You're in a new neighborhood and you're looking for the right place to go. Your usual recommendation app is context-agnostic and actually gets worse at what it's supposed to do the more popular it becomes: it overwhelms you with a flood of information, making your choice harder, rather than easier.
What it does
Momentum gives you the right thing to do, right here, right now - and the drive to do it. It pulls your time and location, and tells you where you should eat/drink/explore/party. Recommendations come from other users who did the same thing you want to do (eat), at the same time as you (Saturday morning), in the same neighborhood (Red Hook). And the more recent a recommendation, the more momentum it conveys on a place.
Users gain and loose momentum, too. Increasing your momentum by giving recommendations moves you up through the ranks, from novice to local - and the higher the peer group you rank in, the more curated the recommendations you receive become.
How we built it
Our team collaborated using Slack and GitHub. We built the backend using Rails and Postgres and the frontend using HTML5, React.js, and Bootstrap. We fetch needed data from the Google Places API. Momentum is deployed on IBM Bluemix.
Challenges we ran into
- Creating the decay function describing how Momentum is lost over time (we decided to add a smoothing function)
- Getting the wording right to convey the purpose of the app
- Minimizing the interaction steps for better user experience
Accomplishments that we're proud of
- Going from nothing to a usable app with a simple, slick user interface in 24 hours
- Getting everything running in docker containers on IBM Bluemix
What we learned
New technologies like React.js or IBM Bluemix. The best thing to do right here, right now!
What's next for Momentum
- Improving the user peer-group allocation model (novice, regular, local)
- Turning the web app into native apps for Android and iOS (Cordova, React native...)
- Testing the app in the market
- Adding initial data for more places so that Momentum becomes attractive to new users
- Celebrating our huge success :-)