Eye Looks Recommender

Bobbi Brown
2020, 10+ weeks
Project Overview
The Eye Look Recommender drives multiple product recommendations comprising a personalized look based on a consumer’s detected facial attribute. It will educate consumers on their unique attributes, showcase how to create set look(s) on the consumer as well as provide a path to purchase related products.
Contributions
UX Design and Research
Looks Recommender screens overview.
The Problem
COVID-19 had disrupted our in-store engagements and we wanted design a solution to deliver the same personalized experience they would get with our beauty advisors in-store.

Our existing Virtual-Try On tool has provided an alternative solution to trying makeup on in-store, however, we have found that it feels impersonal and users drop off due to dissatisfaction or uncertainty about the product.
VTO existing screen (without LR)
Our Goal

Develop a tool that will drive a high-touch experience by providing highly personalized, relevant product recommendations to consumers while educating them on their facial attributes and makeup products.

Process at a Glance

Research
- Conduct generative research to identify requirements and user stories.

Ideation and initial feedback
- Conduct landscape analysis and facilitate sketching session.
- Align with teams to define technical and functional requirements.
- Produce wireframes to share with stakeholders and collaborators.

Prototyping, visual design and usability testing
- Create and test prototype with proxy users on a usability testing website.
- Continue to review UI and intended interaction with collaborators.

Research

Guided by our business and strategy team, we explored existing data for Virtual Try On to identify areas for improvement and error management to have a better understanding of how to develop a rollout strategy to stay aligned with the parent brand.

After gaining a deeper understanding of where VTO stands, we found that it continues to grow in engagement and plan to take advantage of personalization and education in the space. We intend on using facial recognition software to take the experience a step further to educate and guide our consumers about their facial attributed and makeup application.

Defining Requirements

For this project, we would be collaborating with a vendor that would assist our VTO experience with facial recognition and product-match goals. I met with our producer and engineering team to define our requirements and understand any technical limitations.

Ideation

Guided by the requirements and user journey, I designed a user flow to map out all interactions found within the experience to lead our designs.

Understanding the designs

I developed an initial user testing plan with 10 women to understand the usability, concept, and design. These learnings will be incorporated into future executions of the Looks Recommender with other ELC brands as this feature transitions to Product Development. We want to conduct pre-launch testing to discover any opportunity spaces that we can take advantage of to improve engagements and awareness of the feature. 

Once we collect insights from testing, we validate our assumptions and begin high-fidelity designs.

Prototype

Feel free to explore the design and interact with the prototype below. ☺

Projects
Figma tile for project titled "Looks Recommender"Figma tile for project titled "Image Recognition Search"Figma tile for project titled "Lynit"Figma tile for project titled "Talkspace"Figma tile for project titled "Jeweler's Mutual Insurance Company"Figma tile for project titled "Dining Offers"young turks project thumbnail