Ocado, 2024
Smarter stock management that cut unfulfilled order items by 10%
Team
Product manager
Data analyst
Data scientist
Software engineers
Role
UX research
Conceptualisation
Usability testing
Engineering handoff
Constraints
Low user engagement
Contractual deadline
Timeline
Jan - Apr
4 months
AT THE TIME…
Customers weren’t getting their orders,
and warehouse teams were feeling the pressure to deliver.
This all traced back to one culprit: an outdated warehouse rule.
This rule was a rigid time window that didn’t give slower-moving stock and people a chance, causing waves of substitutions and shortages and not-so-happy customers.
6-hour
unchangeable rule
when stock needs storing by
24%
of stock took longer than
6 hours to put away
Stock shown as available
on webshop when it's not
frustrated
customers
Why not just automate stock timings?
The data science team did. They trained a machine learning model to predict put-away times based on real behaviour. The result? About 12% fewer substitutions and shortages.
But we realised the model didn't account for future disruptions and so we wanted to introduce a tool that allowed managers to override the predictions as needed.
Stock piled and ready to be picked for customer orders, the warehouse folks worked hard to reach the 6-hour goal
TALKING TO WAREHOUSE TEAMS
Critical put-away decisions are made as issues arise
I identified 2 main scenarios from users that guided our designs: to be able to store and sell very short shelf-life products, and to allow for longer put away times for when unexpected disruptions occur.
When stock need more time
Public holidays and weekends meant fewer staff were available to put stock away.
Bulkier and promotional products with larger packaging require extra time to be put away.
Freezers in the warehouse can become non-operational, putting strain on supply chain.
When stock needed less time
Short-life products like sushi have a higher risk of going to waste and must be stored and sold quickly in the chilled part of the warehouse, well before the 6-hour mark.
When suppliers deliver late to the warehouse, teams strive to put things away quicker.
With low engagement, we risked gaps. I prioritised scenarios with users to ensure we focused on the most important ones.
With all the prioritised scenarios I had, I asked the question…
How might we let warehouse teams react to
real-world conditions so products are available when customers shop for them?
BRAINSTORMING
Surfacing the right information so managers can
plan with confidence
After a design session, the team and I created a UI flow of how the machine learning model would work alongside a user override function.
The goal was to have a single place to view the current put away times,
OUTCOME
An easy way to
After a design session, the team and I created a UI flow of how the machine learning model would work alongside a user override function.
The goal was to have a single place to view the current put away times,
Challenge:
One of our partners weren't too sure about
REFLECTIONS
Our tool rolled out to Kroger in the USA
A month after we launched the tool, my product manager visited the USA, where one of our partners, Kroger, adopted the feature. Their analyst tracked the impact, and since the trial period, we saw an average decrease of 100 unfulfilled items per day — a 10% reduction!
Keep stakeholders close
There were A LOT of people involved and interested in the project. Noting down what level of involvement teams wanted and how they wanted to be informed helped me practice my communication styles for different folks.
Speak up for good design
I was the only UX'er on this team and the nature of the work was very data-driven. We were under pressure to build a good experience and I made sure we were able to do that by engaging with users as early as I can to design around the core user needs.
