Thoughts, resources, articles
woman wearing beige and red floral top leaning on gray concrete slab with white leather bag ontop
Photo by Tamara Bellis / Unsplash

Tel Aviv's Aspectiva has found a new home in Walmart's Store No. 8.

Leveraging on Natural Language Processing (NLP), the team aims to enhance the customer journey from discovery to purchase through product reviews.

Recommendation systems are notoriously tricky given that no single system seems to be enough to solve for every use case. Here are 3 basic ways to illustrate:

  1. Recommend based on shoppers who have the same attributes as me (e.g. Asian, young, middle-income)
  2. Recommend based on items that were commonly bought together
  3. Recommend based on my previous shopping history

There are natural drawbacks to any approach but Aspectiva seems to attempt to reconcile at least the first two. With Walmart's treasure trove of consumers' shopping history and respective product review, Aspectiva can leverage on that corpus of data to pick out specific attributes of various products to add into Walmart's technological arsenal.

Such technology could prove to be more important than we currently realise given the rise of user-generated content. Signals on social platforms and blogs have also risen and retailers are in the race to catch up.