Say hello to Michelle’s List. Michelle’s List is a website for renters who want to know more about their landlord before moving in to a new rental.

The Story

One day a couple of weeks ago my sister sent me a link to OpenIgloo, and said, “Could you build this, but for everywhere?” I’d been looking for a project to work on, so I said, “Sure.”

I started looking into what OpenIgloo is all about. It turns out that OpenIgloo is one of only a handful of websites that help renters look out for problem landlords. The problem with OpenIgloo is that it only works in a few large markets. For my sister who is renting in a small town, it wouldn’t work. Another site, ratethelandlord.org, is available all over the US, and actually a couple of different countries, but you can’t search by the address of the property. You can only search by general location or the landlord’s name.

One other reason this idea attracted me is that I had been noodling with my business partner, Alex, on how to fix reviews. We both believe that what are used as reviews these days, a general 1 to 5 rating, is not a good way to help users differentiate between options. Instead, it would be more helpful to use natural language processing to parse out the reviews for sentiment analysis and key phrases. That way, users still get the Good/Neutral/Bad quick review, while also getting more nuance about what makes the good or service unique. Amazon appears to be the only place doing this style of review.

Getting Started on Development

Coming off the back of the now defunct eccentric.health work, I knew I could build a front-end fairly quickly using Next.js and Vercel. The question was what kind of back-end would I need. Could I use Airtable, like I had last time? Did I need to build my own back-end server, with a set of APIs, and my own database?

The answer to both questions turned out to be, “Yes”. Taking more of what I learned from eccentric.health, I knew that iterating quickly was key to keeping my excitement for the project high. So in order to build an MVP quickly I could use Airtable to stand up a database quickly, and once the front-end was working, transition to a proper Node.js server with RESTful API architecture and a SQL database.

Future State

As already mentioned, I want to build out a proper back-end server, but that’s only necessary if I have more than a couple thousand rows of data. So how am I planning on getting to more than a couple thousand reviews? As my business partner put it, “The issue is the step of reading reviews in the user journey is not 1st… in order to read the reviews, you gotta leave them first.” This means users are not going to arrive just because I have reviews. I need to get them in some other way. Right now, the plan is to start showing listings for open rentals alongside reviews. This provides a potential revenue stream in addition to getting users to add their reviews. Then as time goes on, and I get more reviews on the site, I can work on creating the type of review system that I think is better than just a 5 star rating system.