WeOS Shopping Assistant

Customers want to buy your products but sometimes need help finding what's right for them. With WeOS Shopping Assistant, you can convert any Square store into an AI Powered conversational experience.

Comment

Inspiration 🤔

It takes forever to find what you’re looking for when you shop online

  • 30% of eCommerce consumers use on-site search
  • Accenture showed that “40 percent of consumers have left a business’s website and made a purchase elsewhere because they were overwhelmed by too many options”
  • Online shopping is preferred for convenience and price comparison, but in-store shopping is still preferred for the overall experience due to better social interaction. Despite the convenience of online shopping, people still prefer in-store shopping for the overall experience.

So, how can we make online shopping a more approachable and enjoyable experience?

What it does 🔥

WeOS Shopping Assistant transforms any Square website into a conversation mimicking that in-store shopping experience

  • A user search triggers a conversational search experience
  • Our shopping assistant asks relevant follow-up questions just as if they were in the store
  • As they respond the results are narrowed down until they find what they want
  • At the end of the journey other items are recommended

How we built it 🪄

WeOS Shopping Assistant is built on our WeOS platform which makes it easy for businesses to leverage data that lives in their SaaS applications

  1. Integrated Square into WeOS so that products can be returned when the user does a search. WeOS has a novel approach to integrating with SaaS to make data more accessible
  2. Created an interface that combines search and generative AI to create a conversational search experience

Challenges we ran into 💪

  1. The Square Items API is not ideal for product search. Even searching in the square dashboard for items isn't a consistent experience
  2. The Square API is deeply nested and APIs are optimized for performance though they lead to having to make many calls to get the full information.
  3. LLMs can be slow and e-commerce consumers expect fast results. We combined keyword search and generative API to mitigate this

Accomplishments that we're proud of 🙌

  • Pivoting and executing in 2 days!
  • Leveraging our existing technologies to get the job done

What we learned 🤓

  • Square has a lot of great APIs that cover the length and breadth of the platform.
  • Setting up a site on Square is very easy
  • The Snippet API was a great alternative to using the Custom Tracking option to get things installed on square site

What's next for WeOS Personal Shopping Assistant 🤩

We’re excited to get this into the Square marketplace

We think there's a future for us in the Square ecosystem! Here are some things we have planned!

For Business Owners

  • We want to create a more integrated experience
  • An admin view to review conversations and even take over if/when required

For Consumers

  • Make voice search available on all platforms
  • Add Visual search to support a user uploading an image and saying “I want a dress like this but shorter”
  • Handle customer support queries (Hey, where's my order?)
Share this project:

Updates