How Jamie built an online shopping recommendation system with Joonbot

Sometimes, we don't know what we want. In this case, filters are useless, an online shopping recommendation system is what we need.

by
in
5 MIN READ | November 29, 2021
Jamie e-commerce shopping assistant

In a nutshell

Challenge: Provide an interactive and frictionless shopping experience to make Jamie & I the favorite way to shop online.

Solution: Create a virtual shopping assistant that learns about your personal style and gives you a curated selection of outfits tailored to your taste.

Result: Make customers satisfied by having them discover and purchase the perfect outfit that fits their personality 100%!

Context

Jamie is not an online clothing store like others, the brand has a strong and well-defined positioning.

Jamie’s promise is to provide a curated collection of wow pieces from independent, sustainable designers you’ll love.

For that, Antonia, the founder of Jamie, is betting on three pillars:

  • Unique: Jamie showcases a curated selection of boutique brands creating stunning pieces of high quality and unique designs.
  • Personalized: Jamie strives after finding the perfect fit for your personality.
  • Inspiring: Jamie helps you experiment and get new style inspiration, stimulating the creativity napping within us.

Challenge

On online shops today, the only manner to find what you need is to play with filters.

The issue is that it supposes you already know what you are looking for. Also, it means that you like spending a huge amount of time scrolling over many articles to decide what you’ll buy …

Look at the Asos filters:

Now, think about the experience you have when you go to a shop and get the help of a shopping assistant. You don’t have any idea of what you want and the shopping assistant helps you by asking you some questions and then provides a selection of outfits tailored to your taste. In the end, you can even discover some articles you would never think you would wear and they become your preferred outfits.

So much easier and inspiring, right?

Question is… How to provide a recommendation for online shopping? 😉

Solution

Antonia found the solution and went beyond what a shopping assistant does. Jamie is not an online clothing shop, it is an online shopping recommendation system with the design of a virtual shopping assistant. The system learns about your personal style, while regularly sending you a curated selection of outfits tailored to your taste. Antonia calls Jamie, the Spotify for Fashion. Amazing, right? 😉

So, Antonia created a style game with Joonbot that plays the role of the shopping assistant.

Wondering how the style game looks like? Here is an extract:

The advantage of using Joonbot to do this style game is that the experience is interactive and frictionless. You don’t write anything, you only look at pictures and click on a “like” or “dislike” button, like on Tinder in fact!

Antonia built with Joonbot a scoring system that determines, depending on the answer given, if the user has a classic, modern, feminine or relaxed style. To do that, she used our logic jump and calculation blocks!

After a user ends the style game, he is redirected to a specific web page with a curated selection of outfits tailored to his taste thanks to the scoring system.

She also sends the style profile given by Joonbot to her database so that she can send regular emails with outfit inspirations that match the user style.

Want to experiment with it by yourself?

Here is the light version of a style game:

You can download our personal shopper chatbot template in your Joonbot account to get inspired.

Next steps

After testing Joonbot as a customer acquisition system, she plans to create new Joonbots for customer retention.

It means an online shopping recommendation system available 24/7. 🚀

I’ll share it with you in another use case!

 Camille Franceschi
 CEO, Joonbot

Our articles by category

Boost your conversion rate now

Get your access for FREE to build your own chatbot with our chatbot builder.

Blog → Case studies → How Kiloutou is fully automating its material return process