Marketing Analytics: Clustering Customer Data Using K-Means Algorithum | R Programming

marketing analytics
Marketing Analytics at WWW.PLANETANALYTICS.IN

Marketing Analytics is a collection of processes and technologies which enables a marketer increase revenue by collecting, measuring, and analysing the available data.

Importance of Marketing Analytics:

1. Revenue: The sole aim of marketing team in a company is to increase the revenue and analytics helps them achieve the same.

2. Forecast: Marketing analytics enables a marketing team to predict the future outcomes to take better decisions and formulate better strategies. 

3. Monitor Marketing Activities: Marketers launch various initiatives and programmes for various purposes like new product launch and promotions and analytics helps them monitor and analyse the success of these initiatives.

In this article we will try to solve a business problem using marketing analytics and analyse how we can implement a statistical tool to strengthen the marketing efforts and increase ROI.

Business Problem: An online men fashion retailers wants to cluster or form segments on the basis of available consumer data so that they can be targeted more efficiently instead of using the already existing practice of one promotional technique for all to reduce the marketing cost per target audience and increase the return on investment.


Download The Dataset

Understanding the dataset:

The dataset provided belongs to an online men fashion retailer and it includes data of customers who bought from the online portal last month.

UID: Unique ID of the customer.
City: Origin of the customer.
Age: Age of the customer,
CasualTop :Number of casual tops bought,
CasualBottom: Number of casual bottoms bought,
FormalTop: Number of formal tops bought,
FormalBottom: Number of formal bottoms bought,
CasualFoot: Number of casual footwear bought,
FormalFoot: Number of formal footwear bought,
Innerwear: Number of Inner-wear products bought,
PlusSize: Number of products bought from Plus size category,
Accessory: Number of accessories bought,
Traditional: Number of products bought from the category traditional.


#Set working directory
>setwd("/Users/planet analytics/Documents/RDirectory")


# Read File
>Data <- read.csv("onlinestoredataset.csv")


#Explore Variables
>names(Data)


#See the structure of the data
>str(Data)


#Check the summary
>summary(Data)


#Removinf UID
>CustomerData <- Data[-1]

#Training the model
>library(stats)
>set.seed(123)
>clusters <- kmeans(CustomerData[ ,3:12], 5)



#Explore Clusters
>clusters$size


>clusters$centers


If we focus on the Third cluster it is clearly visible that people in this cluster are interested in buying casual tops and casual bottoms hence it will be profitable for the company to target them first when they launch new products in these two categories. 
On the other hand cluster 5 is inclined towards buying products listed under plus size category and it will be a wise decision to target them for plus size promotions instead of targeting each and every customer which will help reduce the marketing cost and increase the revenue. 

Thanks for visiting Planet Analytics, Connect with us at info@planetanalytics.in

__________________________________________________________

Marketing Analytics | Perceptual Mapping