In my inbox first thing this morning was some extremely interesting research, published by Alexander Coppock and shared by Tom Baker of Bond International. I think it’s worth sharing further – and it goes roughly like this:
So…you send half your twitter audience a request to sign a petition and please retweet, by including a link in a tweet from your Charity twitter account.
To another quarter you send a direct message, saying “You are one of our most valuable followers…please RT this petition…”
And a final quarter get a message that says “You are one of our most valuable organizers…please RT this petition…” (N.B. this was a US – experiment!).
Then, after the petitions are signed, you send a third of the petition-signers an email that asks them to retweet, including a button to make it easy for them.
So what do you think happens next?
Well firstly around 4% of people who receive a direct message do indeed sign the petition…good stuff! Far fewer (just over 1 per cent) send out the retweet, but still there is definitely some sort of network outcome.
However, before we get too excited let’s not forget that when these users get a DM, they generally also get an email by default, AND they may previously have seem the general Tweet – so either message reinforcement or simple email responsiveness may actually be what’s working here…
Secondly, of those who DO sign the petition, around 50 percent also send out a tweet. This is, in and of itself, a very significant pushback against those who argue that online activism only works because of the ease of taking online action – a.k.a. the ‘slacktivism’ simplification. Clearly, a higher level of engagement is at work from these petition-signing individuals.
Thirdly, and very interestingly, the users who are addressed as ‘Valuable Followers’ significantly outperform the ‘Valuable Organizers’ in terms of their propensity to tweet. This may be because the reference to Follower has invoked an “identity prime” which evokes a higher responsibility to act within the paradigm of Twitter; or it may conversely be that the Organizer identity has evoked a form of cognitive dissonance, by over-aggrandising the action and making it all seem like too much responsibility. Alternatively maybe this language simply seems inauthentic in a Twitter context? More research is needed.
Finally, the most interesting finding of all is that over two experiments of essentially the same structure, reaching roughly 7,000 followers, not a single one who received the broadcast tweet sent out a retweet – or signed the petition. Not one. The known relative lack of social cohesion in this particular network, and a relatively low level of engagement are the most probably causes, but what we are also beginning to see here is a feature of this particular platform – Twitter – and its endemic behaviours compared to other social networks.
The report makes fascinating reading. Despite the relatively small sample sizes, the research discipline is profound. We need a lot more of this sort of research conducted if we are make good use of social networks for more than just reputation-management, and PR news triggers.
In conclusion, it seems that online social network activism does sort of work; and it works in much the same way as offline activism. But making it work requires a good dose of campaigning common sense, as well as some basic social identity theory and good insight into behavioural responses to language. Finally it requires realism and patience in anticipating emergent network structures, and platform-specific behavioural norms. Frankly it’s not easy.
What’s clear from all this, is that the current reductionist management instruction to many digital managers to “Drive up our twitter followers” is a dumb answer to an even dumber question.
Twitter is not just a social network. And it’s not just for individuals. Twitter followers are not a technical abstraction. They are social ‘beings’. That respond to contact in broadly predictable ways.
The former BBC Knowledge Management expert, Euan Semple is fond of saying: Organisations don’t tweet; people do. What’s clear from this analysis is that both can, and both do. And productively so. But only when intelligently-managed!
Tim Kitchin is client service director and director of consulting at Copper, the digital marketing agency.