Sunnyland Farms was experiencing an unusually high number of abandoned carts. Customers frequently added items to their cart but were unable to complete checkout, leading many to contact the call center for help. I worked on this project as part of a five‑designer student team, collaborating directly with the CEO of the business and coordinating with their developer to understand technical constraints.
To understand why customers were abandoning checkout, we evaluated the existing experience using a combination of qualitative and evaluative research methods, including surveys, interviews, journey mapping, competitive analysis, and current‑state usability testing.
Across usability tests, a consistent issue emerged during the shipping step. Sunnyland Farms allowed customers to purchase multiple items in a single order and ship them to different recipients as gifts. While this capability was valuable, users struggled to understand how to assign specific items in their cart to specific shipments. The interaction model did not align with their mental model, causing confusion at a critical moment in the checkout flow.
This confusion often resulted in customers abandoning the experience altogether or calling the business for assistance to complete their purchase.
Early design exploration focused on making the shipping assignment process more intuitive. One concept I proposed involved a drag‑and‑drop interaction that allowed users to visually move items from their cart into different shipments.
As the team evaluated this approach, usability and accessibility concerns surfaced, and conversations with the CEO’s developer highlighted technical constraints that would make the interaction difficult to implement and maintain.
The direction ultimately shifted toward a simpler and more explicit interaction pattern: allowing users to assign a shipping address to each item via a dropdown selection. This approach preserved the flexibility needed for multi‑recipient gifting while reducing interaction complexity and improving accessibility relative to the drag‑and‑drop concept.
In parallel, the redesign also included improvements to the visual layout and structure of the home page, navigation, and product pages to make it easier for customers to find products and add them to their cart.
In the months following implementation, the CEO reported approximately a 15% reduction in call center volume. While abandoned cart rates were not directly measured, earlier research showed that customers frequently called when they became stuck during checkout. The reduction in calls strongly suggested improved self‑service completion and better digital containment for high‑intent customers.
The redesigned experience reduced friction during checkout while also improving overall product discovery and flow through the site.
Looking back on this project now, with several years of product design experience, I see it as an early lesson in balancing “intuitive” interactions with accessibility, feasibility, and long‑term maintainability.
At the time, I gravitated toward a drag‑and‑drop pattern because it felt efficient for organizing items across shipments. What I didn’t fully account for then was how quickly that approach introduces accessibility risks for keyboard and assistive‑technology users, as well as engineering complexity that can slow delivery or limit adoption.
This experience shaped how I work today. I now pressure‑test concepts earlier against accessibility needs and technical constraints, and I treat those conversations as first‑class design inputs rather than downstream checks. The stronger solution wasn’t just a different UI pattern—it was learning to surface tradeoffs early and design for the broadest set of users within real platform constraints.
