Over the last couple years, drugstore cosmetic products have been putting high-end brand products to the test. With the rise of social media influence on product trends, drugstore cosmetic brands have been taking advantage of this influence by starting to create what are referred to as “dupes,” or similar products to high-end ones, and showcasing them compared to the expensive products. Particularly throughout 2023, drugstore brands released multiple dupes of high-end products, making the comparison of them a new, popular social media trend.
As a company that sells a variety of drugstore cosmetic brands, these new trends pose the question of how can we take advantage of the newfound popularity of their products to optimize our sales in the cosmetics department?
By analyzing Regork customer shopping patterns, we can assess what kinds of ways we can optimize sales in the cosmetics department, as well as strategizing their placement across stores to increase sales in other departments. Along with product analysis, it is important to identify target groups to market to in order for the strategy to be effective. By identifying our top selling cosmetic brands and the demographics of frequent purchasers, we have been able to draw conclusions on who we are trying to sell these products to and how we can advantageously place these products in stores.
We are confident that our findings will be able to maximize sales in the cosmetics department and boost the sales of products they are frequently purchased with.
The following R packages are required to run the code for this analysis:
Package definitions adapted from “Data Wrangling With R” by Bradley Boehmke, https://bradleyboehmke.github.io/uc-bana-7025/
The CompleteJourney data is a collection of 2,469 unique households and their grocery store transactions from January 2017-December 2017. Information includes, but is not limited to, demographics, transactions, products, coupons, etc.
CompleteJourney was developed by data scientists at 84.51 for educational purposes. More information on it can be found here.
Of the package’s eight datasets, the three primary ones utilized for our analysis are Products, Transactions, and Demographics.The details of each dataset are below:
Source of tables and table descriptions: The Complete Journey User Guide
After assessing what datasets need to be utilized for the analysis, the following variables were prepared to be incorporated into our visuals.
cos_makeup_data <- products %>%
filter(department == "COSMETICS") %>%
filter(product_category == "MAKEUP AND TREATMENT")
cosmetic_basket_id <- cos_makeup_data %>%
filter(department == "COSMETICS") %>%
inner_join(transactions, by = "product_id") %>%
distinct(basket_id)
non_cos_prod <- products %>%
filter(department != "COSMETICS")
baskets_cos_noncos <- transactions %>%
semi_join(cosmetic_basket_id) %>%
inner_join(non_cos_prod)
As previously discussed, it is initially important to identify what the top selling cosmetic brands are at Regork. By filtering all transactions data to specifically display cosmetics department information, we are able to visualize the brands that generate the most sales:
From this visual, we are able to observe our top 5 selling brands. It is evident that Covergirl and Maybelline are the top two brands, with Revlon, L’Oreal, and Almay respectively trailing the two.
Now that we have discovered our top brands, we are also interested to
see what kind of demographic groups are purchasing from the cosmetics
department the most. Below are the age groups and income ranges in
relation to cosmetics sales:
After examining these plots, we can note that cosmetics sales are highest in the 35K-74K income ranges and within 35-54 years of age. Let’s take a look at the sales of those two demographics individually relative to our top selling cosmetics brands:
Now that we have identified who generates the most sales in the cosmetics department, we are able to use this information to identify what they frequently purchase with cosmetics products to make our final conclusions:
From this visual, we are able to identify that the top products most frequently purchased with cosmetics are:
In the next section, we will utilize this information to make recommendations on how we can optimize cosmetics sales and increase sales in other departments.
The Goal:
As aforementioned, drugstore cosmetics popularity is currently climbing. The objective of this analysis is to identify how Regork can untilize customer data on cosmetics purchases to assess who we should market these products to and where in the store it may be optimal to advertise them.
The Analysis
To address this concern, first, Regork’s top selling cosmetics brands, Covergirl, Maybelline, and Revlon were identified. Second, the income ranges and age groups that generate the largest amount of sales for the cosmetics department were drawn. The top-purchasing income ranges (35K-74K) and age groups (35-54) were examined separately in comparison to the top brands, making it evident that Covergirl generates the most sales overall, with people 45-54 years old in the 50K-74K income range produce the most sales for the brand.
After looking at the baskets of cosmetics buyers in our top income and age ranges, the top products purchased with costmetics were able to be identified, listed under the Exploratory Data Analysis sectiion.
Interesting Insights:
In our target income and age ranges, it is intriguing that Covergirl is the top selling brand except within the 35-44 age range, where it falls second. It is also notable that cosmetic sales are drastically higher in our target demographics compared to the others. While our demographic trends are clear, there are no obvious correlations between cosmetics purchases and other departments.
Proposal:
As a result of our analysis, our first recommendation would be to stock a variety of Covergirl, Maybelline, and Revlon products. Since these have been identified as the top selling brands in Regork cosmetics departments, it is important to have a range of their product lines so that those who are loyal to those brands can have a one-stop location for obtaining their products.
Our second recommendation would be to highlight cosmetic displays frequented by 45-54 year olds. From the visuals under the Exploratory Data Analysis, is is noticeable that this particular age group generates the largest sales for the cosmetics department. To target this group, it may be beneficial to display cosmetics suitable for mature skin. Additionally, since milk, soft drinks, and alcohol are frequently purchased with cosmetics products, it also may be advantageous to place some displays or advertisements near these sections.
Our final recommendation would be to provide coupons and discounts for the popular cosmetics brands. Since cosmetics typically have a common retail price across all stores they can be found in, it may be helpful to send out coupons for these products across all demographics so that Regork’s cosmetics department may be more attractive than another grocery chain’s.
Limitations of Analysis:
It could be argued that our data is outdated (2017) and unable to accurately predict 2024 trends. However, we feel that our strategy will still have a positive impact due to the drastic increase in popularity of drugstore cosmetics since 2017. Additionally, the dataset did not allow us to identify specific product lines, only brands. Due to this, we were unable to identify what products from even the most popular brands actually sell, but along with our second recommendation, we would like to think that any product line can be sold with adequate advertisement.