You just crowdfunded a research meth lab!
On Saturday morning November 24th, our 52-day grassroots campaign to crowdfund basic biomedical research, which we dubbed Crowdsourcing Discovery, began its final day. We were just south of the $20,000 base camp, and still another $5,000 trek from the $25k summit. My heart told me would make it, but our fundraising data called for a black swan.
Spoiler alert: we got our outlier, and it was awesome:
Here’s how it went down.
On the previous day, Black Friday, we received a jolt of fresh media, which when combined with my exhortations on Facebook and Twitter netted us 25 donors, tying the previous one-day donor record set on Day 1 of the campaign. These 25 donors contributed around $1,000.
So we needed to do 5-times better the very next day to close the gap. But we hadn’t experienced a day anywhere close to 100 donors, nor raised more than $2,000 in a 24-hour stretch. What’s the difference between a campaign that stalls at 80% versus a campaign that reaches 100%? Did we have the secret sauce?
Then I remembered what I had calmly written a week ago in my final campaign update. According to stats published on crowdfunding site blogs, a successful campaign raises 50% of the goal in the first and last 10% of the crowdfund drive, and slightly more in the last 10% than in the first 10%:
We were on pace out of the gate, but then we grew more slowly than average, so we had to make up 37% of our goal in the last 5 days of the campaign. Thanks to a great turnout of Facebook and Twitter supporters, we raised most of the shortfall in the final 24 hours:
Throughout that last day, we average 4 donors per hour, and at our peak just before midnight on Sunday November 25th, we had 14 donors in a single hour. Sure, that’s nothing compared to the tens of thousands of dollars raised per hour by Mathew Ingram (aka @Oatmeal) during his crowdfunding campaign to build a Tesla museum. But for us, we were raising in an hour what we used to raise in an entire day!
By the time the dust had settled, we’d raised $6,000 from 124 new donors, almost exactly 5-times higher than the preceding day. Who were these donors? As I just mentioned, my Facebook network turned out in a big way, but my tweeps really take most of the credit, as I’ll explain in the next post-game analysis, which will be out this weekend. In the interim, let’s focus on my Facebook network, because I started visualizing it a few weeks ago.
These are graphs of my Facebook network, which is comprised of 698 friends (nodes) and over 3000 connections (edges). The size of each node is proportional to its degree, which is the number of edges to other nodes. In other words, the large nodes are the hubs, the small nodes are orphans. The graph on the left is color-coded by friendship clusters. For example, the large red cluster is Grad School. The graph on the right has the connections wiped away for ease of viewing, and now there are only two colors: donors are in yellow, and non-donors are in blue.
17% of my Facebook network contributed to our campaign, and with the exception of a cluster of science friends I’ve made in the last 2 years, being a donor doesn’t depend on how I know you or how well you are connected to my other friends. Content over connection, perhaps?
I’ve only begun to think about these questions more carefully now that the campaign in behind us. Which brings me to my last point. The experiment part of our project will start in the weeks ahead, but the meta-experiment has already yielded a fundraising dataset that will be of use to the greater community.
I will be analyzing the Crowdsourcing Discovery campaign dataset and then writing it up as a paper in the open, a process that will serves as a model for how we intend to analyze our actual dataset when experimental results start to trickle in.
Thanks again to our 385 supporters — vive le Open Science!
For a summary of week 6, please go here.
For a summary of week 5, please go here.
For a summary of week 4, please go here.
For a summary of week 3, please go here.
For a summary of week 2, please go here.
For a summary of week 1, please go here.
And for a summary of the first 96 hours, please go here.