Data Study on Laptops market on BestBuy

Posted on Feb 19, 2018
The skills the author demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

You may have a bad time to buy a laptop. But did you ever think maybe laptop makers have really bad time to sell a laptop too?


Did you ever struggle to choose the best laptop that fits your need? I did. The reason is that data shows there are so many laptop brands on the market right now and it is really difficult to pick the best from them because they are so similar to each other. But, did you ever think that it also could be difficult for laptop maker to sell their products since this market is quite competitive and saturated? Maybe you didn't.

Recently, Dell, the largest private laptop maker announced on February second, 2018 that it could be going to merge with its subsidiary company Vmware or going to have its IPO. This marked the possible move for Dell to explore the new market and transformation of its business. And I checked some Asian laptop makers' stock, they didn't perform well in the recent years. The stocks of them are either moving in the consolidation or moving in the downtrend.

Data Study on Laptops market on BestBuy Data Study on Laptops market on BestBuy Data Study on Laptops market on BestBuy

I am quite interested in the laptop market. Therefore, I try to find the possible reasons behind this. But the problem is that some laptop companies are not listed on NYSE or NASDAQ, it is hard to find their financial statements and target market. Therefore, I try to use data science to analyze laptop market. If you are interested, you can check my Github for the code.

Where is the data?

Web scrapping Data

In order to find the target market and possible market strategies of each brand, I scrapped the laptop's information on the Bestbuy which is the largest laptop retailer in the US.

1. What do I want?

In order to analyze the target market, I need to scrap the price of each product and the reviews of each product. Besides, I also need to have the selling rank of each product. But I didn't scrape the information of products that have zero reviews because these products are lack evidence to analyze.

2. How do I get?

I used the Scrappy package in Python to get my data. First, I sorted the laptops by using the tab 'Bestselling' and I set the laptop condition as 'New'. The reason for this is that I can get the selling rank and the market price of each product.

Becuase there is JSON code on the website. So, I used two spiders to scrap, one for the price, the name of each product and selling rank of each product, another one is for the details of each review. For the reviews, I scraped the name of the author, the percent of recommendation, pros, cons, number of helpful, number of unhelpful, stars given, titles of each review and the number of reviews.

3. How to clean data?

  • Because I used two spiders, so I had two CSV files and I needed to merge them together.
  • The score given in review is an integer. It is hard to analyze the data that is not consecutive. So, I use the number of helpful and the number of unhelpful to adjust the score. I multiply the score by the ratio between helpful and total of helpful and unhelpful to get 'my_score', and I further dropped the review that has the ratio less than 0.5 which means most people don't agree on the points of the review.
  • The information of brand and features are included in the name for the laptop and I need to find a way to extract them. So, I created several duplicated rows so that I can group by the reviews by different features or brands.

Data Analysis

1. The popularity of brands

This plot is the count of reviews each brand has. We can find one interesting thing that there is no single Japanese brand listed on BestBuy. It could be nobody left a review on the Japanese brand or there is no Japanese brand. So, I tried to search the whole BestBuy but I still can not find a single Japanese brand. In my opinion, this could be the protection of other brand listed on BestBuy and later we could find the possible candidate for this kind of protection.

Other than that, we can find the top three most popular brands on BestBuy, which are HP, Apple, and Microsoft. And it is also interesting to find that the gaming laptops like Alienware, MSI, Razor are not popular on BestBuy. In my opinion, most hardcore gamers may be more concerned about the benchmark of graphics card and overall computer gaming capability. As a result, these gamer may go to other more sophisticated websites to find their gears. Therefore, we can find from this plot that the parent, businessman, students who don't want to spend too much time to buy a laptop could be the possible customers on the BestBuy.

2. The price box plot and my_score

These two plot is quite informative, we can find many interesting insights from them

  1. There are basically three kinds of laptops on BestBuy

  • The gaming laptops represented by MSI, Alienware, and Razor
  • The ultrabooks represented by Apple, Microsoft. What's more, we can find Microsoft has many outliers on its price box plot.
  • The laptops for daily casual usage. We can find there are many brands locate in this category. They are HP, Lenovo, Dell, Asus, Acer, Google, and Samsung. And we can also find that Dell has some ultrabook outliers which compete with Apple and Microsoft.

2.  The possible protected candidate

  • After I search the internet, I can still find some Japanese brands that are sold on Amazon and the official sites. And I find that the target market of these brands, especially VAIO,  is quite similar to HP' s target market. So, in my opinion, HP could be under protection from Japanese brands.

3.  The insights from my_score box plot

  • We can find that there are some outliers on the my_score box plot. And I think that the outliers of Dell, Microsoft, and Apple could have some insights since they are more expensive than other brands and they don't have the laptops which are really cheap, especially the outliers of Dell and Microsoft. Could these outliers be the same laptop as the outliers on the price box plot?

3. Find the outliers

We can use different filter to find the outliers of Dell, Apple, and Microsoft


  • Dell - We can use the price filter to find the outliers. It looks like Dell' s ultrabook are not favored by the customers. They get some bad score and cons word, and they are quite expensive. It is reasonable to think that these ultrabooks are not only the outliers on price plot but also the outliers on the my_score plot.


  • Microsoft - We can search by using price. We can find that the laptops are not bad when we use price to filter. So, we can know that the outliers of Microsoft are not the expensive ultrabook. But if we directly use the my_score as the filter, we can find that there are some products have cons which are 'sleep mode'. And after I check the internet, I find that these laptops would close the backstage application when they go into sleep mode. This is quite annoying if you want to use backstage applications, for example, the music players.


  • Apple - We can use my_score to directly find the laptops have bad reviews. We can find that it is interesting that the 'touch bar ' appears in both pros and cons word. In my opinion, it is a strategy Apple uses to promote these products among people. Because this kind of controversial design can create the topics among the customer who bought these laptops and these topics can bring its products to more people. But it looks like that 'Thunderbolt' is a really terrible design and nobody likes it.

4. Dive deeper into pros and cons

What is interesting to investigate all cons word is that we can find the price which people tend to think that a computer is expensive. So, we use the 'price' as the filter to check the price of these 'expensive' laptops.


The average price of 'price' laptops is around 1300 - 1500. Based on that, we can go back to our price box plot to check the target market of each brand. Let's check price box plot first. It looks like that the price range of Apple is half below 'expensive' price and another half is above 'expensive' price. let's find out the possible strategy Apple uses based on that.

Apple's Strategy

For example, if I was a student who works really hard at the part-time job and I got 1000 dollars to buy a new laptop. I go to the BestBuy and use 1000 dollar as a filter to search laptops. And I can get 8 brands that I can choose from. It is really difficult for laptops makers at this range to get customers. That could be the reasons that the profit margin of casual usage laptop market is shrinking.

Let's think another kind of situation.

If I was still a student, but I had a rich uncle. He gave me some money to buy a new computer. So, I had about 1200 dollars. And if I used this price as a filter, I could buy an Apple laptop right now! Most students would like to buy an Apple laptop if they don't want to spend too much time to chose a brand. So, in this situation, I become the target customers of Apple. But, in reality, I don't have the rich uncle and nobody would like to give me money for free. So, Apple says 'No worries, I could be your rich uncle as long as you are a student. I can not give you money but I can offer you a discount which I only give students!'.

By far, the students could be the largest group of customers who want to buy a laptop on BestBuy. By doing so, Apple successfully 'steals' target customers from other brands and puts them in its own pocket. As a result, Apple makes the casual usage laptops market get worse.

Comparing Price Ranges

Some brands notice that the ultrabook has more profit margin. So, brands like Dell, Microsoft has its expensive ultrabook products. Apple also has expensive ultrabook products. Let's compare how other brands and Apple perform in this price range which is higher than 1500.

We can find that the Apple outperforms competitors that have the similar expensive market.

5. What's about the cheap laptop?

If the price range above 1000 is dominated by Apple. What's about the price below that. We use HP, and Lenovo because they have the similar target market.

We can find that they have similar popularity and selling rank. In my opinion, in this price range, there are no significant differences between two brands. The market share is based on the preference of customers.

6. Conclusion

From the data I got, I can find a really competitive market among laptops makers who's price range is less than 1000 dollars in the USA. In this price range, laptops are really similar to each other, and they can be easily substituted by each other. The most important force that drives the selling is the preference of customers. But Apple makes the shrinking profit margin even worse because it has the power to steals customer from other brands which have the laptops less than 1000 dollar.

Some brands like Dell and Microsoft notice that, and they have the capital to differentiate themselves from other brands by producing the ultrabook to compete with Apple directly. The result is that Surface takes away a small amount of market share from Apple but Dell's ultrabook is demolished by Macbook.

So, it is reasonable for Dell to have the plan to transform its business and explore a new market. Good thing for Dell is that it still has some nice subsidiary companies like Vmware which has many intangible assets and technology. But what about Lenovo, Asus, and Acer?  These brand's  large amount of profit in the USA comes from their laptops. It could be difficult for them to transform easily and to be competitive in the laptop market.

About Author

Zhe Yang

Hi, My name is Zhe Yang. I got my master degree in Financial Analyst at Rutgers University. I love challenges and solving difficult problems. I used to be a trader in the T3 trading company. During I worked...
View all posts by Zhe Yang >

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