A toothbrush tied to a paste make up for a luring combo, for there’s a special offer for the taking. Affinity analysis is helping a toothpaste brand to perk up sales of toothbrush or stimulate demand by way of this promotion. The combo offers don’t end there. There are many more depending on customer buying patterns, purchase behavior and affinity towards products. Market basket analysis props up retailer spirits in the way it helps understand customer behavior and make personalized recommendations to drive sales of products.

The toothbrush and paste apart, as a retailer turns the lens of customer transactions, market basket analysis springs to life in the way it helps unearth products that occur frequently together in customer transactions and understand association between products by focusing on ‘product combinations’ in those transactions.

A snapshot of market basket analysis

Let’s dig into customer transaction database. Transactions revolve around bread, milk, sugar and rice. Needless to say, data also includes the Item ID, purchase date and the quantity purchased. Taking this transaction detail into account, unleashing market basket analysis on the referred datasets yield the following.

  • 70% of customers purchase bread, milk and sugar together, which points to co-occurrences
  • 55% of customers who brought bread and milk also bought sugar, which points to association rules
  • 50% of customers who purchase bread first also buy rice within two weeks, which alludes to sequential patterns

Market basket analysis yields better informed decisions

As retailers look beyond the POS data, there’s customer data spread across disparate sources enriching the POS data. The combo of POS and customer data serves retailers well in understanding the product affinity. By using transactional data residing in POS along with data across various customer touch points, retailers mine customer purchase patterns and preferences, make the most of ML algorithms and advanced analytics to evoke statistics in the form of support, confidence as well as lift to unearth product affinities lying untapped in a market basket.

In effect, customer purchase patterns lying hidden in mountains of data become a valuable feed to drive actionable insights which when disseminated to marketers and other retail stakeholders spark timely action to bring about desired business results.

Recommendation engine atop market basket analysis

With market basket analysis, retailers comprehend associations between items bought by customers. Jack is someone who puts meat in his basket while his shopping turn arrives. While the meat is in the basket, Jack also has the habit of putting snack items or beer into his basket. His buying history reveals his purchase pattern.

By having a recommendation engine built over Market basket analysis, retailers extract value from personalized recommendations and stimulate demand amongst customers. The recommendation engine, taking user taste, preferences and behavior into account, helps the retailer to suggest relevant items to customers such as David, stimulate demand and facilitate enriching user experience.

If there is a will to influence customer decisions, the way then is about reaching them at moments ripe for decisions. Affinity analysis entwined with recommendation engine come to the retailers’ aid in capturing decision moments, driving sales and promoting personalized user experiences.

Market basket analysis for optimized retail operations

Insights derived out of market basket analysis help retailers to augment:

  • Store Layout – Optimizing store layout by placing co-occurring products together
  • Targeted marketing – Leverage association rules applied to customer shopping baskets pave way for targeted marketing aimed at increasing customer spend
  • Product Recommendation – Recommend product based on customer purchase patterns
  • Inventory Optimization – Market basket analysis becomes the input for predicting future customer purchase, and with sales data, it becomes easy to maintain optimal stocks of products and items
  • Ad Optimization – Using market basket analysis, retailers want to make the most of predictable advertising enabled through knowledge gained from knowing how customers will respond to messaging and communications as well as offers

When an online retailer approached Saksoft to uncover insights around market baskets, association rule mining and K-Apriori algorithm made up for the core of market basket analysis and helped the retailer perk up promotion and pricing strategies, recommend better items to increase product sales and promote better customer experience.

Wondering how this technology suit you? Talk to us.