Scraping Pet Supplies Portal Chewy.com
With spending on pets in the United States rising annually and projected to reach almost $100 billion by the end of the decade, the pet industry is considered virtually recession-resistant. It feeds on factors such as the multi-generational appeal of pets, the positive correlation between good human health and pet ownership, and a pet culture that spawns spending on luxury goods for animals and global expansion.
Riding on the back of the lucrative pet supplies industry is the online portal Chewy.com. With a wide product offering from leading brands, it is where pet lovers shop. For this project, I used Scrapy to Web Scrape pages on Canned Dog Food at Chewy.com.
The main objectives were to uncover:
- Market Shares
- Product Segments
- Product User Ratings
- Product Marketing Strategies
Key targeted information obtained include Product Name, Maker, Food Texture, Price, Weight, Rating, Number of Reviews, Product Key Benefits.
A total of 780 products are offered by the Top 20 Brands in this segment. Most notable are Bill's Buffalo and Purina with about 100 products each. They are followed by names like Cesar, Wellness and Hill's.
Number of Reviews and Average Ratings are both equally important indicators of the Product Popularity. A combined score was formulated with components of both metrics using the equation
Score=Rating + 5*(1−e(Reviews/Q))
where Q is a tuning constant
(Score : 0 - 10)
Highly skewed Distributions of Ratings & Reviews are transformed into a Normally Distributed Score. The upper range of the distribution approaches a score of 10 for products.
It is undeniable that Customer Reviews and Feedbacks are highly valuable encompassing both Positive and Negative ones. Sentiment Analysis was performed on selected reviews.
Additionally, Word Clouds were used to provide interesting insights on most commonly used words in: Positive Reviews
Product Key Benefits (Marketing)