Hotel Booking Analysis
The presence of advanced technology makes booking hotels seem easier and more accessible to everyone. Not only can they book, but they can also choose the place, date, number of people, how long they are going to stay, or even reschedule or push the date booking through an app or website. Thus, to stay ahead in this competitive industry, deep analysis is required.
Dashboard can be accessed in this link
Project Overview:
This project aims to provide a thorough analysis of Hotel Booking from a sales and customer behavior. Using a detailed dataset, the project explores various facets of the business, including sales trend, product popularity, and more.
Requirements:
Get Yeni W’s stories in your inbox
Join Medium for free to get updates from this writer.
On this project, the hotel booking dataset is collected from Kaggle, in CSV format. Analysis was performed using Tableau for creating interactive dashboards. Below are the columns:
- hotel: Indicates the type of hotel (resort or city).
- is_canceled: Specifies whether the booking was canceled or not (0=not canceled, 1=canceled).
- lead_time: Represents the number of days between the booking date and the arrival date.
- arrival_date_year: Denotes the year of the arrival date.
- arrival_date_month: Indicates the month of the arrival date.
- arrival_date_week_number: Specifies the week number in which guests arrived at the hotel
- arrival_date_day_of_month: Represents a specific day of arrival within a month
- stays_in_weekend_nights: Indicates how many nights (Saturday or Sunday) guests stayed or booked to stay at a hotel during weekends
- stays_in_week_nights: Represents how many weeknights (Monday to Friday) guests stayed or booked to stay at a hotel during weekdays
- adults: Indicates the number of adults included in each booking
- children: Indicates the number of children included in each booking
- babies: Indicates the number of babies included in each booking
- meal: Describes what type of meal was booked (Breakfast only, Half board, Full board, or Undefined/SC — no meal package)
- country: Denotes the country-of-origin for each guest who made a reservation
- market_segment: Shows various market segments that individuals belong to when making reservations (e.g., Online Travel Agents, Offline Travel Agents, Corporate clients)
- distribution_channel: Specifies different channels through which bookings were made (e.g., online travel agencies, direct bookings with hotels/tour operators, corporate arrangements)
- is_repeated_guest: Indicates whether the guest is a repeated visitor (0=not repeated guest, 1=repeated guest)
- previous_cancellations: Represents the number of times guests previously canceled their bookings
- previous_bookings_not_canceled: Denotes the count of previous bookings made by guests that were not canceled
- reserved_room_type: Identifies the type of room initially reserved
- assigned_room_type: Identifies the type of room that was assigned to guests
- booking_changes: Represents the number of changes made to the booking
- deposit_type: Indicates the type of deposit made for the booking
- agent: Represents the ID of the travel agency that made the booking
- company: Represents the ID of the company that made the booking
- days_in_waiting_list: Represents the number of days the booking was on the waiting list before being confirmed
- customer_type: Indicates the type of customer (e.g., transient, contract, group, or other)
- adr: Represents the average daily rate (price per room) for the booking
- required_car_parking_spaces: Indicates the number of car parking spaces required by the guest
- total_of_special_requests: Represents the total number of special requests made by the guest (e.g., extra bed, room amenities)
- reservation_status: Indicates the status of the reservation (e.g., canceled, checked-in, no-show)
- reservation_status_date: Represents the date on which the reservation status was last updated
Business problems that can be answered using dataset:
- What was the best month for sales? How much was earned that month?
- What was the preferred hotel of most visitors? Did the season affect the choices?
- How did having no or having kids around affect the type of hotel booking preferences?
- How long in an average the customer would stay? Did they have additional requests during the stay?
- How many visitors per booking?
- What was the most preferable channel of hotel booking?
- What was the most favorite room?
- How diverse the visitors are ? The top 10 countries where the most visitors come from.
- Did the visitors book the hotel with a meal as well? What type of meal?
Analysis
- What was the best month for sales? How much was earned that month?
Insight:
August ’16 emerged the highest sales generating $1,809,324.75 with 3.238 bookings. In contrast, October ’16 recorded as the highest total bookings with 3.689 yet only generating $1,072,101.84. This could happen due to the high season, which caused the price also had a raise. August is warmth month before approaching the winter and school holiday season; therefore, family tends to go out to have a holiday or trip together.
2. What was the preferred hotel of most visitors? Did the season affect the choices?
Insight:
Overall, staying in a hotel had always been preferred over a resort. Compared to the trend of hotel, resort reservation in August had a slight increase with 1.107 bookings. October was the peak reservation for the resort, while the hotel had a slight decline from the previous month.
3. How did having no or having kids around affect the type of hotel booking preferences?
Insight:
The difference in choices between with or without kids could be seen in the diagram, where without-kids preferred to stay in city hotels, while with kids only had a slight difference in choices between city and resort.
4. How long in an average the customer would stay? Did they have additional requests during the stay?
Insight:
In June and August, a customer who was still single would take a 4-day stay, while the customer who brought kids preferred to have a 3.5-day stay as the longest. The second graph showed that families were the most frequent customers who made special requests during the stay compared to customers without kids. It was reasonable since the family may require more beds, meals, etc. In February, 52.5% customer-with-kids made special request from its total booking whereas the customer-without-kids only 26% of its total booking called for special request. It means marketing team could make a special program for with-kids customers so they are able to stay longer.
5. How many visitors per booking?
Insight:
In 2016, customers were categorized as customer-with-kids and customer-without-kids. On average, most customer-with-kids consisted of three people, including their kids; it could be a mother, father, and child. On the other hand, customer-without-kids consisted of two people or a couple in bookings, with 23.704 bookings. This means, sales and marketing team could make a campaign program targeting these customers with put specific criteria on total person booking for example.
6. What was the most preferable channel of hotel booking?
Insight:
In 2016, customer were comfortable to book the hotel through Travel Agent or Tour Operator which generated 9,42 M. Data in Market Segment also said 71.40% of the customer were booking from Online/Offline Travel Agent as well. It means, marketing team could target these travel agencies as all customers tend to reach out to them.
7. What was the most favorite room?
Insight:
In 2016, more than half of total visitors were booking on this type of room. This room was very popular with 25.403 total bookings. This means, hotel might consider to make more A type of room so more visitor would come and stay.
8. How diverse the visitors are ? The top 10 countries where the most visitors come from.
Insight:
In 2016, 17.810 people from Portugal became the guest of the hotel or resort. On the following, there were +/- 9K people from UK and France visited the hotel & resort.
9. Did the visitors book the hotel with a meal as well? What type of meal?
Insight:
In 2016, >85% of the visitors booked the meal as well during their stay. Most of them preferred to have a breakfast at the hotel. It was a good sign for the hotel especially for the sales and marketing team to make a campaign program for it. A full-board meal was the least popular option booked by guests. Meanwhile, 10% of the visitors were not interested at all in having a meal made by the hotel.
Note:
This project may really out of perfection since I just embarked my journey on #DataViz and #DataAnalysis, therefore would be happy if we can share some feedback and connect each other. Find me in Twitter or Linkedin
