top of page

AI Case Study

Stitch Fix assists personal stylist staff in choosing matching clothing items for customers by filtering products to closely match customer preference using recommendation algorithms

Stitch Fix, a US based online styling service, selects products for customers based on surveys, descriptions, feedback and order history over-time using collaborative filtering and mixed-effects modeling. They analyse style trends, body measurements, customer feedback and preferences to assist human stylists identify suitable products for each customer.

Industry

Consumer Goods And Services

Retail General

Project Overview

"Stitch Fix has combined the expertise of personal stylists with the insight and efficiency of artificial intelligence to analyse data on style trends, body measurements, customer feedback and preferences to arm the human stylists with a culled down version of possible recommendations. This helps the company provide its customers with personalized style recommendations that fit their lifestyle and budgets."

Reported Results

StitchFix has 2.2 Million active customers

Technology

"Collaborative filtering problem: given different clients' feedback on different styles, we must fill in the gaps in the (sparse) matrix to predict the result of sending a style to a client who has not yet received it. Mixed-effects modeling, which is particularly useful because of the longitudinal nature of problem to learn (and track) clients' preferences over time, both individually and as a whole.

Trained neural networks derive vector descriptions of 'pinned' images, and then compute a cosine similarity between these vectors and pre-computed vectors for each item in inventory."

Function

Marketing

Customer Management

Background

Stitch Fix is an online subscription and personal shopping service in the United States founded in 2011.

Benefits

Data

"Clients' self-descriptions, Clothing attributes.
Photographic and textual data: inventory style photos, Pinterest boards, and the vast amount of written feedback and request notes "

bottom of page