At this point, I think it’s safe to say that Grassroutes went a little viral during SOPA Blackout. It wasn’t like bird flu or SARS or anything, it just made a couple corners of the internet a little sniffly. (Pardon the pun.) Drew posted some numbers up over on our makeshift Grassroutes blog, if you want to see them.

By all means, Grassroutes probably shouldn’t have gone viral, not even a little bit. We started sharing it with people less than 12 hours before the blackout started. We hadn’t polished it much at that point. Our homepage desperately needed—and still needs—a bit of a redesign. One of our friends naively and prematurely posted it to Reddit, where it proceeded to get 0 upvotes.

So why?

  1. We chose a catchy way to promote it. Grassroutes was actually conceived as a counter-point to the SOPA widgets we were already seeing. (If you want to see where you’re coming from, you can watch my pitch at PennApps, which I still haven’t managed to watch all the way through—too self-concious—but which does explain our idea pretty well.)But then we realized that those SOPA widgets were only up on a few big websites. What if we told people to put a widget on their website—no matter who they were, no matter if they only had a prefab blog—and gave them a really easy way to do it? SOPA might have widgets already, but nothing so accessible as this. And Reddit, Wikipedia, et. al. were forcing SOPA to be a really, really hot topic.And aside from SOPA being a hot topic, it was a “good” topic. Making a SOPA fighting widget meant that we were the good guys. Unlike some of my other projects, this wasn’t something that seemed frivolous. I immediately went about the business of creating a “grabby” homepage. Forget the mellow browns of our main page, I wanted something that would yell at people. Red, white and gray, with a bold font and a little humor.
  2. We posted this catchy page on Hacker News. Hacker News was a great place to post something like Grassroutes, for a few reasons reasons. Firstly, a lot of the people on HN have blogs or websites, so we were reaching some of our users directly. Additionally, many HN users get excited about this kind of thing, and they have broadcast outlets like Twitter. (Shortly after posting to HN we started noticing Twitter buzz.)But maybe most importantly we got a lot of good feedback from HN users. People on HN are generally sharp and happy to give constructive criticism. For example, we needed to add a keypad: often officials make you work your way through a phone tree before you can actually leave a message or talk to someone. This hadn’t occurred to us. Hacker News straightened us out in short order.
  3. We started engaging people on Twitter.(Man, I love Twitter.) I think that this not only encouraged people to actually post the widget, but also made them like our team. This led to a handful of really nice posts, which in turn probably led to more use of the widget.We got a fair amount of traffic just from Twitter: We got retweeted by Reddit founder Alexis Ohanian and Boing Boinger Cory Doctorow, among others.
  4. We have awesome friends.We had a couple of experienced friends who were happy to give us advice, and that was indispensible and I am so grateful. But we also had a handful of other freshman-friends, from Seattle to Philadelphia, help us out in various ways. One started compiling websites that had promised to go dark for the blackout and found their email addresses. One started tweeting the thing at everyone.One of my friends, the incomparable Ben Schiffler, even made this awesome graphic and posted it to his tumblr, with a clickthrough to Grassroutes, where it garnered a respectable number of notes.

    It’s both so thrilling and so humbling when other people embrace your project like this.

  5. We reached out to the media.Ugh, I hate self-promotion, but sometimes it has to be done. I actually think that our biggest mistake here was not reaching out earlier. I waited until we had hundreds of thousands of unique visitors—from sites like Twitpic and Occupy Wall Street—before contacting the media. The result of this was that we ended up mostly on smaller, Seattle-centric sites. (Drew and I are both from Seattle.) While this was no loss, we probably could have gained more traction if we were on a couple of bigger sites, too. In retrospect, I should have reached out earlier and made more of a story out of our team. (Two tips from a friend of mine who is quickly becoming an old hat at getting on TechCrunch.)
  6. We didn’t rest on our (technical) laurels.When traffic started heating up (“Holy shit guys, we’re on Twitpic!”) we started worrying about keeping everything together. We were advised by a Penn upperclassman (who is an awesome authority on many things) to switch all our static assets to S3 while adding a dyno on Heroku. This seemed to be an easy and cost effective way to scale—and I’m not sure what would have happened without these changes.

I guess in sum it looks like we did just enough right. Hopefully I’ll be able to take these lessons and apply them the next time we whip up a project we’re really passionate about. And who knows what I’ll learn that time.



Powered by Facebook Comments