• Fortune 100 retailer

    Improving the customer experience with "Snap it and find" visual search

    How a Fortune 100 retailer tripled their visual search accuracy for their shoppers.

  • Background

    In this age of instant gratification, consumers expect to be able to purchase on the go. However, thanks to widespread adoption, mobile apps are no longer differentiators for retailers. It’s all about improving the customer experience. A $70 Billion retailer with 2000 store locations and over 200,000 employees wanted to do just that by helping connect customers with precisely what they were looking for at the moment of inspiration. 

  • Information

    Use Case

    Visual Search

    Industry

    Retail

    Client

    Fortune 100 Retailer

Details

Challenge

To remain competitive against the market leader in their space, the retailer needed to improve the customer experience by providing visual search in a mobile application across their product catalog. Their expected outcome was to lower shopper bounce rates by offering more engaging search and recommendation features, and improving the customer experience by reducing the time and effort spent on product searches — which would lead to more revenue by increasing the number of conversion opportunities.

 

With over 2 million products and 8 million images across 20-30 categories and over 5,000 taxonomies, helping customers search and find what they were looking for based on words alone and the way they thought of things was certainly a challenge. The challenge was compounded by the fact that the retailer had a shifting base of 1 million product catalog changes a year.

lowes-upgrade-your-kitchen-tool

 

Solution

The retailer turned to Clarifai to help them build a Snap and Search mobile application. Snap & Search offered their shoppers the ability to use their mobile phones to take pictures of items they like, so they could find what they were looking for or find similar products in the retailer’s catalog faster and more easily than a word search. Snap and search also provided the retailer with the ability to offer their shoppers product recommendations based on their search.   

 

The retailer chose Clarfiai not only for its ease of use, but because the accuracy metrics achieved by using the platform in a proof of concept was 3X greater than what their data scientists had previously achieved. Clarifai’s exponential improvement in accuracy was driven by the platform’s AI-assisted data labeling capabilities to deliver more accurate training data than the internal had produced.

 

Also, the retailer was looking for more than a visual search point solution, they wanted to use the full capabilities of Clarifai’s end-to-end deep learning computer vision, natural language processing AI platform to evolve their AI strategy to continue to solve multiple use cases. In addition, by having a full AI solution, the retailer was able to  avoid having to use a third party system integrator, and multiple deployment options meant that they could choose to deploy the solution on premise.

 

  • Bulk removed thousands of duplicates from an 8 Million image catalog to save time and costs
  • Outperformed their data scientists search accuracy exponentially —by 3X 
  • Improved data labeling time to value by 10X  
  • Lower bounce rates by 12% due to more engaging search and recommendation features
  • On track to increase revenue by 2% by increasing number of conversion opportunities



Results

Clarifai was able to bulk remove thousands of duplicates from an 8 Million image catalog to save time and costs. In addition, they were able to outperform their data scientists search accuracy exponentially —by 3X. The retailer also improved data labeling time to value by 10X. The retailer's digital marketing team lowered bounce rates by 12% due to more engaging search and recommendation features, and the company is on track to increase revenue by 2% by increasing number of conversion opportunities.

 

“Clarifai used a cascading approach starting with classification and then moving on to visual search and then created a custom script to connect the two projects.” said the retailers head of digital development. “By doing this, they were able to break a complicated problem into smaller solvable chunks to achieve 3X the performance of our internal data scientists.” 

16x

Improvement in operational efficiency and moderator productivity

300 to 5,000

Increase in images moderated manually per day per moderator

12%

Reduction in website bounce rates

  • Ready to get started?

    Whether you're a start-up or a Fortune 500, contact us to discuss how to enrich your data with AI assisted labeling to annotate datasets accurately, quickly and affordably.

    Learn more about Clarifai

    Schedule a demo

    Discuss your solutions options

  • Already a customer?

    If you are encountering a technical or payment issue, the customer support team will be happy to assist you.


    Contact support >