Look at this fucking hipster has got to be one of my favourite blogs ever. Not sure about these knob-ends, but true influencers are golden. And every marketer is after them on the web. So what's the best way to measure influence? As a planner, finding out this kind of stuff is getting really important - though sometimes we already know the answer, but feel pressured to reverse engineer it through decks of data...
David Armando wrote
a nice little article on this, and I think his points make a sweet pt 1 for this unplanned trip into unearthing and measuring online influence.
Twitter influence
I see a lot of marketers farming as many followers as they can, and going after others with lots of followers - assuming that they're 'influencers'. But just like a FB like, a follower count is a pretty flawed indicator. Plus it can pretty easily be gamed. So, as a starting point, here's a more complete view:
- Lists - This is good indicator for a couple of things a) visibility and b) what kind of 'topical influence' you have, based on list categories
- Retweets - A RT says "this is good enough for me to spread"
- @Replies - This says: a) how much people want to engage with you and b) how conversational vs broadcast-focused you are
- Follower ratio - Common rules of thumb: a) a 50/50 ratio suggests someone who follows back anyone that follows them (including spammers and crackheads) and b) users with significantly more followers than they follow signals a selection process - and maybe quality
- Tweet volume - Consider this: a) high volume users with healthy engagement, list counts, retweets etc are likely to be offering value, while b) high volume users with low scores are just making noise
- Favourites - This is a good indicator of something...but Twitter users use this feature for lots of different reasons, so it's difficult to discern exactly what that is
- Quality and relevance of followers - While being the most difficult to quantify, this has got to be where it's at - targeting the right audience and earning their trust
More on quality and relevance
'Influence' is meaningless unless its attached to a category. Tools like
Klout and
Sulia attempt to do this kind of analysis - and this is where I think I'll take the next installment.
How do you measure Twitter influence?