The document discusses recommendations to increase revenue for Airbnb based on an analysis of their New York property listings data. It finds that Manhattan and Brooklyn make up 85% of listings but recommends targeting Queens, Bronx and Staten Island for acquiring more properties. It also recommends focusing on listings with a price range up to $200 and minimum stays between 4-30 nights. Unpopular properties could be promoted through incentives like free airport transportation and breakfast.
1. High Bookings at Low Price and
Least Minimum Stay
By:
Harshad Surya Chandolu
2. Business Understanding
• For the past few months, Airbnb has seen a major decline in revenue.
• Now the restrictions, due to covid -19, have started lifting.
• People have begun to travel more and Airbnb wants to make sure that it is
fully prepared for the upcoming business opportunities.
3. •Improvise theAirbnb business to generate more revenue.
•To build the properties according to the customer
preferences.
•Raising the number of bookings for the unpopular
properties with added incentives.
Business Objectives
4. Contents
Domain Understanding:
The Head of Acquisitions and Operations, NYC and Head of User Experience, NYC at Airbnb want to
understand some important insights based on various attributes in the dataset so as to increase the revenue
such as -
1. Which type of hosts to acquire more and where?
2. The categorization of customers based on their preferences such as: Location, Prices and so on.
3. What are the most popular localities and properties in New York currently?
4. How to get unpopular properties more traction?
Data and Business
Understanding
Domain
Understanding
Recommendations Insights Conclusion and
appendices
5. Recommendations
1. The properties listed at Manhattan and Brooklyn which contributes about
approx. 85% of New York properties while in Queens, Bronx and Staten Island
have least number of properties. So, we must target Queens, Bronx and Staten
Island for acquisition of more properties.
2. As Manhattan has higher no. of Entire home / apartments and Queens has
higher no. of Private rooms, So we must look out for acquisition of more Private
rooms in Manhattan and more Entire Home /Apartments in Queens. Customers
mostly prefer Entire Home/Apartments and Private rooms.
3. Customers mostly prefer properties with “Budget price” which is ranging up to
$200 and they rarely go for properties which has price range exceeding $400.
6. Recommendations
1. About 98.86% customers mostly opt for below and up to a month worth of
minimum stay. We must keep the number of minimum nights stay from 4
nights – 30 nights.
2. Queens, Staten Island and Bronx properties need to have minimum stay of
3 nights.
3. Manhattan and Brooklyn are mostly unaffected by the least minimum stay
of 11 but preferably it must lie between 3-7 nights.
7. Recommendations to bring unpopular
properties more traction
1. We can offer free pick up and drop facility from airport.
2. We can provide complementary Breakfast.
3. We can provide membership discounts atAirbnb and encourage more
people to apply for those unpopular properties.
4. We can provide discounts by collaborating with the local recreational
facilities like indoor games or swimming pool or Snooker or Library and
so on.
5. Occasional minimum stays of 30 nights/ 15 nights could be introduced
during festive seasons like Christmas and so on.
8. Acquisition Properties at Queens, Bronx and Staten Island
Observations:
1. The properties listed at
Manhattan is 44.3% and
Brooklyn is 41.1% which
contributes about approx.
85.4% of New York properties
2. While in Queens, Bronx and
Staten Island have least
number of properties.
3. We must target Queens, Bronx
and Staten Island for
acquisition of more properties.
9. Customer prefer price range up to $200
Observation:
1. Customers mostly prefer Budget price which ranges up to $200.
2. Customers highly refuse to go for price ranges exceeding $400.
3. Customers still do not find “reasonable” price range of $200 -$400 as reasonable, so we need to reduce the
price range up to $200.
10. Customers preferences based on minimum nights of stay
Observation:
1. About 98.86%
customers mostly
opt for below and
up to a month worth
of minimum stay.
2. Customers rarely go
for minimum stay
above a month.
3. We must keep the
number of minimum
nights stay from 4
days - 30 days.
11. Customer Preferences based on Location and type of the room
Observation:
1. Overall Room Type percentages:
Entire home/apt = 51.96%
Private room
Shared room
= 45.66%
= 2.37%
2. Major difference in contribution of
Overall Room Type and Within
Neighbor Group (based on delta
percentage calculation):
a) Manhattan has 14.7% higher
contribution of Entire home/apt
compared to the Overall
contribution of Entire home/apt.
b) Queens has 23.2% higher
contribution of Private room
compared to the Overall
contribution of Private room
12. Locations affected by high minimum night stay
Observation:
1. Queens, Staten Island and Bronx properties need to have minimum stay of 3 nights.
2. Manhattan and Brooklyn are mostly unaffected by the least minimum stay of 11 but preferably it must lie
between 3-7 nights.
3. Occasional minimum stays of 30 nights/ 15 nights could be introduced during festive seasons.
13. Density of properties in each location
Observation:
1. Staten Island is scarcely populated with properties.
2. Manhattan, Brooklyn, Queens and Staten Island are denser than Staten Island.
3. All the location show more than 95% of the properties are either private rooms or Entire Home/
Appartments.
14. Customers preferences of location based on availability of the properties
Observations:
1. We see that customers prefer
Brooklyn over Manhattan
due to reasonable prices.
2. Customers only go for
Queens when both Brooklyn
and Manhattan are not
available or are expensive.
3. Customers prefer to go for
Staten Island and Bronx only
in later days of the year.
15. Conclusion
1. The properties listed at Manhattan and Brooklyn which contributes about
approx. 85% of New York properties while in Queens, Bronx and Staten Island
have least number of properties. So, we must target Queens, Bronx and Staten
Island for acquisition of more properties.
2. As Manhattan has higher no. of Entire home / apartments and Queens has
higher no. of Private rooms, So we must look out for acquisition of more Private
rooms in Manhattan and more Entire Home / Apartments in Queens. Customers
mostly prefer Entire Home/Apartments and Private rooms.
3. Customers mostly prefer properties with “Budget price” which is ranging up to
$200 and they rarely go for properties which has price range exceeding $400.
16. Conclusion
1. About 98.86% customers mostly opt for below and up to a month worth of
minimum stay. We must keep the number of minimum nights stay from 4
nights – 30 nights.
2. Queens, Staten Island and Bronx properties need to have minimum stay of
3 nights.
3. Manhattan and Brooklyn are mostly unaffected by the least minimum stay
of 11 but preferably it must lie between 3-7 nights.
All these driving factors will lead to increase in revenue for theAirbnb
17. AppendixA: Data Understanding
Data Sets provided are:
• AB_NYC_2019.csv: This dataset contains the details
regarding all the properties in New York which areAirbnb. It
also contains the customer reviews.
Shape of Data Sets:
• AB_NYC_2019.csv has 48895 rows and 16 columns.
18. Appendix B: Data Cleaning
For AB_NYC_2019.csv:
• Finding the percentage of missing values in the columns and
removing the columns with missing values.
• Checking out outliers for the numerical columns
('price','minimum_nights','number_of_reviews','reviews_per_month',
'calculated_host_listings_count','availability_365’) and treating
them.
• Binning of continuous variables for segmented analysis like price
and minimum_nights.
19. 1. Checked for outliers for all numerical variables.
2. Performed univariate analysis on categorical variables and numerical
variables.
3. Checked correlation for all numerical variables if any.
4. Performed bivariate analysis on numerical variables and categorical
variables.
Appendix C: Analysis of AB_NYC_2019.csv data set