Designing an automation tool to manage 1,000+ stores

I designed an automation tool for the Stockwell’s retail team. Leveraging machine learning and crowd-sourcing, the tool provides stocking suggestions for Stockwell machines, significantly boosting efficiency. As the lead designer, I worked with a product manager, two engineers, and two retail buyers.


It was unsustainable for our five person retail team to manage 1,000+ Stockwell stores.


If we provide the retail team with an automated workflow tool, we can reduce headcount, boost revenue growth, reduce operational costs, and efficiently manage large store groups with a lean team.


Designed a dashboard as an automation tool to streamline the retail team’s workflow.


After the tool launched, I had already departed Stockwell in 2019. Unfortunately, the company closed in 2020, and I believe the tool had only a brief period of use before the onset of COVID-19.

The Process

Gaining insights from the user research

I joined the project three months after initial user research and had to quickly get up to speed.

Unveiling the need for workflow automation

Here’s what unfolded as I delved into the project and identified the key challenges:

Understanding our users and their workflow

Our primary users were five members of our retail team, while two were deeply involved in daily store operations. The need for workflow optimization became evident after we discovered they manage store tasks manually via cumbrecome spreadsheets.

Identifying the challenges

As we dissected the issues, we distilled our findings into the following key challenges:

Envisioning solutions and potential outcomes

Envisioning the adoption of a workflow automation tool by the retail team, we hypothesized on these potential outcomes:

Mapping the new workflow

Before designing the automation tool, our team investigated the retail buyer’s store merchandising process. Armed with this insight, we mapped out the new workflow:

Sketching the blueprint before the hi-fi designs

I sketched wireframes to visualize the tool’s structure and core components. These sketches get translated into UI designs to facilitate communication of interface decisions to the team. Following feedback sessions, I iterated on each feature to ensure alignment with our objectives.

The Solution

Upon my departure, the Stockwell Automation Tool was launched in 2019, marking the culmination of our efforts. While I was not privy to quantitative data post-launch, I firmly believed in the positive impact it would have.

Contributing to Stockwell’s design system

Amidst my primary project, I seized the chance to collaborate on auxiliary endeavors. One such endeavor involved contributing to Stockwell’s design system,a cornerstone of our design ethos. Below is a snapshot showcasing my contributions:

My Learnings and future improvments

Asking the right questions – Throughout the project, I leaned on my product manager for insights into the technical aspects. However, I discovered the invaluable perspective offered by engineers. Learning to frame questions effectively became a habit, shaping my approach at every stage.

Understanding user behavior – Reflecting on past experiences, I realized the importance of observing user actions rather than solely relying on their verbal feedback. Moving forward, I prioritize user testing sessions to glean insights into their behaviors and preferences.

Harnessing the design system – Initially I felt constrained by Stockwell’s design system, but after, I saw it as a chance for growth. While the existing color palette posed limitations, I learned to advocate for necessary changes and initiate discussions with colleagues to enhance our design patterns.

Check out the next work here

👉🏻 Financial Portfolio Model