Shedd App

Pre-loved clean fashion

Research
Brand Identity
Design System
UX + Virtual Design

Most vibrant preloved fashion app for pre-loved second wear clothes. Shedd app evolved brand transforms secondhand from a musty compromise to a proud fashion statement. Celebrating style and body diversity, Shedd sought to encourage consumers to reduce their fashion waste by buying and selling clothes, shoes and accessories from and to like-minded girls near them. Delivered 9.5k app installs via social media from 23,674 clicks – a 40.1% conversion rate

Role

I was given the opportunity to develop a design atomic pattern library for Shedd. My mission was to document my methodology and provide initial concepts of the molecules, elements, graphics, animations, and pages.

Mission

Introducing Shedd's most vibrant, preloved fashion app. Celebrating style and body diversity, Shedd sought to encourage consumers to reduce their fashion waste by buying and selling clothes, shoes, and accessories from and to like-minded girls near them. Shedd's purpose is to make better use of the world's possessions.

Impact

Shedd, as a startup, has become a women's fashion marketplace, enabling people in Spain, Bulgaria, Romania, the UAE, Portugal, Colombia, and Australia to trade second-hand fashion since the project's launch. It differs from other classifieds and facilitates pick-up and delivery operations between the seller and the buyer.

To create an excellent visual design, Shedd needs to focus on modular building, which consists of three main components: a pattern library, a style guide, and a design system. However, it's not enough for the system to be built with purpose; it should also align the team with a shared vision of the brand's voice, visual style, and behavior. A shared definition of how the brand should look, feel, and behave is crucial. Moreover, it should allow for meaningful experimentation without feeling constrained.

Creating products that feel cohesive is essential. When design systems function harmoniously, their workflow becomes more efficient, and the user experiences they generate feel more meaningful and coherent.

To reimagine Shedd brands, each one of us was asked, "If Shedd were a person, what would her personality be like?" We also did a survey and found the following strong points: accessible, feminine, fun, vibrant, unique, and humble. The product succeeds by running quietly in the background, focusing on simplicity and efficiency.

Link to view all the design system on Invision App

I came up with this style guide after completing a 4-day sprint workshop on transferring our brand values to a living pattern library and style guide with our Shedd character on top of it.

The initial draft of the Shedd ecosystem and design principles. Starting with design principles is a great way to begin. These principles are shared guidelines that capture what good design means for the team and provide advice on achieving it. Running workshops to define these principles can lead to constant learning within the team.

Checking our usage metrics and using the CUPID system.

The objective was to find a feature with an interaction that will get used the most by asking, "Which steps are always required to complete this selling process?" With that in mind, we've spent more effort working on the seller flow, particularly this page.

We experimented with adding more information to the feed, including size, brand, and asking price. Other additions included users adding items to their favorites while browsing, which was previously only possible on the product page. We saw reduced detail page views by surfacing relevant content to the listing page but a stable purchase intent conversion.

Bottom sheets

Component documentation example

Bottom sheets slide up from the bottom of the screen to reveal more content. We needed to use them to introduce new contextually relevant content on a unique surface and to display content equal in value to the primary content.

Link to view all the design system on Invision App

Seller getting paid

Educating users on what's next using visual cues and setting user expectations when an order is placed, such as by using clear and concise language and using visual cues. I aimed to deliver a simple and short copy without focusing much on UI design.

The thumbs up/down feedback helps us to actively discover issues when completing a task with the current interface. Alongside doing first-look testing, this allowed us to save time and effort while providing a more comprehensive look at the overall effort required.

Buyer Journey

These are high-profile and high-risk pages to make adjustments on, as always. We rolled this out as an A/B testing experiment with an extensive control group. Analytics showed us that a significant traffic source toward the product page was from the user's profile page, which was complicated to navigate. We initially decided to make this page more profile-driven, including the user's number of followers, items for sale, and quick navigation to the other things the seller had listed.

The outcome drove more customers to purchase the product page with a stronger intent. As a result, more sellers and customers who viewed these sections were twice as likely to convert than any other DPV source.

Listing feed page to showcase all our pre-loved clothes

We experimented with adding more information to the feed, including size, brand, and asking price. Other additions also included users adding items to their favorites while browsing, which was previously only possible on the product page. We saw reduced detail page views by surfacing relevant content to the listing page but a stable purchase intent conversion.

Listings feed outcome: we saw a reduction in detail page views by surfacing relevant content to the listing page but a stable purchase intent conversion.

*WAU - Weekly Active User

Product page and outcomes

These are high-profile and high-risk pages to make adjustments on, as always. So, we rolled this out as an A/B testing experiment with an extensive control group. Analytics showed us that a significant traffic source toward the product page was from the user's profile page, which was complicated to navigate. So we initially decided to make this page more profile-driven, including the user's number of followers items for sale and quick navigation to the other things the seller had listed.

The outcome drove more customers with a stronger intent to purchase the product page. As a result, the 'More from this seller' and 'Customers who viewed this also viewed' sections were twice as likely to convert than any other DPV source.

The left graph shows replies per Active User and ad Shares per Active User. The new listing page promotes the sharing of ads and exploring through the "more from this user" section, driving better traffic and higher conversion.

The right graph shows the distribution of in-app traffic by Source (iOS) - 'Recommendations' and 'More from this seller are now twice as likely to convert compared to other listing page sources.