going to meaningfully analyse and measure influence for me, it would need to understand both my target audience, and the followers of whoever retweeted on my behalf.  Is the data in someone’s Twitter bio solid enough to form a judgement on? Typically, I’d say not – so the only measure available would be the keywords in previous tweets. Going down this route implies the platform would need to both analyse said tweets – but more importantly, actually access them.  It’s not the easiest thing to go back in time with Twitter…

So what other measure could be a useful part of the mix?  For me, I’d be interested in the type of thing tweeted. If I got retweets from someone who only ever retweeted others, I doubt it would carry much significance.  Ideally for me, a Twitter account contains a rich mix of interesting short tweets, links to longer material of relevance and reciprocated @mentions to other Twitter users – showing the level of relationship. Capture all that, I think you get somewhere nearer to a measure of influence that might actually interest me.

Can this realistically be done on Twitter? Possibly – but I think the answer may lie in opening up to other platforms in order to better cover the target audience part of things.  Using the more detailed profiles of LinkedIn might be one route to explore.

However it progresses – and I’m sure it will – this post has been something of a pain to me.  I have an idea in mind, a model if you will – but I’m not happy with the way it’s converted on the blog.  I feel like I’m still missing a key part and I’d welcome any thoughts in getting it clearer – if only for the sake of my sanity!

  1. […] This post was mentioned on Twitter by James Mayes, James Mayes. James Mayes said: Blog: Calculating influence http://jmay.es/fFyj43 […]

  2. Sleepless nights for you then, as this will run an run!

    Measuring people’s influence on Twitter is one of the more nebulus areas that needs work. As you say Klout is trying hard, but it just doesn’t represent the real deal yet.
    I still think that many still default to the numbers – numbers of followers,the number of tweets, the reach (through RT’s) and the tweet frequency. Obviously that is not the whole picture by any stretch of the imagination, but I would suggest that for the everyday person ‘on the street’ that suffices. Blimey any more for them and they will get ‘even more’ confused!
    Marketers, brand managers, promotion and product launchers will of course have a completely different opinion. THEY WANT the metrics.

    Good luck with your quest Sir Twitter Knight, but I feel that you may have to find the pot at the bottom of the rainbow before the answer is discovered!


    • Those metrics are, by and large, the easy bit. It’s tying those to relevant audience when it gets both interesting, and difficult. And if I find that pot at the end of the rainbow, rest assured I shan’t be worrying about Twitter influence for a while…

  3. Hi,

    Those are some interesting questions you’re posing…I attended a social media measurement seminar today where we discussed this very topic!

    I am just getting to grips with Twitter to be honest, but it seems to me that there isn’t one preferred tool which can deliver a ‘bigger picture’ view of the reach and influence of social media channels. Since starting to use social media properly, I have been advised to try various platforms that can monitor different areas of my activity – a fairly disparate approach really!

    As you pointed out, the levels of engagement vary greatly with the user, and there definitely seems to be the call for some cohesive way to measure the quality of your following.

    Please do let me know if you discover what it is!!


    • Hi Laura! If you decide to blog on today’s seminar, do let me know. I’d certainly agree the approach out there is disparate at present. I’ve seen a few things which come close, but only at the (very expensive) enterprise end of the market. Fine if you’re Pepsi, but what about the rest of us?! If I do find something closer to my needs, I’ll be sure to blog a review.

  4. I think calculating influence via social networks is nigh on impossible, but I can say that I have gained business via social media that I otherwise would not have been in with a shout of getting including, as recently as yesterday, what will hopefully turn out to be regular work from a Dutch company. Do I care what my influence is? Not overly. I will keep putting my mix of blogs, professional opinion, RTs and lame attempts at humour out there and if collectively they reap rewards so be it, but anyone who tries to suggest there is a secret formula or that they can measure down to the enth degree what your influence is on social networks is, in my opinion, telling porkies. In short, it seems to me to be hit and miss and very random.

    • I agree it can seem hit and miss – but I can also say for certain that in my own use, there are patterns to be drawn out. It’s not easy, it takes time and analysis – and in search of greater efficiency, I’ll constantly seek tools that can help in this analysis. I think you’re right in suggesting it’s almost impossible to nail the whole enchilada, but many of the tools I’ve tried to date are so far wide of the mark it’s just baffling!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: