Analysis

In this dataset, some columns, including 'id', 'image url', 'description','region url', 'url', 'VIN', and 'county', are irrelevant to the analysis; since the aim of this project is to ascertain what factors influence a car's sale, urls cannot be used to extract a lot of information. The VIN is also irrelevant since it is a car's identification number. The 'description' and 'county' columns were removed since they contain mostly invalid (NaN) values. In addition, the seller also drafts the description, which means it could be prejudiced and not fairly describe the condition of the car. Those columns were removed from the dataset.

According to the descriptive table above, the majority of used cars were made around 2011. The newest vehicles listed on Craigslist were created in 2022, while the oldest vehicles listed there were created in 1900. The 'year' column's 25th percentile, 50th percentile, and 75th percentile values are 2008, 2013, and 2017, respectively, showing that the majority of the cars were produced in the 2010s. The peak of production that occurred in the 2010s was due to the establishment of multiple car factories in Mexico (H, 2014). Mass production back in the 2010s resulted in an excessive number of vehicles listed in the market. Since most of the used cars in the market are one decade old, the target consumer for the dealer might not be a buyer who feels that a vehicle from the 2010s is too old.

The minimum, mean, and maximum odometer readings are 0 miles, 98043 miles, and 10,000,000 miles, respectively. The maximum odometer reading in the original dataset is possibly an outlier, because according to AAA, the vehicle with the highest mileage in the world is a Volvo with 3400,000 miles (Doyle, 2022). This outlier could be a typo in the listing. The odometer readings' 25th, 50th, and 75th percentiles are 37,704 miles, 85,548 miles, and 133,542 miles, correspondingly. The percentage indicates that there aren't many used cars available on the market with less than 37,000 miles on them and that the majority of them have between 80,000 and 90,000 miles on them. This makes sense since, according to data by Lending Tree, a car loses the greatest value in the first five years of ownership, therefore most new car owners would not want to sell their vehicles during this time because they risk losing a significant amount of money (Wollenhaupt, 2022). Customers who find a car with 37,000 miles undesirable may be persuaded by a dealer to consider a new vehicle because it is uncommon to locate a used car with lower mileage.

The average cost of a used car is $7519, and the highest price recorded in this dataset for a used car is $3,730,000,000. Because this price is so much higher than that of the second-most expensive car on the market, it is possibly due to a web scraping error. Due to the possibility that some individuals may simply give their vehicles away, the minimum price for a secondhand automobile is $0. $5900, $13,950, and $26,485 are the 25th, 50th, and 75th percentiles of the "price" column, respectively. This demonstrates that the majority of used automobiles on the market are priced between $10,000 and $20,000, which gives the dealers some sense of how to price a car. Customers with budgets in this range should be given preference since they are more likely to locate a car that suits their requirements and close the transaction sooner.

More insights and details on how the dealers can create an efficient procurement and pricing strategy, and what factors the dealers should be aware of will be discussed in the following sections.