How Does Traffic Congestion Affect Starbucks Locations’ Ratings? Does Being A Company-Owned Or Licensed Starbucks Bring In Higher Ratings?
I wanted to discover if traffic data correlates to bad business reviews on Yelp, particularly in Boulder, CO, with the most current CSV data files I could find (not formulating a time series like my past module with these files, but just correlating current data with the Yelp API requests). I then wanted to correlate my ratings and locations found on Yelp to a CSV file of Starbucks’ Directory, where it deciphers each location as either a Company Owned location or a Licensed location. This way, I could determine whether being company owned or licensed brings more customer satisfaction.
I first created a Yelp API Application. I wrote a CSV file titled “boulderStarbucks.csv” as well from the Yelp API data I requested for the 9 Starbucks locations in Colorado that have at least 3 reviews. Here are all the parameters for the Yelp API, to better understand. There were originally 12 locations that came back off the request, but 3 of them only had 1 review, which doesn’t really give me enough evidence to conclude that the review was fair (one location’s single review was ⅕ stars); therefore, I scraped those out through an IF statement. This is what I ended up with:
There could be many reasons as to why these ratings get the ratings they do. Some could be that there are more workers, better workers and workers with more experience/capability, the amount of time that location has been open for, and so on. Starbucks’ 29th St. Mall location having the highest ratings does not surprise me. People are usually in a good mood when they’re shopping, and far from in a rush, unlike on 28th St., which is a busy roadway dragging off of Highway 36.
The CDOT traffic data I used was an exported CSV file of AADT (Average Annual Daily Travellers, “AADT.csv” ) on all the 60 routes provided in Boulder. I sorted these in ascending order, and then made a bar chart to visualize which areas are the most congested. This is what I got for this graph:
The Foothills Parkway has the most AADT in Boulder. Following that leader in descending order of AADT is Iris Avenue’s merge with the Foothills Parkway, Canyon Boulevard crossing over Broadway onto 9th St., all of Broadway and Baseline, then onto 28th St. and Arapahoe Ave. This makes a lot of sense, as Foothills is a continuing highway off of Highway 36, and is from my experience very crowded. I was however surprised that Canyon Blvd. was more congested than 28th St., as 28th seems to be very busy as well, rooting off of Highway 36.
So the next move for me was to pull the 9 closest locations from the CDOT in proximity to the Starbucks’ locations, so I could aggregate the data further. For the Starbucks location on 29th St., I used the same station ID to count AADT as I did for the Starbucks location in Target and on the 28th St./Pearl St. merge, as these 3 Starbucks locations are within a 5 minute walk of each other. There was really no better option otherwise for a different station ID. I think this is what gave me difficulty in finding a really strong correlation in my data, as there are just so many locations right next to each other; nevertheless, the ratings varying but all using the same station (which has the same number of AADT) created a scatterplot that couldn’t tell me anything significant at first glance, especially if you don’t know the area yourself.
As you can see, the three locations all within a 5 minute walk of each other are lined up vertically at the beginning of the x-axis, and the 1.5 star location is right in the middle of AADT. In my opinion, the ease and accessibility of being able to make a right turn into stores like the one on Canyon Boulevard will generate less already dissatisfied customers. Safeway’s Starbucks having only 1.5 stars is likely due to a few reasons: people on Arapahoe going west towards the store don’t want to take a left as it’s much easier to take a right, and there isn’t a lot of parking. You also have to walk through a grocery store. The speed limits are slow as well, with a bunch of school zones being nearby. Baseline’s Starbucks is really difficult to get into as well, with very little parking as well, and impatient drivers not letting you get over to turn in, no right on red to turn on Broadway, and then once again having to make a U-turn to get in. The lights are also extremely short, and everyone is driving on those roads when it is coffee time.
For the last part of the project, I ended up collecting a Starbucks Directory CSV file, which listed all the locations around the world for Starbucks. I scraped out just Boulder’s locations, and had to do even more scraping on top of that, so I could narrow down to the 8 locations which already matched on my data retrieved from the Yelp API. I then calculated the average ratings of Starbucks locations that were company owned vs. licensed.
From this, we can conclude that the locations in Boulder that are company owned have higher ratings than the locations that are strictly licensed. In conclusion, the more control corporate has, the ratings tend to be higher. “Licensed locations use their own employees and their own equipment. They are licensed to display the Starbucks logo and to sell Starbucks branded products. Drinks such as lattes, frappuccinos, etc., should be pretty much the same as in a regular Starbucks location” (quora.com). Operation methods could be different at these licensed locations, which are in Target and Safeway. The full menu is never really at these locations, so I wasn’t surprised that there was a lower average review for licensed locations.
If I was to continue working on this project, what I would do next is use that Starbucks Directory CSV and correlate it with a generated CSV file from the Yelp API. This would help me find similarities and differences within different cities around the country. If not that, then I wouldn’t just delve into Starbucks, but other chains, seeing where certain coffee shops prosper the most, what locations that generate Yelp reviews are supporting small businesses over larger chains (like Starbucks), and so on.
