Data Price-Setting Strategies for Electronics Sellers

, and
Posted on Jul 31, 2020

Photo by Glenn Carstens-Peters on Unsplash

GitHub Repository | LinkedIn: Aniruddha Dhar, Lucas Kim, Sita Thomas

Data Science Background

Our fictional eCommerce employer, TVStore, wants our data team to explore competitor pricing for 55-inch TVs and recommend a pricing strategy, with no specific products in mind to sell.

The dataset we’re given is the only dataset available, despite the data being roughly two years out of date.

  • There are 187 observations of 55-inch TVs across three groups:
    • High-end models at a minimum cost of greater than $2,000
    • Mid-range models costing between 1 and 2 thousand dollars
    • Low-end models costing less than a thousand dollars
  • Only 4 brands are represented in the data: Hisense, LG, Samsung, and Sony
  • There are 4 competing eCommerce websites: BestBuy, Walmart, eBay, and B&H Photo Video

We chose to explore the following strategies:

  • Cost-Plus Pricing
  • Competitive Pricing
  • Price Skimming
  • Penetration Pricing
  • Value-Based Pricing

Data Analysis around Cost-Plus Pricing

Cost-plus pricing is simply taking the base price of a product and adding a markup, typically a percentage of production costs.

Since we have no specific products that TVStore wants to sell, we do not have enough information to make specific suggestions.

The plot above shows that some TV models have a high variation in pricing. Moreover, cost-plus pricing for high-end models in particular could be more profitable due to the higher range in which markup will fall, versus cost-plus pricing on mid-range and low-end models that tend to be priced within a smaller range across merchants.

Competitive Pricing around Data

Competitive pricing is setting product prices at, above, or below the prices of the same or comparable products sold elsewhere.

First, high-end models fluctuate by about $3,000 at competitors, with a median minimum price of $2,098. If TVStore offers Premium incentives to justify high prices, these TVs could turn a high profit. High-end models could alternatively be loss leaders, deliberately sold below market rate to spur other purchases. TVStore may be able to pair them with other high-end products that have a greater markup. Customers shopping for high-end products likely have the income for additional purchases and/or recognize the value of quality items.

Secondly, low-end product prices fluctuate by $600, making them the best candidate for price matching or loss leading. Price matching is likely the safest route, allowing TVStore to make a small but less risky profit, while loss leading may result in overall losses if TVStore cannot persuade consumers to buy additional products.

Lastly, mid-range prices fluctuate by $800. Price matching is also a safe route for these products, but loss leading here may have a slight advantage over low-end products, under the assumption that mid-range customers have more flexible income than low-end buyers.

Price Skimming: Product Exclusivity

Price skimming is setting a high price at initial product release and lowering the price over time as competition appears. This database does not capture the initial market appearance of any given product (only when they first appeared in the database), so we cannot conclusively demonstrate the existence of competitor price skimming.

However, carrying an exclusive product may indicate an opportunity for price skimming due to lack of competition. None of these products are exclusive to a retailer, as the graph below shows, indicating that TVStore could likely price skim at a high profit if they had a unique product on the market, as they would be the only website to do so.

Price Skimming: Price Duration

The above plot illustrates, low-end models maintain a consistent price for shorter durations in relation to their total time in the database than mid-range and high-end models, demonstrating their market volatility, which is likely due to the volume of competition and the extensive prevalence of sales in that price group. Adding more, the mid-range and high-end models, however, hold their value more consistently for longer, making them better candidates for price skimming than low-end models.

Penetration Pricing

Penetration pricing is introducing a new product below the market rate of similar products and increasing the price over time as product/brand recognition improves. Because, again, we have no data on when products were first introduced on competitor platforms, nor do we know what specific products TVStore wants to sell, we cannot make specific suggestions for penetration pricing. Instead, we explored what pricing models competitors did seem to be using, as there may be evidence of past penetration pricing.

In the Price Per Model graph above, Best Buy shows some potential evidence of price skimming, but as we know, we have no data for any product’s initial market appearance, and many Best Buy products have no multi-date observations, so this evidence is circumstantial.

Walmart probably focuses on a combination of psychological and time-sensitive pricing strategies. They carry predominantly mid and low-range products, and runs a lot of sales, as shown in the Sales by Price chart below.

eBay prices hardly vary at all, and there is low product variety, so competitive pricing is probably the strategy of most sellers on that platform.

B&H Photo Video may rely on a variety of product or customer clustering strategies because they carry a relatively even distribution of high, mid, and low-range products, but they do run a fair number of sales like Walmart (shown below).

Given that this data is insufficient to determine penetration pricing for the same reasons it is insufficient to determine price skimming, we cannot say if it is a worthwhile strategy for new products. In order to be successful, it would have to not only draw customers away from competitors, but keep them purchasing other products at TVStore in order to offset the losses of the initial penetration price.

Therefore the risk of penetration pricing would be poor profits at best, losses at worst, if TVStore were unable to keep customers. Penetration pricing may be most beneficial when combined with other pricing strategies, in order to balance overall pricing risk.

Value-Based Pricing: Market Share

Value-based pricing sets prices based on customer perception of a product’s worth. The most highly-valued products are high-end models from manufacturers with extensive brand recognition. The graph above shows that Sony and Samsung have more products (and therefore likely better brand recognition), as well as more high-end models than Hisense or LG in this dataset, making them better candidates for value-based pricing. Pricing a low-end model higher than the market rate would require adding value to the product, such as extended warranties or offering generous return policies, similar to Premium competitive pricing.

Value-Based Pricing: Review Scores

Product reviews are an indicator of perceived value by customers, but for this data set are too similar to support value-based pricing.

Customers expect certain technological specifications for models in each price group. Although this data is about two years old, these specifications are still relatively highly valued today as there has been little progress in bringing new technologies to market.

Shipping options may also influence perceived customer value. Mid and high-end models nearly always come with free shipping while low-end models are split 50/50 between free and Standard shipping.

Recommendations: High-End Models

Cost-plus pricing for high-end models in particular could be more profitable due to the higher range of competitor prices in which markup could fall. If TVStore offers Premium incentives to justify high prices, these TVs could turn a high profit. As well, high-end models could alternatively be loss leaders, paired with other high-end products that have a greater markup.

Carrying an exclusive high-end product indicates an opportunity for price skimming due to lack of exclusive product offerings among the competition and the evidence that high-end products hold their value more consistently for longer. High-end models should have at minimum a particular set of specs and high-end models should come with free shipping.

Recommendations: Mid-Range Models

The safest competitive route for mid-range models is price matching. Competitive loss leading mid-range products may have a slight advantage over loss-leading low-end products due to the customer base. Carrying an exclusive mid-range product may indicate an opportunity for price skimming, but may not be as effective as price skimming high-end models due to the smaller price range across the market. Mid-range models should have at minimum a particular set of specs.  Mid-range models nearly always come with free shipping.

Recommendations: Low-End Models

Low-end models tend to be priced within a very small range across merchants. Low-end products are the best candidates for price matching or loss leading. Also, low-end models have a lot of market volatility, indicating that running sales for loss-leading, or adding value for psychological pricing may be good strategies. And low-end models should have at minimum a particular set of specs. Free shipping could add value to low-end models given that only about half of competitors offer it.


Because of the limitations of this data set and the lack of known product offerings by our fictional employer TVStore, no single pricing strategy could be determined likely to produce the most successful results.

Instead, each strategy offers potential benefits and risks that should be considered for any specific product depending on that product’s production cost, competition price range and pricing strategies, uniqueness, technical specifications, and perceived value.

This data provides insights into all of those topics, but the full dataset and additional information, such as TVStore’s target offerings, is needed to maximize TVStore’s profits while minimizing its risks.

The skills I demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.

About Authors

Sita Thomas

Knowledge is power. I leverage fierce curiosity and creativity to deliver immediate and impactful results for mission-driven companies by weaving together 20 years of statistics, engineering, and business development. I've worn many hats across a variety of industries,...
View all posts by Sita Thomas >

Lucas Kim

Born and raised in Brazil, Lucas graduated from the Embry-Riddle Aeronautical University and from Korea Advanced Institute of Science and Technology (KAIST) with a degree in aerospace engineering. Before joining the Bootcamp, Lucas worked in Finance in Brazil.
View all posts by Lucas Kim >

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