Wayfit

Inspiration

Wayfit is intended to address the customer concern “How can I buy this sofa when I’ve never sat on it? What if it’s not comfortable?”

What it does

Wayfit will produce personalized comfort predictions that will be shown to customers as they browse furniture products.

How we built it

First we needed data. We worked with facilities to setup Waystation - an area on the 7th floor with twelve chairs. We asked fellow Wayfarians to sit in all the chairs and rate them for comfort.
We took detailed measurements of the 12 chairs
We conducted a data science experiment to see if we could accurately predict comfort based on the data we collected.

Challenges we ran into

Since we knew almost nothing about chair ergonomics, we had to get educated quickly.

Accomplishments that we're proud of

Working together as a team, we were able to execute on a big idea and get positive results.

What we learned

It's possible to generate personalized comfort predictions!

Wayfarians like donuts and will do anything to obtain them!

What's next for Wayfit

We think Wayfit is a simple, compelling idea that could have a large impact on Wayfair's business and strategic position in the market.

We feel Wayfair should move forward with Wayfit.

We propose to develop a project plan to implement Wayfit. The first phase would be to conduct a larger more comprehensive experiment, to refine the data science and improve predictive accuracy. This experiment could be done on premises at 4 Copley, with Wayfair employees supplying the data.

Built With

Created by

I started thinking about this idea when I first started at Wayfair, and it eventually morphed into this challenge submission! My role was primarily to kick-start things and put this awesome team together. I jumped in on the data science and presentation.

Andy Ullman
I worked on gathering data (including survey building), collaborated on creating the presentation, refining value proposition and providing the voice over. I also had a minor role in the data wrangling component of the project and provided feedback that was ultimately implemented in the visualization of the feature on the website.

steven Jin
When Andy told me about his idea I just knew this is going to be fun and exciting. This was mostly team work, My specific role was writing code for our machine learning models and their evaluation (Andy had very useful feedback here) and I also participated in other parts including selecting the chairs for our experiment,
taking pictures for the survey, collecting data from our participants, measuring chair's characteristics.

Ali Bagheri
Patrick Clark