Data Based Customer Personality Analysis for Businesses
The skills we demoed here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.
Data Science Background
Data supports Customer Personality Analysis which is a detailed analysis of a company's ideal customers. It helps a business to better understand its customers and makes it easier for them to modify products according to the specific needs, behaviors, and concern of different types of customers. While this dataset covers multiple attributes to the sale of multiple products, I will be focusing on three attributes that I believe could have the most impact on a single product. The analysis is focused on age, education, and income and how they possibly correlate with the amount of money customers spend on wine over the last two years in this particular dataset.
What is the importance of age, education, and income customer attributes in determination of the purchase of wine over the last 2 years?
After incorporating my dataset and reading it as a data frame in Jupiter Notebook, I created scatterplots to determine the correlations between my variables. Since I did not use the entire dataset, I did not need to incorporate a correlation matrix to view all of the possible correlations.
First, it is shown that there is no correlation between age and the amount spent on wine unlike I originally assumed.
Second, I compared the level of education to the amount of money that a customer spends on wine. This scatterplot shows that the biggest buyers of wine are those with a graduate degree.
Finally, there is a positive correlation between a customers income and the amount of money spent on wine. For this variable, I generated a pandas profiling report which shows an analysis report of my data frame. It provided me with statistical information on the dataset and I placed a screenshot of the income variable, which gave a me a better understanding of not only the variable, but the dataset as a whole.
Based on the report and going back to the scatterplot, I was able to deduce that the largest customers of wine are those with an average income of around $70,000.
Data Conclusion and Future Work
To summarize my findings, I found that there is no correlation between year of birth and amount spent on wine, the biggest customers of wine are those with a graduate degree, and the biggest customers of wine are those with an average income of around $70,000.
As for future work, I would take an open ended approach to this analysis as I limited the number of attributes for this project. The amount of traits that can potentially have an effect on the sale of various items is extensive. Variables that could be looked into further include but are definitely not limited to the source of the purchase of wine, number of people per household, or the purchase of other items having an effect on each other.
I would like to explore all of these traits and look more into depth of the various correlations with not only the sale of wine but all of the other products included in the dataset as well; incorporating different algorithms, clustering techniques, and additional plots.