4 Areas AI and Machine Learning Can Improve Your Merchandising and Marketing Decision Making

Navin Dhananjaya, Chief Solutions Officer, Ugam,

Make no mistake about it. Artificial intelligence (AI) is a technology to be reckoned with and retailers would be wise to embrace it. As they strive to improve their merchandising and marketing decision making, many have already turned to AI and machine learning. Incorporating these technologies enables retailers to provide for their customers enhanced and improved shopping experiences.

Here are four areas where AI and machine learning can improve merchandising and marketing decision making:

1. Assortment
Product line reviews across categories require effective taxonomy classification and mapping. With AI, you have the ability to match and compare them with competitors’ corresponding products. Instead of dealing with this typically months-long process if done manually, soon, you’ll be able to turn to machine-learning to classify products under various taxonomies at a fraction of the time. This will allow retailers the ability to review many more categories and be more relevant to customers.

Mapping products against the competition, typically done manually or via rule-based engines, will continue to develop, as doing so at scale will be critical. Retailers have already started to experiment with sophisticated methods of product clustering in tandem with self-learning fuzzy logic algorithms to get this done instantaneously.

2. Content
Online retailers are striving to include more attributes in their product content to inform confident purchase decisions. Attribute extraction – the process of extracting attributes from feeds or manufacturer websites - is carried out for attributing products at scale. Machine-learning algorithms are key in this area, as the need for speed, scale and diverse sources for attributes continues to grow.

In addition, product images also form a key part of product content. Deep learning algorithms can be used to classify images based on the retailer’s requirements — think identifying fraudulent images and promotional text on images at scale.

3. Pricing
Relying on data from historic sales-based trends and forecasts to make pricing decisions isn’t enough these days. Retailers need to incorporate this data with demand data available outside the firewall to be able to make more competitive pricing decisions. Allowing machines to identify price patterns, allows retailers to make smarter, more timely pricing decisions at scale.

4. Search
Product discovery requires products to be discoverable in multiple nodes across a catalog or in the taxonomy. Multiple taxonomies or a routinely changing taxonomy for product discovery is a key governance that retailers need to reckon with in the future. This simply can’t be done manually. Instead, a combination of classification techniques and constant learning algorithms combined with human intelligence will be the future need for many ecommerce retailers.

Some of the content in this article originally appeared in Total Retail on February 9th, 2017. 


Sorry, your search did not match any relevant results.