Menu

Desktop Customer Service Chat

Amazon, 2018

Role

UX Designer
Product Lead

Team

Amazon Digital & Devices Customer Service

Status

Launched October 2018
Resolved 11.3% of contacts

Go to experience

Let's put an experience—made specifically for mobile—on desktop.

Message Us(MU) is a customer service chatbot platform aimed at decreasing the amount of customer contacts handled by humans, by providing automation from a bot. Built in 2017, Message Us first launched on Amazon's mobile app. In 2018, in order to reap the rewards of automation, we decided to bring MU to desktop.

Built specifically for mobile, there would be a good amount of work to make this experience desktop-friendly.


Open Questions Brainstorm

Putting an existing experience into a new environment is not as straightforward as it may seem. There were many discontinuities, both technically and in the customer experience. In order to champion the customer and align the team, I organized and led a brainstorm to collect all the open questions that would have to be answered in order for MU on Destop (MUD) to be launched.

51 CX related questions
16 Tech related
4 Ops related

Architecture

After listing all of our open questions, I began thinking about how the addition of MU would alter the existing jouney the customer would go though in order to get in touch with customer service. Previously a customer would choose their issue and then recieve contact channels (phone, chat, email) to choose from. Because issue identification already occurs within MU, we chose to allow the customer to first tell us the channel they want to use.

Ingress Sketching

In order to support the new paradigm of channel selection first, I sketched out what this might look like.

I tried to make the page extremely intuitive, later validated by usability studies

Modality

In parallel to figuring out the ingress, the team had to decide what modality MU would use to function on desktop. To help visualize the different modalities, I wireframed the three different options we had to consider. I chose these by looking at industry standards for messenging platforms.

We determined that while the 'Traveling messenger' modality had the best CX, it was the hardest to implement. Thus, we landed on the 'Pop-up' modality for the following reasons:

Concept validation

Before we launched this experience, we wanted to conduct a concept validation with Amazon customers to identify if there were any CX pain points that we had overlooked. With the help of our Front-end Engineering team, I created a prototype that mixed click-through Invision prototypes with live code. After 10 sessions with customers, we identified opportunities for bettering the experience, but nothing large enough to be a blocker.

The final designs for the new MUD pop-up window and channel selection page.
This is the journey customers go through to interact with Amazon contact channels.
Because of the pop-up modality, I had to define the behavior of how MU would redirect the customer to a self-service page.