Sephora Skincare Analysis

Skincare and Budget

Background

I’ve spent the last 7 years diving and exploring skincare products. Before that, I used a maximum of two products: moisturizer and sunscreen. But now, I have close to 12 to 15 products, of course, I don’t use them all at once, however, they all have their own purpose and some of them are used depending on the season.

Since I have so many products and I might buy more in the future — the cost of good skincare products could get very expensive. Definitely, there were days when I bought a product that didn’t work for me.

Ok, I think that’s enough about the back story. Coming to the meat of this article. The reason I wanted to work on this project is that I wanted a product that people loved and did not cross my budget. There are three questions that I wanted answers for:

1. What products did people buy & love the most?
2. What category was mostly commonly purchased?
3. Most expensive products (top 10).

Python

First I gathered the skincare dataset from Kaggle and performed data cleaning and transformation using Python. So here, I wanted the top 10 expensive items available at Sephora. So if you look at the table below, we have a list of expensive products. Most of these products were under the ‘Treatment’ category.

top_10_exp = df.sort_values(by = ['flt_price', 'brand'], ascending = False)…