Anatomy of a crowdfund, week 2: branding and networks

October 23, 2012

I’m happy to report that as of last Friday (Oct 19th), which marked the 1/3 point in our 45-day crowdfunding campaign Crowdsourcing Discovery (CSD), we raised 1/3 of our $25,000 fundraising goal.

 

As of this writing, the headline numbers are: $8,565 raised from a total of 125 contributors. The average contribution = $68; the median contribution = $25.

 

The big news from Week Two was that our project was mentioned in an article about science crowdfunding that appeared in the print and online editions of The Economist! This has to be the most awesomely alliterative title in the history of science journalism: “Many a mickle makes a muckle:”

 

(the tag line under Aahnold reads: Redistribute this, punk)

 

How did this come to pass? Lou Woodley (follow her @LouWoodley) of Nature, whom I met at the Spot On NYC (formerly SONYC) monthly salons, graciously put me in touch with Akshat Rathi, a UK-based science writer. Akshat interviewed me the day after our campaign launched, and we had a great discussion about what motivated the CSD project and science crowdfunding in general.

 

Switching gears, let’s look at some data and charts and networks. Let me begin with what should by now be the familiar donor distribution:

After two weeks, there’s a rougly 2:1 ratio of $25 to $100 donors. So how are the average and median contributions holding up? Here’s an updated table of daily stats:

 

 

To be expected, the average number of contributors fell by around half from 9 in Week One to 5 in Week Two. Similarly, the average contribution dropped three-fold from $766 in Week One to $255 in Week Two. And weekends are always a drag. Another notable milestone from Week Two: we crossed the 100th contributor mark!

 

The question that I get all the time is how many contributors are already in my social network? Let’s start by crunching some numbers. Plotted here is a breakdown of all $25 and all $100 contributors according to whether they are connected (or not connected) to me in a social network:

 

More strangers gave me $25 vs. $100. Well duh. How often do you give $100 to a stranger?

 

What I find interesting to think about is the connectivity between contributors who are also my friends. I used a freely available program called Gephi to create the first of many social-network representations of contributors and their connections to me and to each other. In other words, people are nodes (the filled circles), and friendships are lines.

 

Consider this real-life 10-person social network comprised of me (the green node) and 9 contributors, 8 of whom are my friends. To protect the innocent, let’s refer to everyone as A thru J. Notice the triads connecting me to different pairs of mutual friends, e.g., A-D-E. Also notice the two different kinds of singletons, F and I:

 

 

 

There’s a lot more I can say, but I’m interested in hearing your feedback first. The comment pool is open. Jump in!

In the next update on Week Three, I’ll take a more bird’s eye perspective on these social networks of science benefactors. In other words, more fun with Gephi!

 

  • LJStewartTweet

    Ethan. What amazes me is actually how hard it is to crowdfund things. You’ve got an important project, and coverage in Wired and Economist, and still $25K target hasn’t been hit. You note 1/3 of the way through 45 days (15 days) and you raised ~1/3 of the target (~$8K), but is the rate of contribution dropping over time.. Would be nice to see a graph of the amount of donation per day since start and maybe the rate of donation per week to see what the current trajectory is to getting this fully financed. Who’s reading Wired and Economist? Why aren’t there more non-connected people donating out of the blue?

    • ben

      As I understand it, crowd funding projects usually start with a burst, have a bit of a lull in the middle then end with a burst. Means there’s a lot of hard work to be done yet (as I’m sure Ethan is aware) but I think it’s definitely doable.

      I like the mapping of the larger funders with Gephi. I presume it’s possible to do that with the smaller funders? A much larger network but it would be interesting to see how it would be clustered.

      • http://twitter.com/eperlste Ethan Perlstein

        Actually, to clarify: the small network graph in this post was not filtered based on contribution size. It’s just 9 people on FB, 8 of whom are my friends, and I was curious to see how these friends connect to each other.

    • http://twitter.com/eperlste Ethan Perlstein

      The contribution rate has dropped over time. But that’s to be expected. These campaigns are U-shaped, meaning the contributions are not evenly spaced across the campaign but are bunched up at the beginning and the end. There’s not any data out there from real-life science crowdfunding campaigns that I can find reporting on the flatter part of the U-shaped curve. So I don’t know how day-to-day contribution rate stacks up.

      We have gotten a fair amount of strangers who’ve given because of media buzz. But not all media buzz is made equally. The Economist article didn’t include a direct link to our donation page. That means a reader would have to Google my name and RocketHub, find the project, register, etc. On the other hand, the Wired article did have a link to the project page, and that definitely drove traffic to our project page.

      Live and learn! Thankfully, we have the whole second half of the campaign still ahead of us.

  • A nonny mouse

    How often do you give $100 to a stranger?

    I’mma plead the fifth on that one…

  • Arnie Perlstein

    LJ is correct, it really is very difficult to crowdfund at this point in time. Ethan has done a remarkable job so far, and while I know he wishes it were easier, I think he has learned a great deal about the process because necessity really is the mother of invention.

  • http://twitter.com/BraveBosom Brave Bosom

    I have some thoughts on how you could improve your fundraising efforts (though my fundraising goal isn’t as big as yours!). I am doing something similar here: http://www.causes.com/actions/1695873