The data were sourced from an article published in the Journal of Data in Brief by ELSEVIER, authored by Nuno et al. (2019), and available on ScienceDirect. It encompasses a total of 119,390 booking transactions from two hotels: an unnamed resort hotel in the Algarve, Portugal, with 40,060 observations (H1), and a city hotel in Lisbon, Portugal, with 79,330 observations (H2). Both datasets share the same structure. The dataset includes bookings scheduled to arrive between July 1, 2015, and August 31, 2017, covering both completed stays and cancellations.

hotel_bookings <- read.csv("hotel_bookings.csv")

Setting up the environment for analysis.

library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.4.4     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.1
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(ggplot2)
library(dplyr)
library(lubridate)
library(ggthemes)
library(forecast)
## Registered S3 method overwritten by 'quantmod':
##   method            from
##   as.zoo.data.frame zoo
library(tseries)
library(padr)
library(rsample)
library(scales)
## 
## Attaching package: 'scales'
## 
## The following object is masked from 'package:purrr':
## 
##     discard
## 
## The following object is masked from 'package:readr':
## 
##     col_factor
library(recipes)
## 
## Attaching package: 'recipes'
## 
## The following object is masked from 'package:stringr':
## 
##     fixed
## 
## The following object is masked from 'package:stats':
## 
##     step
View(hotel_bookings)
hotel_bookings %>% 
  count(hotel, market_segment, is_canceled) %>% 
  group_by(hotel) %>% 
  mutate(total = sum(n),
         ratio = n/total,
         is_canceled = ifelse(is_canceled == 0, "No", "Yes")) %>% 
  ungroup() %>% 
  mutate(market_segment = tidytext::reorder_within(market_segment, ratio, hotel)) %>%  
  ggplot(aes(ratio, market_segment, fill = is_canceled)) +
  geom_col(position = "dodge") +
  labs(x = "Percentage", y = "Market Segment", 
       title = "Hotel Demand by Market Segment",
       fill = "Booking Cancelled") +
  facet_wrap(~hotel, scales = "free_y") +
  scale_x_continuous(labels = percent_format(accuracy = 1)) +
  scale_fill_manual(values = c("darkgreen", "darkred")) +
  tidytext::scale_y_reordered() +
  theme_pander() +
  theme(legend.position = "top")

Based on the results, majority of bookings are made through travel agents (TA), both online and offline, which together account for more than 40% of the total non-cancelled transactions. The other segments do not have as many transactions, but it might be beneficial to examine the revenue generated by each market segment by looking at the Average Daily Rate (ADR).

hotel_bookings %>% 
  filter(is_canceled == F) %>% 
  group_by(hotel, market_segment) %>% 
  summarise(adr = sum(adr)) %>% 
  mutate(total = sum(adr),
         ratio = adr/total) %>% 
  ungroup() %>% 
  mutate(market_segment = tidytext::reorder_within(market_segment, ratio, hotel)) %>% 
  ggplot(aes(ratio, market_segment)) +
  geom_col(fill = "blue") +
  facet_wrap(~hotel, scales = "free_y") +
  tidytext::scale_y_reordered() +
  scale_x_continuous(labels = percent_format(accuracy = 1)) +
  theme_pander() +
  labs(x = NULL, y = NULL, 
       title = "ADR Contributions by Market Segments")
## `summarise()` has grouped output by 'hotel'. You can override using the
## `.groups` argument.

The Average Daily Rate (ADR) is calculated by dividing the sum of all lodging transactions by the total number of nights stayed. Thus, a higher ADR signifies more revenue generated per night. The total (ADR) generated from online travel agents is the highest for both city hotels and resort hotels. The segments of offline travel agents/tour operators and direct bookings have a narrow margin between them, with the direct booking segment contributing more to resort hotels, despite having a lower number of transactions. This identifies the leading three market segments, both in quantity and profitability.

ggplot(data = hotel_bookings, aes(x = lead_time, y = children)) +
  geom_point(aes(color = children)) +  # Color by 'children'
  scale_color_gradient(low = "skyblue", high = "darkblue") +  # Gradient color
  labs(title = "Relationship between Booking Lead Time and Guests Traveling with Children",
       x = "Booking Lead Time",
       y = "Number of Children",
       color = "Children") +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5))  # Center the plot title
## Warning: Removed 4 rows containing missing values (`geom_point()`).

Explored the relationship between booking lead time and guests travelling with children.The plot shows a group of guests who typically make early bookings, and this plot showed that many of these guests do not have children.

If the marketing department aims to launch a family-friendly promotion targeting key market segments. They may seek to identify which segments generate the highest number of bookings and determine the preferred locations of these bookings, whether in city hotels or resort hotels.

A bar chart may be useful to display each hotel type alongside the market segments, utilizing distinct colors to represent each segment.

ggplot(data = hotel_bookings) + geom_bar(mapping = aes(x = hotel, fill = market_segment)) +
  labs(title="Hotel Type and Market Segment")

After creating the bar chart, it becomes apparent that comparing the size of the market segments at the top of the bars may be challenging. Furthermore, a clearer comparison of each segment would be beneficial. Consequently, the decision is made to create a separate plot for each market segment to facilitate a more detailed analysis.

ggplot(data = hotel_bookings) + 
  geom_bar(mapping = aes(x = hotel), fill = "steelblue") +  # Adding color to the bars
  facet_wrap(~market_segment) + 
  labs(title = "Comparison of the Size of Each Market Segment") +
  theme_minimal() +  # Using a minimal theme for a cleaner look
  theme(plot.title = element_text(hjust = 0.5))  # Centering the plot title

Now, that a separate bar chart for each market segment. We can have a clearer idea of the size of each market segment, as well as the corresponding data for each hotel type.

After considering all the data, the stakeholders may decide to send the promotion to families that make online bookings for city hotels. The online segment is the fastest growing segment, and families tend to spend more at city hotels than other types of guests.

In addition, it would be better to visualize a plot that shows the relationship between lead time and guests travelling with children for online bookings at city hotels. This will give a better idea of the specific timing for the promotion.

Moreover, we may break it down into the following two steps:

1) filtering the data;

2) plotting the filtered data.

onlineta_city_hotels <- filter(hotel_bookings, (hotel=="City Hotel" & hotel_bookings$market_segment=="Online TA"))

View the assigned data

View(onlineta_city_hotels)
ggplot(data = onlineta_city_hotels, aes(x = lead_time, y = children, color = as.factor(is_canceled))) +
  geom_point() + # Keep the original point aesthetics
  labs(title = "Online Bookings for City Hotels",
       x = "Lead Time",
       y = "Number of Children",
       color = "Booking Cancelled") +
  scale_color_manual(values = c("0" = "deepskyblue", "1" = "coral"),
                     labels = c("No", "Yes")) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5)) +
  guides(color = guide_legend(override.aes = list(shape = 15, size = 6))) # Use square shape for legend keys, adjust size as needed
## Warning: Removed 1 rows containing missing values (`geom_point()`).

Based on the filter applied, the scatter plot displays data for online bookings at city hotels. This visualization indicates that bookings with three children showing a notably shorter lead time (less than 200 days). On the other hand, bookings with two children and below have longer lead time. Therefore, promotions targeting families with three children can be strategically scheduled closer to the relevant booking dates for greater effectiveness.

(The filtered data will enable the creation of varied views and facilitate a deeper investigation into more specific relationships within the dataset.)

Recommendations

If the booking lead time is longer—meaning people are planning their trips well in advance—there are several marketing strategic actions it can take to make the most of this situation;

1. Early Bird Discounts

Offer discounts or special packages to guests who book their stay several months ahead of time. This encourages guests to lock in their reservations early, securing revenue for the company ahead of time.

2. Flexible Cancellation Policies

To attract early bookings, companies can offer more flexible cancellation policies. Knowing they can cancel without penalty makes customers more likely to book early.

3. Personalized Marketing

Use the lead time to engage with customers through personalized marketing. This could include sending them information about the destination, tips for their upcoming stay, or offers on upgrades and additional services.

4. Revenue Management

Longer lead times give companies more data to optimize their pricing strategies. By analyzing booking patterns, they can adjust prices to maximize revenue, perhaps increasing prices as the arrival date gets closer and demand goes up.

5. Improved Resource Planning

With a good idea of future occupancy rates, a company can better plan its resources, including staffing, inventory, and maintenance work. This helps in providing a better guest experience and managing costs efficiently.

6. Loyalty Programs

Encourage early bookings by offering points or rewards through a loyalty program. Guests could earn more points for booking early, which they can redeem for perks like free nights, room upgrades, or other services.

7. Special Experiences

Offer exclusive experiences or access to special events for guests who book early. This could be a dinner at a top restaurant, tickets to local attractions, or a unique tour that’s only available to early bookers.

8. Forecasting and Strategy Adjustments

Use the data from booking patterns to improve demand forecasting and strategic planning. Understanding why lead times are longer can help a company adjust its marketing and operational strategies to meet customer needs better.

Adapting to longer booking lead times is all about understanding customer behavior and preferences, then aligning your business strategies to meet those expectations while optimizing your revenue and operational efficiency.

Shorter booking lead time

If it faces shorter booking lead times—meaning people are making their plans closer to their travel date—it suggests a shift towards more last-minute bookings. It may adapt to several strategies;

1. Last-Minute Deals

Offer attractive last-minute deals to encourage spur-of-the-moment bookings. This can help fill up unsold inventory and ensure rooms or services don’t go unused.

2. Flexible Pricing Strategies

Implement dynamic pricing strategies that adjust based on demand. When the lead time is short, and demand is high, prices can be adjusted upward. Conversely, if there’s a lot of available inventory, lowering prices can help increase occupancy.

3. Targeted Marketing Campaigns

Launch targeted marketing campaigns aimed at last-minute bookers. Use social media, email newsletters, and other direct communication channels to reach potential customers with special offers and incentives that encourage immediate booking.

4. Streamline the Booking Process

Make sure the booking process is as easy and fast as possible. A streamlined, hassle-free booking experience is crucial for capturing last-minute bookings, especially through mobile devices.

5. Package Deals and Extras

Create package deals or offer added extras, like free breakfast, parking, or Wi-Fi, to make last-minute offers more attractive compared to competitors. This can also enhance guest satisfaction and perceived value.

6. Partnerships and Cross-Promotions

Collaborate with local attractions, restaurants, and event organizers to create comprehensive last-minute packages. This not only adds value for guests but also promotes local tourism.

7. Efficient Resource Management

8. Utilize Technology for Real-Time Insights

Invest in technology solutions that provide real-time insights into booking patterns and guest preferences. This data can inform decisions on pricing, marketing, and resource allocation.

9. Encourage Reviews and Social Proof

Positive reviews and recommendations can be particularly persuasive for last-minute bookers. Encourage satisfied guests to leave reviews and share their experiences on social media to attract more last-minute bookings.

10. Loyalty Programs

Adjust loyalty program rewards to offer immediate benefits or perks for last-minute bookings. This can encourage loyalty members to choose your company even when booking late.

Companies facing shorter booking lead times should focus on flexibility, efficiency, and targeted marketing to attract last-minute bookers while optimizing their revenue and ensuring a great guest experience.