What we learned building Gildme.com: you need to build a community #baylaunch
We thought Gildme.com would be a fantastic startup idea. Something we could build and launch in one week. Something that was economical. Something that people wanted.
We saw that Redditors had a strange existing behavior: they “gild” other users’ comments for $3.99 a pop. All you gotta do is click “give gold” on any reddit comment, and voila, the author’s comment becomes emblazoned with a little gold icon.
Yea. People are actually paying $3.99 (enough to buy a sandwich and a drink from Safeway) to put a gold icon on a random stranger’s comment on the Internet.
According to Reddit Gold’s about page, having “reddit gold status” means you also get access to some exclusive features, but after studying the list of services, you realize these are features only a truly hardcore Redditor would appreciate, but most casual Redditors probably don’t find super valuable.
So then, giving Reddit Gold is about all creating feelings of goodwill. This makes sense, since studies show that spending money on other people makes us happier than spending it on ourselves.
So given this behavior, we thought perhaps people would want to “gild” users with something more tangible: they get mailed a toy gold coin along with a card.
We dreamed that if this service proved to be valuable, people would also want to gift each other flowers or even real gifts on Amazon. We thought perhaps people would want such a service for Twitter as well. However, in the interest of launching quickly according to lean startup philosophy, we decided to focus on toy coins for Redditors, and expanding to these other ideas and social networks if there was demand.
So in 1 week, we built and launched Gildme.com.
We marketed it by submitting to a few different subreddits. We didn’t actually get that many upvotes (I think we had a score of only 16 on our top post), but we got quite a few visitors this way. We also got some love on ProductHunt, with 59 upvotes. On launch day, we ended up getting about 2,000 unique visitors, 34% from Reddit, 24% from ProductHunt, and the rest from other sources. We had a small percentage of traffic (1.24%) from Facebook, even though we spent $5 on a Facebook campaign and posted status updates to our friends. And we got basically no traffic from our LinkedIn status updates. We didn’t get around to tweeting about it, although a few strangers did when they found out about our service.
Ironically, the night before our launch, we learned about redditgifts.com, a service that’s gotten a lot more traction by organizing “Secret Santa” gift exchanges, and helping out charitable causes like giving to schoolteachers’ classrooms. We watched their founder’s TEDx talk.
“Doh!”
That was definitely a “Doh!” moment, as we smiled goofily at one another and realized we just spent a week building something that already existed. It was the same vision of giving a stranger a gift without having to know their address. Although they say it’s all about execution not the idea, I don’t think we were going to be more passionate about “gift-giving to strangers” than these guys.
Lessons Learned
This week was really a lesson about user acquisition channels.
We naively thought that since giving gold was an existing behavior on Reddit, that Redditors would magically adopt the behavior of mailing toy coins and notes. It turns out we have to think more carefully about how exactly that process of adoption actually happens.
For Reddit Gold, there’s a “give gold” link on every single comment, inviting people to give gold. There’s visibility after someone is gilded, in the form of a gold icon. And very often, it’s good reddiquette for the recipient to edit their comment to then thank people for the gold. This creates an enormous amount of visibility, that a service like Gildme currently doesn’t have. In fact, we probably won’t have many repeat visitors to our website after our initial launch.
On the other hand, redditgifts.com’s way of organizing Secret Santas and gift exchanges around charity and other causes is a much more viable approach: you get to repeatedly launch to users, whereas we can’t repeatedly keep posting to Reddit that our service exists.
So in short, it’s not enough to be able to get users to know about your service, you also need a strategy for retaining initial users and building a real community around using your service. For a service like this to succeed, we really need to learn to be deeper thinkers about community-building.
In the end, we had a couple Redditors try to place real orders, but only one went through because of a bug. With a grand revenue total of $5, this is not enough traction to justify further investment. So we’re moving on and building a grocery delivery service next week. Stay tuned.
Special thanks to Michael Lee and Gautam Jain for working on this project with me. Thanks also to the Reddit community and ProductHunt community for the valuable feedback. To see our other projects, check out Baylaunch.com
