Twitter for speed; email for sure
Jimmy Leach on how the data on crowdfunding at Unbound draws some useful conclusions.
We’re getting more and more into the data at Unbound, to the point of having an actual data scientist, Noelia, whose own crowdfunding project is here. She’ll be looking at the Big Picture, profiling the science behind a great crowdfunding project. That level of technical will increasingly drive the recommendations and advice we give to authors and could even fit into the recruitment process. We will always be a mix of editorial judgement and crowdfunding nous, but the nous will be driven more by the numbers than by the grizzled experience of the past (and I speak partly of myself here, rather than the estimable Georgia, who is experienced, but never grizzled).
We’ve had some long conversations already, effectively circling the numbers and looking just where to start to make the changes and the nudges that will makes things, gradually but definitively, more efficient on the funding front.
And as part of that, we need to give out more advice, more widely, more often. Hence the resurrection of this Medium blog, which will be the discursive ‘guess what we learned’ while we also build a set of author resources that will be the encyclopedic version.
So after all that circling, where do we start?
We looked at referrals. The lifeblood of a campaign. Any crowdfunding author will spend a great deal of their time driving traffic to their project page. It’s vital, therefore, that they know what platforms are driving the most traffic to their page. There’s a correlation between traffic to the page and pledges. Obviously. How else are they going to pledge?
When advising authors on their crowdfunding, we’ve always advised that targeted, one-to-one emails. Pushing a personalised campaign to known networks, with personal and professional connections make logical sense. And it has the added benefit of working…
But while it works, it’s no longer the dominant force that it seemed, from the data we’ve looked at.
Now, first of all, a slight disclaimer. What follows is extrapolations from a very broad data set. It is all referrals to all books over the last 12 months. In here, there are exceptions, and your book/campaign may be one of them (and we’ll come to a theory about that shortly). So take the point. Look at your own campaign and assume this pattern applies to you, right up to the point at which you prove that it doesn’t.
So, the broad data still shows that the ‘direct’ route (by which we almost always mean personal email) is still the most effective route, around 28% of the revenue generated (we’re using the revenue figures as they’re the important ones)
The next strongest is Twitter (23%) which is (a) higher than we expected and (b) needs some caveating. And, at its most basic, that caveat is this:
Twitter works best for people who are good on Twitter
You don’t get that level of insight in many places.
But the point is this: The Twitter figures are, arguably, more skewed by outliers than others, because when it works, it works very well. A certain cadre of authors drive their audiences through Twitter (and did so before they launched on the platform). Authors with sizeable audiences and high levels of engagement. Such as Beth McColl, Aaron Reynolds, Tom Cox, and more. And it worked well for them. Because they were already there, and that was one reason why they wanted to crowdfund their book in the first place.
And what that also did was to drive the pace of the funding. These books were characterised by the speed of funding too. Well above average. When Twitter works well, it can work very well.
The other significant social channel is Facebook (lump together Facebook and mobile Facebook in the graphic above), which accounts for 13% of transactions. A decent enough figure, but bear in mind that the majority of the people connected to an author through FB will also, normally, be on their email list. Facebook and email can reinforce each other, rather than be entirely separate channels.
Beyond that, the other social channels are YouTube (previously dominated in Unbound history by Stuart Ashen) and Instagram which, even for highly visual projects, is not a big driver, due to the sheer clumsiness of the linking process.
The transactional email (the automated nudge when a pledge is started but not completed) is chunky supplier of pledges, but the original source of the traffic is difficult to ascertain. But you could maybe assume the same patterns as the rest. And, in a different way, the same might be said for Google search. We know they know about the book, we don’t necessarily know how they know about the book. There’s always known unknowns.
So what does this mean for an individual’s campaign? In many ways, much the same advice as we’ve always given. Direct, personal contact remains the bedrock of a campaign, and should be your first instinct. But when Twitter works, it can be transformational.