Recently, me and two of my friends launched adore.ly, a Facebook-enabled web app that allows you to “adore” someone, and your identity doesn’t get revealed unless they “adore” you back. As the app has been online for a few weeks now (2 weeks as of this entry), I have had a chance to reflect back and think about what I did, what went right and what went wrong. This post serves as a way to record my thoughts and to share my experience with other students trying to get something out there.
We built most of the app during the Facebook hackathon that was held at UW. 24 hours of coding and brainstorming later, we were able to present a crude version to the crowd. The next week was spent polishing the app and working more on the front-end. We basically had adore.ly up and running in under 7 days.
It was very easy to find flaws in our app and we were tempted to push the release date further and further back. However, just launching it was the best move we made. It made us fix bugs and iterate faster since we knew people were looking at the site.
A slight caveat: We launched Feb.7 and posted on Hacker News and Reddit (with good results) on Feb. 8. In hindsight, waiting till Feb.14 (Valentine’s Day) would have probably been smarter.
I think this is one area where we didn’t do as well as expected. We felt that these “anonymous” adores wouldn’t be “creepy” since they came from your friends on Facebook, and so you knew all of them. We were 60% right. Turns out that guys agreed with us, but some girls didn’t. As a result, these girls did not want to send out adores, or did not respond to adores that they were receiving.
The point I’m trying to make here is to understand the psychology of the user and whether your app will be in their comfort zone, as far as their thinking is concerned.
It’s one thing to have a market or target audience. That’s something every app should have – otherwise why are you making it? However, things get tricky when you don’t know how to reach your target audience. For example, the idea behind adore.ly could go viral in high-school or in more conservative countries (such as India), but we weren’t able to figure out exactly how to target those people. Most high-schoolers dont read Hacker News or TechCrunch or tech blogs in general. They are on Facebook and Twitter, but our friend lists didn’t have this age group. I think luck is important in succeeding here.
We did really well here. We had two coders, me and Victor, and Fei Jing was in charge of business and promotions Victor is a backend guru and coded up Python super-fast, while I worked on the entire front-end. Having specific roles for each person really helped, as we all knew exactly what to do. At the same time, we were small enough that we could take decisions really quickly.