Helping Warehouse Teams Prioritize Restocking
The Problem
At Flink, warehouse workers regularly restock products from storage areas back to their designated shelves.
The challenge was that the restocking list gave no indication of urgency. Products were simply sorted by shelf location, so it was hard to know which items should be restocked first.
In some hubs, workers could be dealing with more than 80 products at a time, and not all of them had the same impact on daily operations.
The data existed to identify which products were more urgent, but that information wasn't visible anywhere in the workflow.
My Role
I worked on the technical design and frontend implementation of the solution.
My responsibilities included:
- Defining API requirements with backend engineers
- Refining how priority data would be exposed
- Implementing the frontend experience
- Adding sorting and filtering capabilities
- Rolling out the feature across the restocking workflow
The Solution
We introduced a priority system that classified products as High, Medium, or Low priority.
The score behind these labels was based on factors such as product demand, stock levels, and how the restocking request was created.
On the frontend, I added priority indicators, sorting options, and filters so workers could quickly focus on what mattered most.
What looked like a simple UI change was actually powered by a scoring system that transformed operational data into actionable information.
Challenges
One of the most interesting challenges was deciding when the priority score should be calculated.
Some of the data required expensive calculations involving product sales rankings across an entire hub.
Instead of calculating everything every time a restocking task was created, I helped define a solution that precomputed part of the data through a scheduled process and reused those results later.
This significantly reduced the amount of work required during normal operations while keeping the priority information accurate.
Impact
The feature was progressively rolled out across approximately 160 hubs.
Warehouse workers could immediately identify which products needed attention first, instead of relying only on shelf order.
The result was a more consistent restocking process and greater confidence that the most important products were always prioritized.
What I Learned
This project taught me that some of the most impactful features aren't necessarily the most visible ones.
A small visual indicator can have a meaningful effect when it's backed by the right business logic and presented at the right moment.
It also reinforced the importance of thinking about performance early, especially when calculations depend on large amounts of operational data.