eCommerce-Customer Segmentation Analysis

This 4 following months, I learn one of the most challenging yet also exciting data analyst vital programming languages is Phyton spesifically using Google Collab. As my practice i took case from one of unicorn tech company to hone my skills in phyton skills as my current job that required much data in analysis. This is not coding source, this article more describe the comman analysis, if you need for the code you can get here.

Gustovo just started an e-commerce startup based in Portugal that recently opened on online website to sell their product. Fortunately, Gustavo is launching their website when the covid-19 hits and making them grow faster than ever. However, Gustavo is still not using targeted marketing which hurts their marketing budget as only a fraction of their user comes back to their website. Gustavo needs your help to increase their marketing conversion rate by doing more targeted marketing using customer segmentation so that it will not hurt their budget.

EDA for Bussiness

Objectives:

• Summarize the main characteristics, started by using any statistics descriptive that show by statistical graphics, and other data visualization methods
• The goals is to get model such as: Number of orders/month, Daily orders trend, and percentage of orders per payment type

Evaluate the Descriptive Statistic:

Insight from Descriptive Statistics :

Based on Gustavo cases, smallest payment value is 0 USD, and the maximum is 344.340 USD. Meanwhile the mean is 110.010 USD. From the percentile 75% (P75) that the payment value is 150.460 USD. In more precise way, we can conclude that 25% of payment value is more than 150.460 USD.

When viewed from the kurtosis, which shows the number 0.50, it indicates that the peak distribution of the data is not too high (value less than 3). and Skewness shows a positive number of 1 indicating that the positive skewness (tail long to the right), so have more payment value in small amounts when compared to large ones.

EDA for Bussiness

1. How many number of order from Gustafo company

From Gustafo company we want to know how many number of order from the latest 2 years. The chart we can see below.

From year 2016 to 2018, the number of orders trend was increased.

2. What busiest day of the week

The busiest day of the week is monday by total 15102 orders, and the least is saturday by total 10166 orders. From this we can turn the slowest day into the most popular day of the week

Recommendation

Marketing team should maintain quality performances in Monday and can make powerfull marketing strategies to grow bussiness faster in another day.

Some alternative marketing strategies:

- Put together a fundraising event for a good cause in busiest day. Customers respond to brands with heart.

- Market to your existing customers and earn more of their business

- Hold a networking event in busiest day-

3. What is Customer’s favourite payment type

Gustavo’s customer the most favourite using credit card as payment type dominate the proportion of 59.2%. and the least is debit card which proportion of 3.8%.

Recommendation:

Gustavo marketing team can provide discount for customer that using credit card and make interesting program for other payment type.

Promo marketing ideas that innovate:

- Get creative with ad visuals

- Whether launching brand or the newest product, launch it big.

- Give exclusive discounts and offers to followers

- Build hype for an announcement or new product with a countdown. By drawing out the hype, you increase the chances of people noticing.

Clustering & Customer Segmentation

Suppose that we have a company that selling some of the product, and you want to know how well does the selling performance of the product. You have the data that can we analyze, but what kind of analysis that we can do Well, we can segment customers based on their buying behaviour on the market.

Keep in mind that the data is really huge, and we can not analyze it using a bare eye. We have to use machine learning algorithms and the power of computing for it. This article will show you how to cluster customers on segments based on their behaviour using the K-Means algorithm in Python. For code source I prepare this for you.

Before we get into the process, I will give you a brief on what kind of steps we will get.

• Manage Skewness

The data variable is still skew to unskew the data using log transformation

The variable is not skew anymore

Then we use a scaler for scikit-learn library: standarscaller to scale our library. After that we determine cluster number using elbow method.

Form the elbow method we can determine the optimal number of K is 4

• Customer Segmentation

****

Recommendation

In this analysis I used Behavioral segmentation that are :

Spending Habits and Interactions with the brand. Behavioral segmentation is possibly the most useful of all for e-commerce businesses. As with psychographic segmentation, it requires a little data to be truly effective — but much of this can be gathered via your website itself. Here we group customers with regards to their: Spending habits, Purchasing habits, Browsing habits, Interactions with the brand, Loyalty to brand, Previous product feedback.

I hope that this article will help you on how to do customer segmentation step-by-step from preparing the data to cluster it.