Jess Chang 〘张美嘉〙   leads design at 

She sees design as a helpful medium for creative problem solving, communication, alignment, and storytelling. She cares significantly about craft, tools, and execution. She believes that form, function, and speed can co-exist.

Enabling NFT curators to curate multiple galleries

Agent rebooking tool for last-minute cancellations

Using ML and micro-interactions to accelerate case note composition

Optimizing for accurate case contact reason selection

product design, taxonomy design, information design

Helping fleet managers compensate drivers and meet customer expectations

Foundational components for Samsara's design system

Bridging brand and product design systems

design systems, brand design, art direction

Index.mdx

Hello, I'm Jess. I currently lead design at   in a growing market for creators and curators.

About.mdx

Enabling curators to curate multiple galleries

I designed a flexible, scalable experience that allows collectors to create multiple galleries to organize their collections in more ways, and visitors to browse galleries in a clean and easily navigable way. This was part of Gallery's push to increase user engagement and retention.

Historically, users on Gallery can create only one gallery with multiple collections displayed on the same page. The experience started breaking for large collectors who had too many things to display on one page. Strategically, we wanted to create more entrypoints into galleries to increase profile engagement and retention.

Large galleries were slow to load, and too much to scroll through.

Strategically enable more entrypoints into multiple galleries

How might we design an experience for collectors to create and manage multiple galleries that is  ,  ,

and a browsing experience that is   and  ?

Design principles: Scalable, Navigable, Clean

Before: An early exploration of "Galleries" subpage on a profile. Primary navigation is crowded. Users complain that the Follow feature and followers list are undiscoverable.

A way to manage and edit galleries that works for both small and large collectors

Building collection management into the primary editor

Gallery.mdx

Samsara.mdx

Agents recognized that the different tabs would load contact reasons for that selection.

Raising the accuracy rate for case contact reason selection

As part of the Nova Case Tools redesign, I redesigned how agents select the contact reason for the case, a $10M feature that agents must use on every case in order to unlock the workflows they must follow, and help Airbnb gather data.

Agents spent ~$10M time determining the contact reason of a support case in 2020, contributing to a low contact reason selection accuracy rate (38.7%).

Agents found contact reason selection frustrating and overwhelming.

How might we help agents select contact reasons accurately?

We studied how 14 participants navigated our prototype of the improved contact reason selection experience.

Several agents explained that they would use the search features to find contact reasons by key words.

Agents appreicated that they wouldn't need to navigate to contact reasons from the top level of the taxonomy.

Selecting a contact reason via navigating a sliding taxonomy, with intelligent filters and an improved search experience

Keyboard shortcuts, surfacing most frequently selected contact reasons from the last week, guessing the contact reason from case details.

Airbnb Changing contact reason.mdx

Medallion.mdx

I redesigned how agents record admin notes, a seemingly small but impactful feature that is used in every support case. Admin notes help other agents on the case review activity that has happened so far. The final implementation has saved Airbnb ~$4MM.

Agents spent 12% of their time writing admin notes, contributing to the ~$9M USD incurred due to ticket resolution time.

Agents often spent more time than needed on admin notes, because they lacked instruction on what to write vs omit.

How might we speed up the writing process?

Airbnb Admin Note.mdx

Helping agents provide rebooking support for last-minute cancelled guests

In 2021, My team shipped a rebooking tool for   that allows agents to proactively reach out to guests who are cancelled on last-minute, and help them rebook, which has brought the rebooking rate from  , and NPS went up from  .

Guests are cancelled on last-minute. Out of frustration, they self-book outside of Airbnb 69% of the time.

1. Agents are not assigned to the case in time

How might we design a rebooking tool for agents that is  ,  ,  , and  ?

Aleck, lead engineer on Batphone, declares that it is non-trivial to spin up a microsite as the rebooking tool for the March 2022 release

Airbnb Batphone.mdx

I designed an analytics dashboard that helps fleet managers compensate drivers and meet customer expectations.

Historically, fleet managers could only measure fleet timeliness at the end of the day since there was no effective way to live monitor their drivers.

Now that fleet managers can use Samsara sensors to track the location of their fleets, how can they understand driving patterns so they can dispatch more efficiently?

We conducted interviews with two large fleet customers.

1. Customers design compensation models based off driver performance.

2. Customers want to see plan vs actual time

3. Customers want to bring products and goods to their customers on time, in order to ensure strong business relationships.

How might we help fleet managers compensate drivers better and meet customer relationships?

Route analytics

Helping fleet managers compensate drivers and meet customer expectations

I designed an analytics dashboard that helps fleet managers compensate drivers and meet customer expectations.

Background

Team

Timeline

Problem

Historically, fleet managers could only measure fleet timeliness at the end of the day since there was no effective way to live monitor their drivers.

Now that fleet managers can use Samsara sensors to track the location of their fleets, how can they understand driving patterns so they can dispatch more efficiently?

Understanding the problem

We conducted interviews with two large fleet customers.

Key findings

1. Customers design compensation models based off driver performance.

2. Customers want to see plan vs actual time

3. Customers want to bring products and goods to their customers on time, in order to ensure strong business relationships.

Reframing the problem

How might we help fleet managers compensate drivers better and meet customer relationships?

Findings

What we shipped

Impact