Scaling ‘nudges’ across the retail ecosystem with Analytics and Machine Learning


Share on LinkedIn

“Choice Architecture” is one phrase that has stuck with me after reading “Nudge: Improving decisions about Health, Wealth and Happiness” by Richard Thaler and Cass Sunstein. The book describes people as being irrational and most of the times, not the best decision-makers for themselves, their family, and for society. To help them in the process, businesses, organizations, policymakers, and governments should organize choices in a way that it nudges or influences individuals to select the desired decision or outcome. There are several successful, as well as poorly designed, choice architectures that span across different industries and the world today.

Behavioral economics and the concepts of choice architecture are not new concepts to the retail world. Retailers across the globe have been leveraging these concepts through techniques that are demonstrated in online shopping and advertising.

Additionally, brick and mortar shopping also has a lot of opportunities to present nudges especially involving in-store associates. Displaying labels on products such as “best seller” or after viewing a product online, including a section called “customers also liked” to refer other products give customers a peer validation and a sense of trust in that product to eventually nudge them to make a decision.

We are also observing a direct relationship between a retailer’s value proposition to their customers and the way the choices are presented. Some websites present choices for different fulfillment options, next day delivery etc., whereas others presents discounts-based choices. All these also mean that there is a larger collaboration at the back-end between retailers-product vendors, retailers-logistics providers that helps retailers in providing these choices to their customers.

Scaling “Nudges” in the era of Machine Learning

Recent developments in the field of Analytics and Machine learning have helped retailers solve the common issue of marketing to a general audience. Several retailers prioritize customers based on their value, their preferred engagement option, so on and so forth. While these have helped retailers in getting better responses to marketing campaigns, promotions and markdowns etc., these are few of many touchpoints in order to “nudge” customers to select the preferred outcome.

There are always risks involved in the “nudging” process that needs to be considered. After applying a general model, retailers should assess and learn from the customer’s behavior to then influence them toward the desired outcome. For example, let’s say a customer named Roger typically selects a direct-to-consumer type of fulfillment during the holiday season. Retailers should consider building a model tailored to Roger’s needs and user experience as well as understands its impact on the organizational goal. It will then nudge Roger towards selecting the buy online and pick up in-store option. By presenting that choice with incentives, this is known as, “precision nudging.” This concept uses the lessons learned from behavioral economics and scales it to an entire customer base through machine learning.

There is a thin line that differentiates personalization and nudges. While personalized content and campaigns are derived from a customer’s previous interactions, nudge models should take that and align it with the customer’s and the organization’s goal. For example, a popular American pharmacy chain offered loyalty points for every step their customer walked. This promoted healthy behavior among their customers and helped the brand in converting loyalty points into potential sales. Although loyalty programs are not a new and innovative concept, there is one outcome irrespective of the variance in customer base; it is a good example of how the pharmacy chain aligned their organizational goal with the customer’s wellbeing and nudged them towards mutual success.

The potential opportunities for retailers and technologists in the field of “Behavioral Economics and Machine Learning” have enormous value. And now is the time to employ the “choice architecture” on our decision and nudge others toward the best ones that solve the organization’s goals.

Sivakumar Thiyagarajan
Sivakumar Thiyagarajan is a Managing Consultant in Wipro Global Consulting Group’s Retail & Distribution Practice. He has over 12 years of experience in the retail and health & wellness industries, focusing on Omnichannel Supply chain, Merchandising and Store Operations programs, working with Fortune 500 companies worldwide. He has been part of several transformation programs that helped companies implement change and growth strategies across fast fashion, off price, mass merchandiser and pharmacy retail sectors.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here