‘Who is our customer’?

This Analysis lays down some points which we can look at while trying to answer a popular question from any brand i.e. “Who is our customer?”  Below mentioned are some of the key areas that we usually analyze to answer such question:

1. Customer Demographics: Age, gender and region etc.

2. Product purchase behavior.

3. Discounts availed by customers.

4. Channel of purchase: Offline or Online.

5. Loyalty Program:

a. Redeemer/Non Redeemer breakup.

b. Points redemption behavior.

6. Seasonality Pattern:

a. Festive/Non Festive Buyers.

b. EOSS/NON EOSS buyers.

Outcome of this Analysis

  • Understand who your customers really are.
  • Target specific groups based on their customer personas
  • Get to know the favourite products for each gender as well as the age group.
  • Understand the product seasonality.
  • To get an understanding of the points redemption in a detailed way and answer questions such as:
  • Which age group customers have the highest redemptions?
  • How many points are getting redeemed in a particular transaction from customers in different age groups?
  • To analyze whether discounts are triggering repeat purchases.
  • Understand the shopping pattern of customers across different seasons in a year.
  • Which age group is more likely to transact in what season of the year?
  • Which LTV group is more likely to transact in what season of the year?

Notebook:

The purpose of this document is to guide the end user with steps to run the notebook and get the insights from the output.

Input fields:

1. Org id:

Enter the org_id of the brand for which analysis needs to be done.


2. Product Database:

Enter the name of the database in which inventory table is present.


3. Product Table:

Enter the name of the inventory table that has product data.


4. Inventory Column:

Enter the name of the product column at the level of which analysis needs to be done.

For eg: Product Category, Brand etc.


5. Product Code Column:

Enter the column that captures ean code in the product master table.

Please note that this column must have same entries that are present in item_code column of bill_lineitems table.

Notebook Flow:

 Customers’ breakup by Age and Gender:


The above chart can help the user in understanding the distribution of the customers across different age segments for the brand.

We can check which gender prefer the products from which categories.

Female Product Seasonality:

Male Product Seasonality:

Female product preference by Age group:

In the table below, The average transactions for dresses from each age group is ~0.7% but for <=23 years age group customers it is ~2.3% indicating that younger people are more likely to buy dresses. Same way insights can be drawn for other product categories.

Male product preference by Age group:

Product Preference by recency and frequency groups:

Each row adds up to 100%, the green cells indicate the recency and frequency which have a comparatively higher proportion of transactions for a particular product.

Preferred redeeming range by different age groups:

The output of the above table can be downloaded from the notebook and user can create a view like below to get some insights out of it:


pointsband
LT23BT24_40BT41_55GT56
[0.0, 200.0)

21.0%

18.2%18%24%
[200.0, 400.0)

23.0%

21.7%24%25%
[400.0, 600.0)

26.5%

26.7%26%24%
[600.0, 800.0)

9.7%

11.1%

13%

8%
[800.0, 1000.0)

7.3%

8.2%7%7%
[1000.0, 1200.0)

4.2%

5.4%4%5%
[1200.0, 1400.0)

3.2%

3.2%4%5%
[1400.0, 1600.0)

1.9%

2.5%2%2%
[1600.0, 1800.0)

1.8%

1.6%1%1%
[1800.0, 2000.0)

1.5%

1.4%1%1%

24 to 40 year old age group prefer to allocate minimum 400 points before redeeming, whereas greater than 56 year old age group customers redeem as soon as they have some points.

Discount trend by different frequency groups:

We can also verify the hypothesis that the ones who are not availing any discounts are one timers.

Seasonality Trend:

By Age Group:

We can determine which Age groups usually shop during which festivals and which are the ones where higher proportion of transactions can be seen for non-festive period. Each column adds up to 100% and the green highlighted cells indicate which age groups are likely to visit more during which period.

By LTV:

We can divide the transacted base into different segments based on their life time spend and then can see their shopping patterns during Festive and Non Festive period. Table can be read in the same way as explained for Age group case.

NotebooksCluster links
Customer Persona AnalysisIndiaSEAEMEA