Data Analysis of Amazon's Shampoo Category

Posted on Jun 29, 2020
The skills I demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.



Data shows Amazon has quickly become the US’s largest online retailer, dominating the ecommerce space across almost every category. One of the main reasons that the company has been able to do this is by offering a vast assortment of products and brands.

Being that Amazon carries almost all brands in every category and that being listed on its site is practically a perquisite to selling anything online I wondered if there were any insights into what makes a brand successful that could be found through scraping product pages. In this project I specifically scraped the shampoo category in order to find meaningful takeaways for a brand looking to enter or expand their presence on Amazon.

Scraping Methodology and Data:

To scrape Amazon, I created a Scrapy spider that was able to extract the ASIN (product identifier), brand name, current price, list price (reflects original price if item on sale), description, product detail bullets, product title, rating, number of reviews, product dimensions, merchant name (whether the item was sold and shipped by Amazon or other) and price per ounce from a product page. The spider successfully extracted this information where available from 4675 shampoo products out of the ~20,000 listed.

Data Analysis:

Much of my analysis rests on the assumption that higher amounts of product reviews is an indicator of a product’s sales on site. While product review amount sales volume may not be a perfect correlation it is well known that they are a primary driver of traffic and are the best quantity indicator of customer engagement available on a product page. Using this assumption I sought to address 3 strategic decisions that every seller must make when working with Amazon:

  1. The channel strategy- that is whether to be set up as an owned vendor that sells their products to Amazon for Amazon to sell their products to customers, displayed as “sold by and shipped,” a “Fulfilled by Amazon” vendor that sells their products to the customer directly but uses Amazon as a warehouse and distribution service for a fee, displayed as “sold by X and shipped by”, or a marketplace vendor that sells and ships their own product, displayed “Sold by and Shipped by X.”
  2. Whether to put focus on brand recognition in marketing or on product quality.
  3. How many products to list.

I also did an analysis on common word stems that came up on product titles and descriptions in order to see if there was a discernible difference between the 4675 product “field” and the top 100 best sellers in order to identify possible market gaps and successful marketing techniques.

SKU Channels 

The charts below display my findings on channel strategy (SKU stands for
“stock keeping unit,” a term referring to a single item):

Data Analysis of Amazon's Shampoo Category

In the shampoo category there is a high supply of marketplace items that receive very little customer engagement compared to the owned and FBA channels. This makes it clear that the customer in this category prefers items shipped by Amazon and that succesful brands are choosing these channels.

Unless there is an extremely compelling reason to list on marketplace unique to a vendor I would recommend listing through the owned channel or FBA. The decision to list as owned by Amazon or FBA needs to be made with a tradeoff of two priorities in mind: driving a higher volume of sales or retaining price control. While the data shows that vendors can expect more engagemnt being owned, being FBA allows them to control pricing and not be subject to price swings that can negatively effect margins.

Detailed Look

 In order to find the importance of branding versus product quality I split the brands into 10 groups based on the percentile of total reviews they received. With the brands split I was able to see how dominant the top brands in the category in terms of total reviews accrued, how their engagement on different at the product page level and how their star ratings differed on their products. The charts below display what I found:

Data Analysis of Amazon's Shampoo Category

There is clear dominance of the top brands in terms of total review share, with the top 10% accounting for over 70% of all reviews. These brands also have 50% more reviews on their product pages than the next group. However, the “Ratings By Percentile Group” boxplot shows that product quality does not appear to be a driving factor for differentiating this top group, with the top 6 groups having similar spreads and medians on their product reviews.

The conclusion here is that product quality does not appear to be the differentiating factor between high and low engagement brands and therefore other factors such as brand recognition and product marketing must be responsible.

Number of Reviews

The importance of brand recognition can be seen in the chart below, where I plotted the SKU count versus the the review totals for each brand:

Data Analysis of Amazon's Shampoo Category

From the chart it appears that for the most part a brand needs to have at least 60 listings to have significantly more reviews than the average brand (remember that this data only accounts for about 25% of the shampoo listings).


In the final part of my analysis I wanted to see if top products were addressing different needs or being marketed differently than the general population. To to this I performed natural language processing techniques to count word stems in the descriptions of the products listed on Amazon’s 100 best selling shampoo brands and compare it to the stems of the 4675 shampoo sample. I plotted my findings below (top 100 being on the left, the larger sample being on the right):


One of the more actionable insights to take away here is that dandruff is the third most mentioned term within the top 100 descriptions while not even appearing in the top 20 terms of the sampl.  This could mean there is a lack of products addressing dandruff issues and thus could be an area for a brand looking to expand their presence to focus on.

Another insight is that the stem “natur” appears as the second most mentioned term in the sample while not appearing in the top 20 of the top 100. This could be an example of a misguided marketing tactic that isn’t resonating well with the customer or an oversaturation of the market- many brands appear to be listing “natural” products with none becoming extremely popular. This means that brands may want to shift investment away from these products.


In conclusion the shampoo category seems to be brand-driven where large brands have robust assortments where customer has a clear preference for listings shipped by Amazon. This means there should be a larger effort and budget being put towards marketing rather than product quality. There is also a possible lack of products addressing dandruff and oversaturation of “natural” products.

About Author

Alex Ellman

Data scientist with background in ecommerce. Analytical problem solver with demonstrated history of leveraging data to achieve business goals. Able to combine previous business experience and domain knowledge with data science and machine learning skillset to create and...
View all posts by Alex Ellman >

Leave a Comment

No comments found.

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI