Designing SmartSort: Roadie’s Warehousing App

Roadie Overview

Roadie is a crowd-sourced delivery platform that enables last-mile delivery all across the US. With more than 50k monthly active drivers, Roadie is a trusted delivery partner for many large-scale enterprises as well as small-medium businesses.

Challenge

Retailers rely on inefficient delivery solutions and struggle with overhead costs and delays. How can Roadie get their products to their customers faster than ever?

Opportunity

Roadie can leverage retailer warehouses and turn them into sorting and distribution centers.

By leveraging warehouses, Roadie eliminates the need of 3PL distribution centers, thereby reducing delivery costs and fulfillment times.

Solution

Working with Fanatics, I designed an MVP product that enables sorting of packages and pickups directly from warehouses.

Discovery

The initial stages of this project focused on understanding how Fanatics operates. Being a 0 to 1 product, I did not get to conduct research in its conventional form by talking to users. Instead, I visited Fanatics warehouse in Jacksonville, FL where I learned more about how packages arrive and move through the warehouse, their size, label format, and the technology in use.

Definition

To turn warehouses into a sorting and distribution center, the design should allow the warehouse employees to scan packages, sort them into bins and hand them off to Roadie drivers.

Key Requirements

The system should support two core functionalities

Designing for Warehouse Reality

Warehouse staff experience high churn rate and limited onboarding time. I had to ensure that the product was:

• Easy to learn in minutes
• Does not slow down users

User Flow & Initial Sketches

I designed the following user flow based on the tasks user needs to complete and then created some sketches to visualize the interaction.

Designing under operational constraints

I reviewed the initial designs with stakeholders including Operation Managers at Fanatics and Roadie’s Engineering team.

• Adding secondary bin scan

  To prevent accidental missorts, the Fanatics operations team wanted a confirmation that user has placed the package in the right bin. As a result, I added a step where user needs to scan the bin after scanning the package.

• Dropping the wave concept

  To allow warehouses to operate in waves (e.g. AM wave, PM wave), I designed an overview to select waves. However, in reality the warehouses did not operate in waves with shipments from waves rolled over into existing wave.

• Supporting driver verification

  To prevent drivers from claiming bins, I suggested that instead of revealing the bin number to the driver, the warehouse user verifies the Batch ID provided by the driver and maps it to the corresponding bin number.

Final Designs

Incorporating feedback from Operations, I designed the following high-fidelity designs of the SmartSort application.

Business Outcomes

As the designer leading this initiative, I worked closely with Operations and Engineering to ship an MVP. Since its launch, SmartSort has enabled Fanatics to run sorting and pickup directly from their warehouses. SmartSort served as a foundational tool in supporting this new operating model.

• Fanatics expanded SmartSort operations to 8 warehouses within 6 months
• Contributed to a 19% reduction in overhead costs and 25% increase in volume as warehouse-based fulfillment scaled
• Enabled faster fulfilments as delivery time reduced from 2–3 days to under 24 hours

Starting Roadie’s Own Warehouses

Following the successful rollout at Fanatics, Roadie launched its own warehouse operations using SmartSort as the primary sorting and pickup tool.

SmartSort now supports the sorting and handoff of hundreds of thousands of daily shipments, contributing to 25M+ total shipments processed through Roadie-operated warehouses.

Takeaway

Designing SmartSort has taught me how to build systems in the absence of established workflows. Without existing processes to rely on, I grounded design decisions in warehouse reality, operational constraints, and engineering feasibility. This experience strengthened my ability to translate complex physical operations into scalable digital systems.