Social recommendations and clickthroughs

Dan Zarella has a great blog post from last year which examines where the highest click-through rate for a URL in a tweet is, depending on where the URL was placed. Surprisingly, he found that links placed one quarter along the length of the tweet actually had the highest clickthrough rate.

This led to a lot of discussion around how tweet should be constructed, and whether we should be more commonly using something like “Headline / Link / Info”, rather than the more conversational “Headlne / Info / Link” model.

Dan’s heat map of clickthroughs

This puzzled me at first, because I couldn’t I couldn’t think of any examples of anyone ever placed a link in the middle of a tweet.

But then I began to think about the old-school retweet and recommendation – where you retweet content, but add a personal recommendation on the end of it. The more commonplace model for this is “Headline / Link / Personal Comment”.

A recommendation-style tweet from the rad Danielle Warby.

 I began to wonder if in the original data used in Zarella’s study, we would be able to see see whether a lot of the tweets involved retweets or links where personalised recommendations are added?

If this is the case, I think this really cements the power of social recommendation: irrespective of someone’s  reach or influence, this data clearly shows that someone personally recommending something is far more effective in creating clickthroughs, than simply by sharing a link. Brand advocates, from a marketing perspective, have a very important role to play in the way that new people approach or see a brand for the first time.

However, in the context of our decentralised media environment, I believe that it’s this structure of personal recommendations is also very important. I think it could be taking the first steps in addressing the challenge of ‘how does quality content rise to the top?': because people actually talk about it, not just share it.

I suspect if we experimented with a “tweet this article” combined with “what did you really love about it” would see a significant increase in page views and traffice over a simple “tweet this” model. Only content which people really love will they add the extra effort to personally recommend to their friends.

Our media landscape is changing – but so is the way were finding and recommending what is the best content. We should keep an eye on how people are finding the best content as the mediums we use change: because this could be the centre point of how our media is found and consumed in years to come.


Got something to say? Go for it!

Links:sac lancel pas cherchristian louboutinair max pas cherair jordan pas cherchaussures louboutinabercrombie pas chercasque beats pas cherair max one pas cherabercrombie pas cherstylo montblancair max pas cherstylo mont blancnike air max pas cherair max pas chermoncler pas cher