Your Scene

A look into the challenges of B2C

Customer Discovery

• Intercepting folks at the pain point we'd like to alleviate
• Craigslist for rent/real estate sections for various cities 
• SubReddit's for cities, e.g. reddit.com/r/atlanta
  • People are looking for city specific information here for various reasons
  • We knew for sure many times people come here to get a primer on the city

Customer Discovery

• So we'd post something like this: 

 

 

• Which would take people to a basic landing page: 

 

Results

From the onset the results were pretty positive:

• We drove about 30-40 hits over the span of 2 or 3 posts
• email conversion rate: ~1%

So We Built

• This was enough of an indication that we should build something functional
• So we got the first iteration of our app off the ground:
• Using a basic walk score metric to match up cities and providing users value

AB testing

We found that we had better conversion rates with a different title, and modified wording, shown below:

AB testing

We also optimized driving traffic and engagement

• Subreddits worked better than craigslist, with posts that seem less bot-like, e.g. : 

• We were able to drive a decent amount of traffic, with about a 3.5% conversion rate over the remaining duration of the project

Adding Value

So far what we had was good, but we needed to provide the user with just a little bit more... ahem, queue live demo please

Potential future goals

1. Improve the algorithm so that it is surgically accurate in finding the perfect neighborhood for our users
2. Visualize the data so users have something to show their friends
3. Improve traffic driving methods
4. Devise monetization techniques