Venue check-ins in Tokyo - how venues can leverage the power of social media

Posted on Oct 12, 2017

A recent article from Harvard Business Review stated that no industry is immune to digital transformation and neither can venues turn on their blind eye to how their business models are constantly being challenged. Especially, location-based marketing (check-ins) has shown to be a key element in digital marketing to leverage the power of So-Me platforms.

In this blog, which is a summary of my first project delivery at NYC Data Science Academy fall 2017, I investigate data from venue check-ins in Tokyo. Given that this domain has unlimited directions to explore, I particularly want to investigate venue check-ins according to location, venue, time, gender and popularity. Thus, the research questions I want to address are the following:

  • In what areas of Tokyo is check-ins more frequently used? Are there any clusters?
  • At what kind of venue do people more frequently check-in?
  • Does the number of check-ins change by time?
  • Do popular people check-in more often than non-popular people, and is there a difference between the number of check-ins between males and females?

The questions have been processed in R and visualised in a Shiny application.

Based on the answers from these questions, it’s my intention to sum up how venues can take advantage of the insights and utilise the concept of location-based marketing.

 

Summary of the dataset

My project is based on dataset containing information about 120,000 check-ins in Tokyo from April 2012 until February 2013. For each check-in, I have information about the user and venue category, latitude, longitude and a timestamp feature. Also, I was able to join the dataset with personal features of the user such as the number of followers and followings on Twitter, from which I created a new variable called popularity. Thus, if a person has more followers than followings he is considered popular, vice versa. Lastly, I decomposed my timestamp feature into time of day, weekday and month, so that the final dataset comprises 15 features.

 

Tokyo from above

The following map shows check-ins in Tokyo represented by the coloured dots. Each check-in holds information about the specific venue, gender and popularity. The color indicates at what time of day the check-in occurred. Thus, red color presents a check-in during the morning whereas the blue color represents a check-in in the afternoon. On the left-hand side of the application you are able to filter the check-ins to be displayed on the map by popularity, gender and weekday. The idea is to visually search for pattern in the data. The right-hand side of the application summaries the top 5 venues and the total number of check-ins based on the input selection.

Intuitively, it seems like the data is spread around Tokyo and its boroughs. However, if one tries to filter by e.g. afternoon, there appears to be certain areas where check-ins happen more frequently. Especially in some of the outer boroughs as well as the city center the number of check-ins occur more often.

 

Let's hang out at the train station!

The Tokyo map didn't quite capture the best visited venues. Likewise, it would also be interesting to explore how the check-ins are distributed throughout the day. By grouping by distinct venues and stack the time of day for each venue, the following bar chart shows some interesting findings.

Surprisingly, venue check-ins at train stations and subways are a lot higher than check-ins at other venues. What is the reason for this? Do train stations and subways in Tokyo provide free Wi-Fi, which attract people to hang-out there, and is it a prerequisite to check-in before using the Wi-Fi? Or is commuting just the time of the day where you use your phone?

In order to get a more sophisticated overview of the rest of the top venues, let's filter out train stations and subways.

The check-ins appear to follow a somewhat expected pattern, where check-ins during the morning typically happens either on the way to or at the office or at a convenience store, at a restaurant during the midday, well distributed by different venues during the afternoon and centered around restaurants and bars in the evening. Overall, people are not prone to check-in when they are at home or during their regular weekday tasks such as being at the gym and at various groceries.

 

 

Summer and winter seasonality

Without conducting any statistical time-series analysis, it intuitively seems like there is a seasonal pattern between the summer and winter months. Running from May and until the end of July, we observe a spike by the number of check-ins. However, the number of check-ins drops drastically during the fall, but as we enter the winter months (especially December), the number of check-ins increases again. Logically, the seasonality could be due to holiday periods and thus increased activity on So-Me platforms.

 

Japanese males check-in more often than females

The first graph visualises the number of check-ins by popularity. It does not seem like that there is a huge difference between popular people and non-popular people according to the number of check-ins. Thus, the two lines are very symmetrical during the whole observation period, although popular people seems to be a bit more prone to check-in compared to non-popular people. This is valuable insight based on the assumption that popular people have a bigger social network and thereby providing additional marketing exposure to the venue when they check-in.

Looking at the other graph it’s quite eye-catching to see the difference between the number of check-ins by gender. Males are much more active than females with a majority of males of almost 90 percent compared to 10 percent females. Because of this, the seasonality pattern from the above graph is also reflected in this graph by the number of check-ins by males, whereas the number of check-ins for females is relatively constant during the observation period.

Since the dataset is heavily imbalanced by a majority of males, we need to be careful about generalisation. But it is not excluded that Japanese males do check-in more often than females and therefore appear more frequently is the dataset.

 

Utilising the insights of check-in data

By only exploring a small number of features, check-in data has shown to be a powerful component to extract insights about user characteristics and behaviour. Location-based marketing has several benefits to a venue such as additional marketing exposure, implicit public acknowledgement and the opportunity for venues to offer real-time sales offerings. As a way to flatten out the seasonality, an idea could be to display offline advertisements on price discounts if the consumer shows the venue that he has checked-in. In that the venue create additional sales but at the same they get additional market exposure through the customer’s social network.

Furthermore, since the majority of check-ins are done by males, they obviously represent the primary target group for venues. However, females should still be considered as a latent segment. Also, since popular people tend to check-in more often, venues can benefit from this insight by directing their marketing effort towards popular people since they have a bigger social network and thereby constitute a bigger indirect marketing channel.

 

 

About Author

Related Articles

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