This case study focused on using Tableau with an AirBnB case study to create visualizations and a dashboard which our stakeholders can use to make data driven decisions. For this case study, the stakeholders are interested in making some investments and buying their own AirBnB's in the Seattle, Washington area, and would like to know what houses are the most profitable and what contributes to their value, whether that be number of bedrooms or location.
The dataset itself is a sample AirBnB listings (Kaggle link to dataset) which covers a wide range of AirBnB data in the form of three spreadsheets. Those spreadsheets are Listing, Reviews, and Calendar; for this analysis, I will only be using the Listing and Calendar spreadsheets, since I believe I will find all of the data I need for my analysis in these spreadsheets. To begin, I joined the spreadsheets using the ID column. Below, you can see the data in Excel before I imported it into Tableau for my analysis.
The Data At a Glance
The Listing spreadsheet contains the majority of the data here. There are 92 columns in this spreadsheet, compared to the 4 columns of the Calendar spreadsheet.
There is a lot of data here that I simply wouldn't use, although I didn't do it for this project, I would have recommended deleting this data from my working dataset, since things like Description, Listing Notes, and Neighborhood Overview would have been pretty useless to me in my analysis.
Getting Started in Tableau
Finding Price per Zipcode
The first visualization I wanted to create was a comparison of the Price per Zipcode for the AirBnBs.
Here we get a nice visualization with each column color coded by zipcode. Based off this graph, my suggestion to the stakeholder would be to purchase property to turn into an AirBnB in the 98134 zipcode, since the Average Price per night for an AirBnB in this zipcode is considerably higher.
Map of Each Zipcode
This is my favorite visualization for this project. Since we are working with zipcodes for these AirBnBs, I wanted to create a map that will go hand in hand with the previous visualization that also shows the average price under the zipcode.
Using this map, you can see the where the most expensive zipcode is, and also see some of the clusters and the pattern of where the top zipcodes are on the map.
Finding the best weeks to list the AirBnB
This visualization shows what the average price of AirBnBs are throughout a year.
This information is useful if the stakeholder is planning on living in this house some of the time, and posting it when it would be most profitable.
This information might seem pretty obvious, but it is still good to know that the data backs up the fact that AirBnBs are more profittable during the holidays and summer time.
How will the number of bedrooms affect the average price?
You can see the average price steadily increases as the number of bedrooms increase, which is another piece of information that may sound obvious, but once again is something that is beneficial to have confirmed by the actual data.
You can also see that 5 and 6 bedroom AirBnBs seem pretty lucrative, as the average price starts to increase more for these places.
My recommendation for the stakeholders would be to try and focus on getting place with more bedrooms if they are interested in making as much money as they can.
Finding the count of listings for different number of bedrooms
This visualization goes with the previous one, since the next natural question would be to know how many of each listing there would be.
You can see here that there are much fewer AirBnBs in Seattle, Washington with 5 or 6 bedrooms. This also contributes to my recommendation of buying a 5 or 6 bedroom listing, since you won't have as much competition with these AirBnBs.
There is also an abundance of one bedroom AirBnBs in Seattle, Washington, so shying away from these would also be a good idea.
Creating the Dashboard
Thank you for viewing my Tableau project! This dataset was taken from a public repository on Kaggle at https://www.kaggle.com/datasets/alexanderfreberg/airbnb-listings-2016-dataset