BOPIS FOR AMAZON PHYSICAL STORES (2020-2022)

For two years I was the sole designer working on a CX to enable BOPIS at 55 Amazon retail stores nationwide.

Context and constraints

1) Our BOPIS experience is built on F3’s stack. Our team took an OLT goal to launch publicly to our full network of stores by EOY 2021. The only path we could find to hitting that date required leveraging F3’s tech (primarily for outbound fulfillment) instead of building everything from the ground up. This meant that our customer-facing elements would also closely mirror that of Fresh and Whole Foods online ordering CX.

2) This project required approval from and collaboration with many partner teams. At last count more than 25 Amazon designers have reviewed and given feedback on our CX throughout the approvals process.

3) We are the first team to build selectable pickup CX on the Amazon detail page. Today, the only way to pick up your Amazon.com order is to select a locker/counter in checkout—not a very discoverable experience. It was a P0 requirement of this project that a customer could discover BOPIS offers as early as search and select a pickup location as early as the detail page. This meant I was not only designing an SRS BOPIS experience; I was designing pickup framework to be used by other teams in the future.

Detail page ui

When I joined this project, the SRS team had already identified a tech-friendly solution for the detail page CX: They planned to use an existing AUI component, swatches, to surface a choice between delivery and pickup. Our VP was already aligned with this approach.

convincing the team to pivot

I didn’t feel that swatches were the correct UI element for our feature, primarily because they were already being used on the detail page for variations—i.e. swatches allow a customer to select attributes about the item itself. Using swatches to select a method of receiving the item felt like it would be breaking a pattern.

The SRS team liked swatches because using an existing AUI element would allow us to build more quickly. Therefore, I suggested another AUI component, tabs, and this was well-received by our team as well as partner teams.

I presented many backup options as well (which can be shared upon request), but tabs were the crowd favorite and became the solution we moved forward with for our pilot.

User feedback

By the end of 2020 we had been through countless design reviews. In the process we discovered that many other teams were either designing their own pickup CX or wanted to use the pickup solution we were building. Out of all of the teams’ proposed solutions, 3 ideas continually floated to the top:

In a usability study comparing all three of these designs, tabs were found to be the winner. In the upsell design, many participants failed to discover the pickup offer altogether. In the toggle design, participants weren’t sure which state (delivery or pickup) was selected and didn’t always understand that using the toggle changed the offers displayed underneath. The tabs were found to be both discoverable and intuitive.

more research

In addition to the comparative study mentioned above (which was run by the DEX team), I ran our CX through two end-to-end usability studies. These studies helped us settle hotly debated topics and catch potentially confusing interactions before launching to the public. I gave a presentation of our findings to the tech and product teams which included video clips of test participants using our CX. Ahead of our monthly update to leadership, I wrote these learnings into our team’s doc as well.