Eniac is an online marketplace specialising in Apple-compatible accessories. It was founded 10 years ago in Spain, and it has since grown and expanded.
Working as a data analyst, I analysed the data to help resolve the ongoing debate on whether or not it’s beneficial to discount products.
-
The Marketing Team Lead is convinced that offering discounts is beneficial in the long run. She believes discounts improve customer acquisition, satisfaction and retention, and allow the company to grow.
-
The main investors in the Board are worried about offering aggressive discounts. They have pointed out how the company’s recent quarterly results showed an increase in orders placed, but a decrease in the total revenue. They prefer that the company positions itself in the quality segment, rather than competing to offer the lowest prices in the market.
- Python
- Pandas
- Matplotlip
- Seaborn
2.1.1 Eniac: Product Categories
- Data Analysis based on completed orders only.
- Classified the products into 19 categories Top 5 Categories
- Desktop - €3.0 mil
- Other/ Misc - €2.9 mil
- Data Storage - €2.8 mil
- Laptop - €2.5 mil
- Monitor - €0.8 mil
- A limitation of the data is that due to time constraints we have 1044 items that we were unable to manually categorise so we added them to a Other/Misc category.
2.1.2 Eniac: Product Pricing Architecture
- Average price of all products just under 500 euros, with a min and 2.99 euros and the max at just over 15,000 euros
2.1.3 Eniac: Discounts
- 93.1% of all units sold were discounted (by 20.8 % on average) representing 95.0% of total revenue.
- High discounts do not directly result in high revenues, indicating they are not applied to high-value-products.
General Effects of high discounts on low-value-products hard to assess:
- No data regarding stock-reduction
- No data on customers
- No data on shop-traffic
2.1.4 Eniac: Seasonality
- Reviewing total sales by month we can see strong peaks for black friday and strong sales in December for christmas
- As we have discounts all year round it looks like these peaks are driven by seasonality and not mainly by discounts
- However looking at comparing Q1 sales for 2017 vs 2018 we can see that we have increased sales
- This could be due to increased traffic over Black Friday and Xmas increasing market awareness
- With the data available I am not able to conclude if this is due to promotions
2.2.1 Limitations/ Data Collection Improvements
Unaware of the Company Goals:
- Build Market Awareness E.g. Is there a need to drive increase traffic to site to build brand awareness?
- Are they a new company? Trying to acquire the customer base
Missing Information:
-
Profits / Cost Price / Margins
- Although we can see peak in revenue we are unaware on how discounts are impacting company profits.
-
Customer Information
- We do not have customer information therefore we cannot see if they are returning or being driven to the site with discounts.
Data Collection Improvements:
- Prices : To be rejected if entered incorrectly (with two decimals)
- Remove/ Reject Duplicates: When updating product-portfolio
- Datetime-Format: Automatically import date-inputs as datetime-format
- Column Names: Columns to have the same name if they contain the same information
- Database: Data to be in a Database not in 4 CSV files
2.2.2 It is Beneficial to Discount Products?
After cleaning and reviewing the dataset provided I am unable to advice if it is beneficial to discount products due to limitations in the data:
- Peaks in revenue are not mainly driven by discounts.
- Assessing the beneficiality is difficult because we do not know the overall strategy of the company.
- We are unaware of the impacts discounting is having on company profits.
- Since the dataset shows discounts over the whole time frame it’s not possible to assess what would happen if we were not discounting or only discounting on special occasions.