Are you looking for some insights on the seasonal trends of bike-sharing services? If so, this blog is for you! In this blog post, I will analyze a Kaggle dataset that contains the leasing data of a bike-sharing company in the east coast of Illinois. This dataset is obtained directly from the company, so you can trust its accuracy and reliability. The data quality is also high with no sparse. Let's get ready as we are going to unravel the layers of information, uncovering seasonal trends, geographic preferences, and valuable insights that could redefine the company's strategies.
The dataset covers the period from January 2022 to December 2022, and it includes information such as the start and end time of each lease, the station location, the geographic location of each station, the duration and the user type. It includes a total of 14 files, totalling to over 1.2GB of data.
Picture: Data Information from Kaggle.
Using some data visualization tools, I will show you some interesting patterns and findings that can help you understand the market demand and customer behaviour of bike-sharing services.
Riding the Wave of Success: The Ascending Moving Averages
Firstly, since this is a real business model, our data odyssey begins with the overall trend of the leasing volume over time to assess the sustainability going forward for the company. As you can see from the chart below, the company is gaining tractions as the moving average trend is up. There is also a clear seasonal pattern, as the leasing volume increases significantly when the weather gets warmer, from March to Fall. This makes sense, as people are more likely to ride bikes when the temperature is comfortable and the sun is shining.
In the realm of business analytics, trends speak volumes. As the temperatures rise, the correlation between weather patterns and user engagement underscores the significance of adapting services to the seasons. The temperature data below is taken directly from WeatherSpark to provide more insight into this correlation. In general, it seems the people of Illinois tends to welcome temperature above 5 degree of Celsius, which leads to the pick up in leasing bike.
Unveiling Daily and Weekly Rhythms
Next, let's explore how the leasing volume varies by time of day and day of week. The heatmap below shows the average number of leases per hour for each day of week. As you can see, there are some distinct patterns depending on whether it is a weekday or a weekend. During weekdays, riders tend to ride more from 16:00 to 18:59, which corresponds to the evening rush hour. This suggests that many people use bike-sharing services as a mode of transportation for commuting or running errands. During weekends, however, the leasing volume is spread out more evenly from 11:00 to 17:00. This implies that people use bike-sharing services more for leisure or recreation purposes on weekends. Interestingly, there is also a slight preference for riding bikes in the second half of a month, which may be related to some monthly events or cycles. Hence, the company might be able to charge extra fees during these rush hour period.
Unveiling the Geographic Tapestry
Now, let's examine a subtle yet crucial detail—the geographic footprint of the company. We can learn about the spatial distribution of the station locations and their associate usage frequency, which can help the company to strategically optimize operational efficiency. The map below shows the location of each station as a dot. Through careful analysis, it becomes evident that the firm operates solely along the picturesque East Coast of Illinois.
As you can see, most adaptation areas are near the coast, with many stations rarely getting used the farther away from the coast. In fact, when we examine those locations where leasing count over 5 (top 80 percentiles) and over 1000 (top 30 percentiles), it is clear that people prefer to ride bikes along the scenic routes or to access some attractions near the water. The company may want to consider expanding its network to cover more inland areas or to offer some incentives to attract more customers there.
More specifically, as can be seen from the picture below, these top 17 stations account for a total of close to 8% of total rented time. Moreover, the majority of them locates near the shore, which reiterates the idea proposed above. Interestingly, among these stations, only 1 station has the majority of bike being rented is electric bike. On the other hand, 10 of 17 (58%) of the stations most successful product is docked bike for non-membership members.
Unlocking the Time Puzzle: Pricing Strategies and Rental Duration
Another important aspect to analyze is the leasing duration. How long do people usually ride bikes for? The line chart below shows the frequency distribution of the leasing duration in minutes. As you can see, most leases are very short, with a peak at around 5-10 minutes. The frequency drops sharply as the duration increases, and there are very few leases that last longer than an hour.
Even at the 5th top frequent leasing duration percentile, there is still a significant drop off near 150 minutes. Thus, the company should focus on pricing strategies below 150 minutes to maximize revenue and profit. For example, the company can offer 5 tiers of standard charging, with one in the range of 0-10 minutes, the other one comes at 11-30 minutes, the third one lies between 31-60 minutes, and the last two tiers focus on 61-150 minutes and beyond 150 minutes.
Unravelling Relationship Between Membership and Bike Preferrence
Now, we shift our focus to the intriguing realm of bike preferences and membership dynamics. As can be seen from the charts above, among all clients, the classic bike reigns supreme, triumphing over its electric counterpart by a substantial 33%. This preference could be attributed to a potential supply constraint, where the company might offer fewer electric bikes compared to the classic ones, prompting users to opt for the readily available and familiar classic bikes. With further information about the revenue brought in by both type of bikes, the company can make an informed decision to adjust their bike offering.
Delving deeper into the realm of memberships, a nuanced landscape unfolds. Memberships, while contributing significantly to the company's clientele, account for just 48% of the total renting minutes. This insight suggests a diverse user base that includes both members and non-members, emphasizing the need for tailored strategies to cater to the distinct preferences and behaviors of each group. Interestingly, a noteworthy observation surfaces—virtually none of the members opt for docked bikes, while docked bike is the most popular product in the most productive stations as observed above! This curious trend sparks curiosity about the factors influencing member choices and opens avenues for targeted marketing and service adjustments to enhance the member experience.
Embracing Insights for the Journey Ahead
In conclusion, our expedition through the East Coast Illinois bike rental dataset has unearthed a trove of insights—a compass guiding the company towards informed decisions. From coastal station preferences and optimal pricing strategies to the nuanced rhythms of daily and weekly usage patterns, the data illuminates a path for refinement and success.
I hope you enjoyed this data-driven adventures and learned something new about bike-sharing services. If you have any questions or comments, please feel free to leave them below. Thank you for reading!
Komentar