Zarrin's Design Portfolio

Projects

Ocado Technology

Ocado Smart Platform:
Stock Putaway Times

Background

I designed an override tool to help people manage warehouse stock so that they can ensure maximum availability for customer orders.

Timeline

6 months

Jan - Jul 2024

Role

Led end-to-end design, from research to UI implementation.

Overview

Opportunity

How might we give supply chain agents the flexibility to manage when products are put away in warehouses?

Problem

Retailers rely on Ocado's default 6-hour putaway time, which is inefficient. For one client, only 76% of their products fit within a 6-hour timeframe, but variations by warehouse, temperature, and schedules highlighted a need for a more flexible solution.

Goals

Supply chain agents have a flexible way to manage product putaway times across warehouses.
Reduce the number of unfulfilled customer orders due to late putaway times.

That's a 20% reduction!

Outcomes

I designed the first putaway override tool in Ocado’s Smart Platform, which has reduced unfulfilled items by an average of 100 per day for a client’s top 250 products.

That's a 20% reduction!

Process

Checklist — Check!

Design Process Tracker

Before starting a project, I align with my product manager on the project goals and timelines. Then, I list and size the tasks I think I'll need to complete for my design process, using my handy document to ensure I deliver quality work with my team.

Finding Out Who the Users Are

What We Know (and Don't)

Capturing initial thoughts helped my product managers and me understand what we knew—and didn’t—about the project space.

With no existing functionality internally from clients, we explored from scratch, allowing us to dive deep into understanding our users and shaping the project from the ground up.

Capturing initial thoughts helped my product managers and me understand what we knew—and didn’t—about the project space.

With no existing functionality internally from clients, we explored from scratch, allowing us to dive deep into understanding our users and shaping the project from the ground up.

Data Download

My data analyst provided a report on how our clients were performing with the current 6-hour putaway setup — the performance seemed bad from a first glance.

I used this report to create some specific questions for user research — like why do chilled products get decanted faster in warehouses? (See if you can find the answer somewhere below).

My data analyst provided a report on how our clients were performing with the current 6-hour putaway setup — the performance seemed bad from a first glance.

I used this report to create some specific questions for user research — like why do chilled products get decanted faster in warehouses? (See if you can find the answer somewhere below).

Let's Talk to Our Users

This generative research study explored how users manage late and early putaway times in warehouses. I worked with product managers to define the goals of the research study, methodology, and any deliverables.

We aimed to understand user groups, their toolkits, pain points, and key use cases for adjusting putaway times.

Challenge #1

This was a big research piece involving users from 3 clients. I started by recruiting 2–3 people per group with different roles to get a mix of perspectives.

What I Learned

I learned that focus groups can work well with our busy users — as long as I give everyone a fair chance to speak and follow-up where needed.

What's the Problem to Solve?

Stories on Putting Things Away

User scenarios were found from the research insights across clients.

These scenarios helped me and my product development team understand when and how often putaway time overrides might be applied, and if these overrides need to be applied at a higher level, like a temperature regime housing many products, or a single product level.

The user scenarios were found in the research I ran.

They helped predict when and how often putaway time overrides might be applied, and if they need to be applied at a bigger level or a smaller specific product level.

Public holidays and weekends mean fewer staff are available to put stock away.

Users want to create multiple overrides to optimise putaway times before and after key events.

Promotional products with larger packaging require extra time to be put away.

Short-life products like sushi have a high purge risk and must be stored quickly in the chilled temperature regime.

State the Problem, Find the Root Cause

I refined the problem statement with my product managers to include it in a user story, using it as a ‘north star’ to address the root cause of the problem.

This problem statement helped me and the team define additional success metrics we can track and begin to think of some solutions.

I refined the problem statement with my product managers to include it in a user story, using it as a ‘north star’ to address the root cause of the problem.

This problem statement helped me and the team define additional success metrics we can track and begin to think of some solutions.

What Do Good Solutions Look Like?

Follow the Flow

I collaborated with 2 engineers and a product manager to brainstorm a UI flow based on user needs.

This exercise clarified how users could manage putaway time overrides when the static 6-hour setup wasn’t suitable. We also explored how users might interact with machine learning and AI-generated putaway suggestions — which was an exciting challenge!

I collaborated with 2 engineers and a product manager to brainstorm a UI flow based on user needs.

A First Look at What Could Be

I took our brainstormed ideas and transformed them into UI concepts, working closely with data science to understand how machine learning models could shape front-end content.

To ensure the designs were practical, I ran several review sessions with engineers and data scientists, making sure we balanced innovation with feasibility.

I took our brainstormed ideas and transformed them into UI concepts, working closely with data science to understand how machine learning models could shape front-end content.

To ensure the designs were practical, I ran several review sessions with engineers and data scientists, making sure we balanced innovation with feasibility.

The first wireframe…

…which was refined from feedback

Challenge #2

I shared the first UI concepts with a client, but they weren’t sold on the machine-learning approach. Their concern was the lack of control over system predictions, which could significantly impact warehouse performance.

What I Learned

Building trust and giving users ultimate flexibility was crucial. We decided to prioritise the override functionality and talk to other clients to better understand how they’d use it. This helped us refine and update user needs.

As a supply chain agent
I need to know what the system-generated putaway times are

So that I can decide if they are appropriate and feel in control

As a supply chain agent

I need to be able to manage overrides

If I know better than the system-generated values

So that I can reduce the number of unfulfilled customer items

These learnings shaped user needs

The Outcome

I ran a usability study to test the UI—making sure the labelling and interactions made sense—and conducted an accessibility audit before release to ensure we met the necessary standards.

A month later, my PM visited the USA, where one of our clients decided to adopt the feature after a great conversation. Their analyst tracked the impact, and since the trial period, we saw an average decrease of 100 unfulfilled items per day—about a 20% reduction!