creator. designer. writer. and everything in between.
modernist (1).png

Philly Happy Hours

Philly Happy Hours, Deconstructed

"Modernist Home" by Alexander Vidal,  Dribbble

"Modernist Home" by Alexander Vidal, Dribbble



Seniors at Penn are constantly looking for new bars and restaurants to visit before we leave Philadelphia come May. Happy hours are perfect for reconnecting with friends - old and new - before we split ways. As a result, we set out to build a guide for various types of students to select the best happy hour for any occasion. With our guide, any student can:

  • Scroll through the tables and filter to find restaurants with different atmospheres or cuisines
  • Identify which neighborhoods have the best deals on food and drinks for the cuisine or atmosphere you're looking for
  • See when most happy hours start and find one that fits your schedule
  • Check which neighborhoods have the cheapest college party bars

For the student drowning in debt

For the maximizing event planner

For the adventurous extra-bubble explorer

For the lovebirds

For peak OCR season

For those slightly too sophisticated for frat parties

For the social butterflies

For the sceney

Atmosphere Breakup


We scraped all of our data from The Drink Nation using First, we scraped the front page for name, neighborhood, happy hour timeframe and specials, then scraped the inside links and chained two web-scrapers together in to get aggregated atmospheres, cuisine, average drink price, and average food price. 

We cleaned the data using Open Refine to cluster similar cuisine names together; split columns to get start and end times for happy hours, individual atmospheres, high and low drink/food prices; and eliminate leading spaces. Next, we used Excel to separate aggregated strings into two different cells using LEFT and RIGHT functions with a semicolon delimiter; separate large text dumps into different cells using "Text to Columns" function with a semicolon delimiter; and delete all irrelevant columns. 

To merge data and create visualizations, we used Tableau to merge and match data based on restaurant name; visualize data through several different chart formats based on usefulness to students in price, offerings, location, and atmosphere; filtered the top 10 based on both record count and proximity to Penn's campus for charts that were cluttered with too many neighborhoods or cuisines; and created visualizations for the top 5 demanded atmospheres in our target viewership, Penn students - these were "Impress Your Date", "College Party Bars", "Young Professionals", "Good for Groups,  and "Trendy".

DISCLAIMER: Only restaurants with Happy Hours on Thursday, Friday or Saturday have been included in this dataset.

Who are we?

Laura Gao and Suriya Sharma are seniors at The Wharton School at the University of Pennsylvania. This data project was conducted for Prasanna Tambe's course, OIDD215: Analytics & The Digital Economy.