Depict.ai’s solution has been to amass a much bigger trove of information by aggregating data from across the internet

Search Filter

Overview
Yahoo Finance
The giants within e-commerce have been able to build huge in-house teams that leverage AI to deliver clear bottom-line impact. For instance, Amazon is reported to earn 35% of its revenue from their recommendation engine (McKinsey report). However, most e-commerce stores don’t have the technical capabilities, nor the large amount of sales data to make existing recommendation systems work. To counteract this imbalance, Depict.ai has developed AI technology that is available to every retailer. In a nutshell, the solution understands all of a retailer’s products, and is deployed as a plug and play solution. Depict.ai utilizes the most recent advances in Deep Learning, training neural networks to understand product images and information to an unprecedented extent. Depict.ai’s engine is unique in that it doesn’t require any user behavior data to deliver product recommendations of the highest quality.
When I’m joining Depict.ai they already have the template/UI Kits for search filtering and any other UI components to be applied, adjusted and blended into the retail e-commerce sites.
Sites URL of customers:
https://www.oscarjacobson.com/en-gb/
https://hope-sthlm.com/
In this case, I adapted the UI styles of each site.


The above designs are for the Oscar Jacobson’s site. Category and list depend on what they have on the sites or we can improve and suggest
Prototype
This is the mobile web prototype. We did some improvements and put the recommendations on some pages such as:
– Frontpage
– Product Detail
– Checkout
– Product Category
– After basket
Our Customer Success Manager has already defined what pages need to be improved. I delivered the design and prototype which would be shown to the customers.
Oscar Jacobson – After the basket
The recommendation will appears after the user check out right at the top of Product Detail.
Hope – Product Detail
Added another recommendation section after their own. ‘Picked for you’ with grid layout. We use a grid layout when we need to put the ‘View more’ text button which means more products than using slides.