Maximizing Airbnb income in Seattle
Airbnb is a platform that has been growing tremendously in the last couple of years. At least before the COVID-19 pandemic, travelling was more popular than ever, especially with the rise of Instagram bloggers that travel the world and share their impressions.
Therefore, many people see Airbnb as an opportunity to earn money. Maybe they are parents whose kids moved out already, which means they have some rooms to spare. Others might invest into real estate and use Airbnb as a platform to find potential tenants.
For those people, of course, one of the most interesting questions would be how they can maximize their return on investment. To assist our decision making with empirical data, I analyzed 3818 Airbnb listings from the Kaggle Seattle Airbnb dataset.
Disclaimer: This post was created within the scope of the first project of the Udacity Data Scientist Nanodegree programme. Please take the results with a grain of salt and do not base your investment decisions on it ;)
For this analysis I used Python in Jupyter Notebooks to predict prices with a linear model. The code can be found on GitHub.
Question 1: Which neighborhood is most expensive?
High prices could be a result of a high demand. If we have the chance to rent out an apartment in one of the most popular neighborhoods, we might increase our income. Therefore, I first looked at the average prices in each neighborhood.
We can see that prices are highest in Magnolia, Queen Anne, Downtown and Cascade. While the bar chart only shows the average prices, it doesn’t take into account other factors that might influence those prices. Therefore I also looked at the neighborhood-related coefficients of the linear model.
Looking at the top 4 coefficients, we can see that the same neighborhoods that have the highest prices on average also have the highest positive impact on price. Interestingly, the order is different, though. It looks like there might be other factors than the location itself that drive prices up in Queen Anne and Magnlolia, which rank #1 and #2 on the average price but only #3 and #4 when it comes to the coefficients.
Question 2 — Which factors influence prices the most?
While answering question 1, we have seen that other factors apart from the neighborhood seem to influence prices. I checked the coefficients of the linear model.
Top 10 factors increasing the price
While 3 of 4 neighborhoods we identified as most expensive are in the list of top 10 factors positively influencing price, we can clearly see that other factors play a role as well.
The property type appears 4 times in the top 10 list. Exclusive and fancy property types like boats, treehouses or chalets have a very strong impact on prices. Boats and treehouses raise the price by 60–75 dollars. This makes sense as I could imagine that people who love to travel want to have a unique experience and are therefore willing to pay more for such unusual housing types.
The number of bedrooms and bathrooms also have a strong positive correlation with price. This makes sense as an increasing number of bedrooms and bathrooms mean that the property can accomodate more people. As you don’t pay per person but usually rent the whole place the price can be shared amongs more people. We would probabaly see a similar correlation when it comes to the size of the apartment, but unfortunately those values were not available for most listings.
Top 10 factors decreasing the price
Now looking at the most negative influencing factors, we see neighborhoods are again present here. Delridge, Rainer Valley, Beacon Hill and Northgate are also among the lowest 5 when it comes to average prices.
If your propoerty type is a dorm, prices will be around 251 dollars lower, according to the model. This aligns with the second strongest negative influence: whether the room is shared or not.
The same trend can be seen when looking at number three in terms of negative influence. A private room also seems to lower the price a potential landlord could ask for. While private room sounds like the opposite of a shared room, it actually still means you could share other rooms and amenities with other people. This is the definition of a private room, according to Airbnb:
”Private rooms are great when you prefer a little privacy, yet still value a local connection. When you book a private room, you’ll have your own private room for sleeping and may share some spaces with others. You might need to walk through indoor spaces that another host or guest may occupy to get to your room.”
Question 3 — Big loft or multiple private rooms with shared bathroom and kitchen?
A friend has an apartment in Seattle downtown that he wants to rent out via Airbnb. It has one big bedroom, a kitchen and one bathroom. Should he invest money to split the bedroom into two rooms so he would be able to accommodate two different guests?
I will look at the following two options and predict the corresponding price with the linear model I built:
- Rent room as a loft for one person
- Split rooms and rent two private rooms that share bathroom and kitchen. Since my friend currently only has one bed, one room is only equipped with a pull-out sofa.
Results
- Option 1 yields $158.08 dollars per night.
- Option 2 yields $102.14 for the room with the bed and $93.31 for the room with the pull-out sofa. This makes $195.45 per night if both rooms are rented out.
That makes a difference of $37.40 per night. Of course, this would also mean more work as two different guests need to be accommodated. There is also the risk of not being fully booked.
Potential follow-up analysis
In a follow-up analysis it could make sense to also look at the amenities. I omitted this data in my analysis, but we could check which amenities lead to the highest return on investment. Maybe this could help our friend to maximize earnings without having to split the room.
