Introduction

Why this report is useful:

  • This report analyses which coupon campaigns have been the most successful and least successful. From there, we determine what demographics and what products are using coupons the most.

  • Within demographics of CompleteJourney data, we discern the household Id, the amount of people living in that house, the age of those people, a persons marital status, and the income per household. This provided a lot of information on the types of people, that being part of a family or alone, and what products they buy with what coupons.

  • Within transactions and coupons of CompleteJourney data, we focused on household ID, store ID, product ID, campaign ID, sales values, and the coupon discounts. This provided insight into how well certain coupon campaigns are doing and what campaigns need to be adjusted for more sales. It also shows which products are often bought with coupons, defining the best and worst matches between individual coupons and products.

  • We chose to handle this problem by first joining transactions with demographics and coupons. We joined transactions to coup by product ID, and to demographics by household ID. Moving forward involved dissecting the demographics data for specifics such as wether a person is married and/or owns a house. We grouped the joined data sets in various ways to form graphs, and we made sure to create a total sales value using a summary function.

Required Packages:

CompleteJourney

  • This package provides all information about the Regork company for analysis.

RcolorBrewer

  • This package provides various color pallets to graphs for better readability and understanding of our analysis.

Tidyverse

  • This package provides a collection of useful data analytics packages regarding design philosophy, grammar, and data structures.
# load data sets
suppressWarnings(suppressMessages(library(completejourney)))
suppressWarnings(suppressMessages(library(tidyverse)))
suppressWarnings(suppressMessages(library(knitr)))
# load data frames
trans <- get_transactions()
prom <- get_promotions()
coup <- coupons
dem <- demographics
glimpse(prom)
glimpse(trans)
glimpse(dem)

The Analysis:

Potential Implications to Consumers:

  • Personalization: create a more tailored shopping experience and make consumers feel more valued and understood

  • Limited Options: If certain coupon campaigns are more successful than others, retailers may limit the products and brands that are included in future campaigns. This can limit the options available to consumers and make it more difficult for them to find deals on products they are interested in.

Limitations of Analysis:

There are some limitations to the analysis mostly due to the specificity of the data in the completejourney dataset. One of the most pronounced limitations was the age section of the data, where anybody over the age of 65 was lumped into the same bin. It would have been interesting to have been able to dive deeper into specific age ranges for older people.

Interesting Insights:

Through wrangling the data in the completejourney dataset, we came across several interesting insights that we though could help with targeting coupons to new audiences.

One of the interesting points that we saw was that in households containing no children, and in households with income ranges in the lower brackets, coupons pertaining to frozen foods and dinners were particularly successful, while losing traction when entering into households with more children and a higher income bracket.

Suggestions Moving Forward:

Building off of the insights that we found in the data, It seems that there is a large market for frozen dinners that drops off at a demographic of higher-earning individuals with children. Something that might help bring those groups into the market would be to advertise more large, family-style frozen dinners, or frozen meals for kids to eat when they go to school.