Customer Analytics

Every business would like to know their customers. The key question is that “how well do you know your customers?” Do you know “What products they buy from you”? Do you think that they may buy from you additional other products? Are there any patterns in their purchase behaviors? Such as buying some products together. Are you flexible enough to release campaign to your  customers that have similarities or relevancy for offers you may provide them” what about their actions to past campaigns?

Customer Analytics can be used to determine the best offer to present to each customer by analyzing historical purchase data to:

  • Identify customer segments for targeted marketing based on their past purchasing behavior
  • Identify products that are likely to be bought by the same customer
  • Understand customers’ propensity to buy products from a different category

When used with historical purchase data that include purchases made during a campaign/promotion period, an analysis can also produce a predictive model to predict a customer’s response to a future campaign.

Our customer analytics approach is a series of analysis templates used for analyzing customers’ purchase behavior. These include the following structural approaches:

  • How to prepare proper data for customer analytics
  • Segmentation Analysis
  • Propensity Analysis
  • Affinity Analysis
  • Real Time Propensity Analysis

 Segmentation Analysis

Your customers’ purchasing behavior tells  you something so that you should be able to group your customers based on their past purchasing behavior to learn more about them. That is what we address in our approach.

clustering algorithm should be first algorithm  to apply for determination of customer segments and learn about the characteristics of each customers.

Clustering is a technique used to group similar objects together. It is up to the person performing the cluster analysis to define what measures to use to determine similarity. For example, a marketing analyst may be interested to group her customers by the similarity of the quantity of products purchased in various categories. Clustering would then result in groups whose members have made similar quantities of purchases in similar categories.

Our segmentation analysis lets you select the product categories to use for generating the segments, specifying the number of segments to identify and let the template’s clustering algorithm to work for you. You should be able to have following questions’ answers;

  • Which product categories are the most important in determining the segments?
  • Which segment is the biggest or smallest?
  • Who buys the most?
  • Who buys what?

Propensity Analysis

Propensity analysis offers some common approaches to find answers the questions such as ;

  • Which customers should we target in the campaign and with which products?
  • How do we know when to promote a product and when not to in order to keep the offers personalized?

In addition to our customer analytics; our propensity approach enable you  to perform analyses like;

  • how we can conduct an analysis to determine products to recommend to customers,
  • identify cross-sell opportunities and make personalized offers.

When conducting a propensity analysis, you can easily find the propensity for a customer to perform a certain action like buying a product. In short, we help you to predict how likely it is that a customer will perform such a action. Depending on cases, we use several algorithms like decision trees, random forests and logistic regression, may be used to perform this prediction.

Affinity Analysis 

Our Affinity analysis approach enable you to find product categories that are often bought together by customers. Together with our other analytics, we help you perform an affinity analysis and identify high-affinity product categories from historical purchases made by customers.

We provide you  a sample set of fictitious data that is created such that the data contains strong purchase patterns for certain categories of products. When categories frequently bought together are found, we can use them to drive further propensity analysis for cross-selling and make recommendations to our customers.

We use use an algorithm to generate an affinity score that measures how often two products are purchased together by the same customer. The algorithm uses a measure called the Jaccard index for the affinity score. It measures the similarity between purchase history of two product categories. The higher the score, the more similar the purchase history of the two categories, and thus the more often the categories are purchased together.

Our affinity analysis approach lets us select the product categories we want to analyze and generates the affinity scores for every pair of product categories within those selected, suing 4 visualizations  used in conjunction with each other to help us identify high-affinity products.