Data Study on Trade Coffee

Posted on Oct 21, 2018
The skills the author demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

The National Coffee Association National Coffee Drinking Trends (NCDT) report is the coffee industry’s leading survey of US coffee consumer behavior and attitudes. According to data from NCDT, more Americans are drinking a daily cup of coffee than they have been for the past six year. Up 2 percent from last year, 64 % Americans are drinking their daily cup of coffee. Since 2001, the NCA began tracking US consumption of specialty coffee, which is defined by the NCA as β€œcoffee drunk hot or iced that is brewed from premium whole bean or ground varieties. This includes espresso based beverages, iced/frozen blended coffee, cold brew, and iced coffee infused with nitrogen.”

Data Study on Trade Coffee

There clearly has been a consistent growth in the specialty coffee market.The US specialty coffee consumer drinks an average of 2.97 cups off coffee per day. The US market share of specialty coffee has grown significantly over the last 7 years from 40% in 2010 to 59% at present.

Americans consume the most coffee in the world and spend an average of $3.16 for a cup of coffee. The average American spends $1,110 on coffee annually. When taking into account the price one has to pay for a cup of specialty coffee, the number is even bigger. For instance, one can find America's most expensive cup of coffee in Brooklyn, NY, which goes for $18 a cup.

As a coffee consumer myself, I am always on the lookout on ways that I can save money and yet still enjoy a great cup of single origin coffee without breaking the bank.

What is Trade Coffee:

Data Study on Trade Coffee

As an alternative to buying from specialty coffee shops, home brewing has allowed me to save hundreds of dollars and indulge in some of the best cups coffee I have put my hands on. As I was searching for my next cup of coffee, I came across Trade Coffee - a website that allows consumers to purchase direct from different roasters across the country and get them freshly roasted on their doorstep.

The Trade Coffee website boasted aΒ curated collection of seasonal, freshly roasted-to-order coffees. I was curious what this new home-delivery service had to offer me and thought perhaps I could further investigate.

As I began my inquiry, these were some of the questions I had in mind:

  • Is TRADE a good resource for procuring and learning about coffee?
  • Do they truly offer a diverse selection of coffees that can appeal to different tastes, price points?
  • Will it satisfy both the needs of the seasoned coffee enthusiast and coffee novices?
  • Is it a good jump off point for those interested in getting into specialty coffee?
  • Would I benefit from using Trade Coffee as opposed to going directly to the roaster.

Data Web-Scraping Trade Coffee:

The web scraped that I needed to build had to scrape the main product Trade Coffee page and the individual product pages. I initially built a Scrapy spider to parse through the web site but ran into some issues such as missing data, hidden elements. As a result, I transitioned to Selenium to scrape the elements on the pages that I was interested in (product name, roaster, description, price, weight, etc.).Β Some of the more detailed product information had to be put under one 'basket' as these elements did not have unique Xpaths.

Data Cleaning and Analysis:

Data cleaning was an important step in making the data easy to work with and manageable. Duplicate entries were removed. Empty strings were removed or filled with a marker if it was necessary in the analyses downstream. Price and weight data entries were converted into numeric vectors. UnnecessaryΒ punctuation and other artifacts were removed.Β  This narrowed down the number of products to be analyzed :Β 470 products down to 452.

Trade Coffee offers beans in different package volumes and it was necessary to convert these into similar units. In order to normalize price data, I decided to use the price per cup data as a measure that can be compared across all products. Following my own brew recipe: I set 20g as the amount of coffee beans used per 300mL of coffee. I used this to calculate the number of cups I could get from a single bag of coffee. Most products on Trade Coffee yields 17 cups of coffee.

Data Study on Trade Coffee

Trade Coffee partners with 51 roasters from 39 cities offering 14 profiles of Coffee which can be either one of 261 Single Origins or 209 Blends. Offering the consumer a great variety of options to choose from. The average price per bag is $18.92 and price per cup is $1.05

Distribution

Diving deep into the data I wanted to look at patterns and relationships, specifically if there was a significant price different between Single Origins and Blends. As I expected, Single Origin coffees were priced higher on average than the Blends.

The same trend follows when price per cup data is looked at. There were several single origin coffees that were far from the median and mean price per cup of coffee. The highest price per cup of coffee was at $6.00 a cup or $85.00 per bag . The average for a single origin coffee is $1.16 per cup and $17.40 per bag. The average for a blend coffee is $0.92 per cup or $ 15.49 per bag. These numbers show that it is much cheaper to brew a cup of specialty coffee at home than going into a coffee shop.

Distribution of the price per bag of coffee.

 

Distribution of the price per cup of coffee

 

Trade coffee offers 14 different profiles of coffee. Below is a look into the breakdown of their products into these categories.

Choices based on Taste

 

An NLP analysis on the product descriptions and flavor profiles, gave a picture on what products one might find on Trade Coffee. the NLTK package was used to process the words from all the product descriptions and flavor profiles. In order to get an informative word cloud, I added words such as 'coffee', 'flavor', 'taste', 'varietal' into the dictionary of stop words that remove redundant and non-informative words from the description content.

Word Clouds

Below are word clouds of the flavors describing the different product on Trade Coffee. One is using the short and more simple flavor notes and the second was created with the more descriptive and verbose roaster notes and descriptors.

Word cloud describing common coffee flavor notes.

 

Word cloud showing the detailed descriptors used in the roaster notes of products on Trade Coffee

 

Some Thoughts and Insights:

Trade Coffee has a wide variety of coffees for a all types of home brewers. Consumers have the option to purchase budget budget friendly coffees as well as special limited and seasonal lots and roasts. The findings of this venture reaffirms that home-brewing is a cost-effective way of enjoying specialty coffee in the comforts of one's home without making a big dent on the wallet.

Next steps:

It is important to note that Trade Coffee regularly changes their inventory depending on the roster of roasters on the site and most especially the seasonality and availability of certain coffees throughout the year. It would be interesting to see trends in their product offerings as well as how often these inventory changes occur.

I would like to take a closer look into the specific details of products such as varietal, origin, sub-region in addition to the price points and shipping costs.

About Author

Denise Sison

Denise is currently a Data Science Fellow at New York City Data Science Academy. Denise has extensive molecular biology and clinical research experience with a focus on infectious disease epidemiology and neonatal births outcomes. Denise was educated in...
View all posts by Denise Sison >

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