Follow Airbnb users from a US city over the course of a year as they interact with the platform for different lengths of time and from different devices.
The purpose of this visualization is to identify patterns in the behavior of users accessing the Airbnb platform. Airbnb open-sourced a one year chunk of user data from a US city for a visualization challange at databits.io. It was a very detailed dataset, so the challenge was to identify and display only a few key patterns - I think the temptation is usually to show too much.
One thing that immediately stood out was the lull in activity between the hours of 14 and 18. You would expect a drop in activity overnight, not in the middle of the afternoon, so it may be that the time stamps in the dataset were shifted. Also interesting that the top 5 users, who accessed the platform >100 times, never booked.
A few other observations (that would need deeper analysis for confirmation):