Skip to content

Instantly share code, notes, and snippets.

@ncase
Last active December 24, 2015 20:59
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save ncase/6861496 to your computer and use it in GitHub Desktop.
Save ncase/6861496 to your computer and use it in GitHub Desktop.
A quick summary of the Open Bundle backer survey stats

#The Backer Survey Data

It's been way too long, but I finally collected all the responses for The Open Bundle's backer survey I emailed out two months ago. Yes, emailed. I trawled through a hundred email responses manually to collect this data. I decided to use email because it feels more personal than a SurveyMonkey, and I wanted to strike up a conversation with some people. (And I did! You people are weird, in an awesome way.)

Anyway, this was the email survey I sent out:

  1. Who are you? (an artist, musician, developer, free culture advocate, other...?)
  2. Why did you back The Open Bundle? (for the vision, for the assets, to support the artists, other...?)
  3. Any other comments? (about you, this project, or anything else...?)

There was a 14% total response rate - about 100 responses out of 744 backers. I just want to say up front right now, take these graphs with a pillar of salt. In trying to make the data quantitative, I had to narrow down their answer to the first two questions to a single thing. (e.g. "I backed it for the vision" or "I backed it for the assets") When in fact, it was almost always a mix. (quote from one backer: "The vision is what pulled me in, but the assets were the goal.") Not only that, I was making up these categories while I was recording the data, because the categories I had set out beforehand weren't comprehensive enough.

But since I created a spreadsheet of responses anyway, I thought I might as well share some graphs anyway.

Here's the raw CSV file, stripped of any confidential, private, or otherwise identifying information. (CONFESSION: I accidentally used To: instead of Bcc: in one group of backer surveys. I care about privacy deeply, and I sincerely apologize for my fuckup. I had used Bcc: in all the other groups, but that time, I slipped. Someone recommended that next time, I should use Tinyletter, which is like a simplified Mailchimp with a focus on conversations.)

##Who are you? responses:

"Developer" just refers to any programmer who's not specifically a game developer. And "FLOSS" refers to people who described themselves mainly as open-source/free culture proponents. Again, people can be in multiple categories, and this definitely doesn't show that. Unsurprisingly, game developers & developers topped the list. I am surprised, however, by the relatively few people who identify as artists or musicians, although that makes sense considering the bundle is selling art & music - things that most programmers struggle to do well.

By the way, this graph does not imply that relatively few people are FLOSS advocates, (in fact the next graph will contradict this conclusion) just that they do not primarily identify as one.

##Why did you back The Open Bundle? responses

For this question, I tried to subcategorize the respones, because I'd otherwise lose even more nuance in collecting this data. I've distinguished between wanting the assets for your own projects vs for the public, supporting the open content/free culture cause vs the open bundle's specific funding model, and supporting these specific indie artists vs indie artists in general.

Two motives are the clear champions: The altruistic motive, supporting the open content/free culture cause, and the "selfish" motive, assets for possible use in one's personal projects. I think that's an important message to take away about human nature. It's not selfish XOR selfless. It's not even a gradient between selfish and selfless.

People are BOTH selfish && selfless.

And you know what? That's awesome. That means the raw ambition we associate with pure self-interest can be used to help the world. That means we can do well by doing good.

##Correlating other pledge data shenanigans

Right off the bat - there's practically no correlation between who a person is, or their primary motive, and how much they pledged. At least, none I can tell from this coarse of data. (Other than the fact that people who identified primarily as students couldn't afford to donate as much.)

For Donation Split, (default, all to charity, all to artists, custom) the only correlation between that and person/motive is that people whose main motive was to support the artists, chose "all to artists" for their donation split. Other than that, no correlation I can see.

Check out the Open Bundle Stats Post-Mortem for a more thorough look at all the other kind of pledge data!

##Conclusions

  1. People are not Selfish XOR Selfless, we're Selfish && Selfless.
  2. Qualitative data is awesome. (But next time, use something easier to manage like TinyLetter.)
  3. Seriously Gmail, why would you hide the Bcc: feature behind a dropdown menu. That's not minimalism, that's just bad design.

Thanks again to everyone who helped make The Open Bundle a success!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment