Market Basket Analysis (MBA)Superstore MBA Insights using Tableau

What can be inferred?• I purchase diapers

• I purchase a new car

• I purchase OTC cough medicine

• I purchase a prescription

medication

Market Basket AnalysisRetail – each customer purchases different set of

products, different quantities, different timesMBA uses this information to:

Identify who customers are (not by name)Understand why they make certain purchasesGain insight about its merchandise (products):

Fast and slow movers Products which are purchased together Products which might benefit from promotion

Take action: Store layouts Which products to put on specials, promote, coupons…

Combining all of this with a customer loyalty card it becomes even more valuable

Association RulesDM technique most closely allied with

Market Basket AnalysisAR can be automatically generated

AR represent patterns in the data without a specified target variable

Good example of undirected data miningWhether patterns make sense is up to

humanoids (us!)

Market Basket Analysis Drill-DownMBA is a set of techniques, Association Rules

being most common, that focus on point-of-sale (p-o-s) transaction data

3 types of market basket data (p-o-s data)CustomersOrders (basic purchase data)Items (merchandise/services purchased)

Typical Data Structure (Relational Database)

Lots of questions can be answeredAvg # of orders/customerAvg # unique items/orderAvg # of items/orderFor a product

What % of customers have purchased Avg # orders/customer include it Avg quantity of it purchased/order

Etc…Visualization is extremely helpful…next

slide

Transaction Data

Sales Order Characteristics

Association RulesAssociation rule types:

Actionable Rules – contain high-quality, actionable information

Trivial Rules – information already well-known by those familiar with the business

Inexplicable Rules – no explanation and do not suggest action

Trivial and Inexplicable Rules occur most often

Association Rules Affinity grouping or rule induction is one of the most popular data mining

techniques. This unsupervised learning technique works by mining through large databases to identify patterns. These patterns are hidden deep inside the data, creating a set of rules and assigning a measure of strength and likelihood of occurrence for the rules.

A related problem with association rules is that they provide a lot of redundant information. Most of the generated rules tell us what we already know – People who buy apples are likely to buy bananas too. Or any other fruit for that matter. But we already know this and so fruits are kept together and often promoted together. But sometimes the analysis can provide deeper information. Wal-Mart may already know about the association between candy and doll sales (both are liked by little children) but the analysis can identify which doll links the most.

Rules can also be used to associate time of day or the day of week with product sales. A convenience store finds that lottery sales peak during lunch hours while newspapers don’t sell much after 11 am. It decides to replace the newspaper rack with a second lottery rack every day before noon. Similarly weekend trends can be identified through association rules as well.

Using a Market Basket Analysis – Data used in TableauData

Superstore Retail DataIt provides transactional records

about customer buying products of different categories & sub-categories

Customers are diversified across geographies

Using a Market Basket Analysis – What insights can help in Tableau?

What are the product categories that are being sold along which product sub-categories? Can those categories & sub-categories be related together? Can there be promotional offers provides for such association?

Which specific region of customer are buying which set of items? Can there be more inventory place in those regions? Any offers for bulk buying in those region of customers?

What about Customer Segments buying items?

Having highest sales where?

LETS USE TABLEAU NOW!!!

http://kb.tableausoftware.com/articles/knowledgebase/market-basket-analysis