The document is structured with the following sections:
As I begin my career in CPGs, I wanted to focus this project on digging into some data from the industry. While I would have liked to look into POS and consumer behavior data, that is something I don’t currently have access to (nor have the funds to obtain). Instead, I found the Largest Food & Packaged Good Companies 2016 data set and will focus on the top companies’ financial growth and wins.
Data for this project is imported from https://myxavier-my.sharepoint.com/:x:/g/personal/rivaj_xavier_edu/ETX1PUd2GmFBqU3AuxZlv6YBw7vb-WazcBO6yytjNk_NlA?download=1. After my cleaning and wrangling, a total of 36 data entries and 7 variables were used of for final analysis. 2 new columns were introduced to code more efficiently (see Dummy Variables for further explanation).
The packages required for this markdown are:
| Package | Summary |
|---|---|
| knitr | RMarkdown documents |
| rmdformats | RMarkdown themes |
| tidyverse | The tidyverse collection of packages all together |
| DT | Making pretty javascript data tables |
| stargazer | Making pretty summary statistics tables |
| corrplot | Various correlation plots |
| PerformanceAnalytics | Building awesome graphics for analytics |
| rvest | Useful tools for working with HTML and XML |
| dplyr | Used for data manipulation |
| pander | Creates summary tables for Markdown |
| scales | Used for editting labels on visuals |
| syuzhet | Provides function that helps with Sentiment Analysis |
| SentimentAnalysis | Perform Sentiment Analysis |
| wordcloud | Create word cloud visual |
Dummy Variables:
Dummy variables were created for these variables ease analysis: - Category -> CatDummy - Profit at a % of Revenue -> PPR_Dummy
Below is a table to reference as it explains each variable of the dataset:
| VARIABLE NAME | NAME OF DATA ELEMENT |
|---|---|
| Company | Name of Company |
| Revenues | Total Revenue in 2016 |
| Profit | Total Profit in 2016 |
| Profit_Perc_Rev | Profit at a % of Revenue |
| Category | Category Company Plays In |
| CatDummy | Dummy Variable for Category |
| PPR_Dummy | Dummy Variable for Profit at a % of Revenue |
Companies by Category
| Category | Company |
|---|---|
| Beverage | Coca-Cola |
| Constellation Brands | |
| Dr. Pepper Snapple | |
| Food | Campell Soup |
| Conagra Brands | |
| Dean Foods | |
| General Mills | |
| Hershey | |
| Hormel Foods | |
| J.M. Smucker | |
| Kelloggs | |
| Kraft Heinz | |
| Land O’ Lakes | |
| Mondelez International | |
| Pepsico | |
| Treehouse Foods | |
| Household & Personal | Avon Products |
| Care | Clorox |
| Colgate-Palmolive | |
| Estee Lauder | |
| HRG Group | |
| Kimberly-Clark | |
| Procter & Gamble | |
| Pharmaceuticals | Addvie |
| Amgen | |
| Biogen | |
| Bristol-Myers Squibb | |
| Celgene | |
| Eli Lilly | |
| Gilead Sciences | |
| Johnson & Johnson | |
| Merck | |
| Pfizer | |
| Tobacco | Altria Group |
| Philip Morris International | |
| Reynolds American |
Stats on Each Category
2016 Revenues for Top 36 CPG Companies
2016 Profit at Percent of Revenue for Top 26 CPG Companies
Comparing Categories by Revenue
Comparing Categories by Profit
Is there a relationship between the a company’s revenue and profit?
Looking at the graph, the positive relationship depicted validates that when revenue increases so will profit.
Because JnJ has brands in its portfolio that fall into personal care, how does JnJ’s revenue compare with those of household & personal care companies?
It was intersting to see that JnJ just topped P&G while playing in two different categories.
For my predictive analysis, I chose to use an ANOVA test to see if there campanies’ play a role in their revenue and proft.
for Revenue:
## Df Sum Sq Mean Sq F value Pr(>F)
## CatDummy 4 1.208e+09 301917045 0.897 0.478
## Residuals 31 1.044e+10 336716696
The P value is high indicating that our results are not significant, thus meaning that the category of a company does not play a role in their revenue outcome.
for Profit:
## Df Sum Sq Mean Sq F value Pr(>F)
## CatDummy 4 286199373 71549843 5.122 0.00275 **
## Residuals 31 433046108 13969229
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
The P value is low (<.01) indicating that our results are significant, thus meaning that the category of a company does play a role in their profit outcome.
Top Products Launches of 2016: Do any of the top product launches of 2016 match up belong to the top CPG companies?
Top 10 Food Product Launches
## [1] "1. DairyPure, $1,163.1"
## [2] "2. Dunkin’ Donuts K-Cups, $204.1"
## [3] "3. Not Your Father’s Root Beer, $114.6"
## [4] "4. Oreo Thins, $110.2"
## [5] "5. Artesano, $102.4"
## [6] "6. Screamin’ Sicilian, $73.2"
## [7] "7. Oscar Mayer Natural, $61.7"
## [8] "8. DairyPure Creamers, $54.9"
## [9] "9. Sargento Balanced Breaks. $54.2"
## [10] "10. Henry’s Hard Soda, $50.3"
Takeaway: 40% of the most successful product launches of 2016 came from the top 36 CPG brands.
Top 10 Non-Food Product Launches
## [1] "1. Flonase, $316.5"
## [2] "2. Gillette Fusion ProShield, $144.9"
## [3] "3. Garnier SkinActive, $117.5"
## [4] "4. Tide Pods Plus Febreze, $87.3"
## [5] "5. Crest Pro-Health Advanced, $80.1"
## [6] "6. The Pioneer Woman Collection, $79.1"
## [7] "7. Persil ProClean, $63.2"
## [8] "8. Cesar Home Delights, $45.2"
## [9] "9. Tampax Pocket Pearl, $41.9"
## [10] "10. Old Spice Fresher Collection, $39.0"
Takeaway: 60% of the most successful product launches of 2016 came from the top 36 CPG brands, and 5 of those 6 were P&G brands.
JnJ and P&G in 2021
Sentiment Analysis
Moving into the qualitative data, I wanted to better understand the sentiment of the words JnJ and P&G they use in their tweets.
This analysis interested me because JnJ and P&G both shared ‘positive’ and ‘trust’ as their top 2 scores. Their third sentiment differed, and it showcases how they differ in the sentiments they market to their audiences. While the two companies are similar, it is clear in P&G’s marketing campaigns that they try to capture joy and JnJ has a heightened focus on anticipation as they prioritize their medical innovation.
Word Cloud