Finding What to Sell Next to your Customer

You need what marketers and data miners call – Market Basket Analysis.

My last blog described finding similar-looking customers using Cluster segmentation. This blog will provide an introduction to Market Basket Analysis.

Why care about DBSync talk about this concept. We believe that centralizing data from CRM, Social & Accounting can provide data for mining the right upsell and cross-sell opportunities.

Introduction

Market Basket Analysis (MBA) is the search for meaningful associations in customer data. Basically Market Basket Analysis (MBA), sometimes called affinity analysis, is a method used to identify items that will probably be purchased together. Many medium to large companies utilizing Customer Relationship Management (CRM) systems and e-commerce sites have accumulated large volumes of data known as “big data” that is used by MBA processing to develop successful sales and marketing campaigns.  Retailers like Target, and Walmart, use market basket analysis to sell product lines to specific audiences.  The FactPoint Group, in their report entitled How Top Retailers are Using Market Basket Analysis to Win Margin and Market Share, ascertained that the majority of the retailers they interviewed for the report were familiar with market basket analysis  “and looking to expand their capabilities in that area”.

Today ability to predict a customer’s purchase habits has become so important to the success of a retail enterprise that software applications, (systems), like Salesforce, Microsoft Great Plains and CRM accounting applications like Quickbooks, have all promoted the fact that their product has the ability to do MBA processing. Why, because as retailers they understand that by using data mining and MBA predictive modeling, they could obtain a competitive edge over their competition.

Market Basket Analysis – How it Works

Market Basket Analysis (MBA) is a modeling technique that is based on the assumption that if a customer buys a certain group of items, he or she is more (or less) likely to buy another group of items.

MBA is an advanced predictive model that is also used to identify events that generally occur in sequence. Using a predictive market basket analysis, Amazon can figure out what a returning online customer will buy, Google can predict what results to display to a web suffer and Netflix can determine what movies or television shows customer will rent.  MBA generally uses the Apriori algorithm to process the data.

(For those who want to go in a bit deeper into the math behind … )

The Apriori algorithm is simply:

Find the frequent itemsets: the sets of items that have

minimum support

– A subset of a frequent itemset must also be a

frequent itemset

• i.e., if {AB} is a frequent itemset, both {A} and {B}

should be a frequent itemset

– Iteratively find frequent itemsets with cardinality

from 1 to k (k-itemset)

• Use the frequent itemsets to generate association rules[ii]

Retailers like Target, and Wal-Mart, use predictive analysis to sell products lines to targeted audiences. In today’s international business world MBA has become so important to the success of an enterprise that CRM software applications like SalesForce, Microsoft Great Plains and CRM accounting interface systems, all advertise the predictive modeling analytic ability of their products.

Market Basket Analysis – Benefits

The results of MBA modeling can be used by enterprises to:

• ascertain  information that can be used to develop effective cross-selling and upselling strategies to increase revenue
• develop effective sales and marketing campaigns
• improve customer service
• aid retail  management with   layouts and endcaps
• help a company manage inventory

In addition the use of MBA makes it possible for companies to process big data to identify “outliers”  “often re-ferred to as outliers, anomalies, discordant observations, exceptions, faults, defects, aberrations, noise, errors, damage, surprise, novelty, peculiarities or contaminants.” Outlier data is used in fraud detection, to identify cyber security fraud and military surveillance.

Summary

Retail use of MBA predictive analytics for corporations has increased because of the need to control business practices to increase revenues during the current recession. CRM & Accounting applications enable enterprises to accumulate the data it needs to effectively process sales and customer information. However MBA predictive analytics data mining uses that historical data to give company management the ability to effectively plan for the future.  This is important because more and more companies rely on current customers using relationship marketing rather than mass marketing, as the model to increase sales for their enterprises.