Smart waste management system
Company
Role
Lead UI/UX Designer
Duration
6 Months
Team
3 Devs, 1 Designer
2 Data scientists , PO
Analytics became the most-used feature within the first month of release.
Problem insight
Design Goal
I along with my product owner and the customer success team conducted more than 20 user interviews across three user types: facility managers, cleaning supervisors, and environmental/sustainability officers. We also interviewed our internal sales team to understand the objections they heard in demos.
The objective of interview was to understand how informed decisions user can make with the current dashboard.
Pain Points
Key insights from research
Analytics was the most requested features. There was no way managers to view operational metrics of waste production in their buildings.
User Stories
Created user stories to ensure the focused solution for the problems.
Design System Creation:
Since we were rebuilding the product from scratch, I created a design system from the ground up establishing type scales, colour tokens, spacing, and a component library. This ensured consistency across the dashboard, cleaning planner, and reporting flows, and gave the dev team a single source of truth.
This is one of the most unique aspects of this project.
Before designing the analytics views, I was directly involved in testing the physical IoT sensor outputs measuring their precision, refresh intervals, and failure behaviours. This isn't typical for a UX designer, but it fundamentally changed how we built the product.
The key finding: Sensors updated approximately every 15 minutes not in real time as we had originally assumed. If we had built a "real-time" dashboard without knowing this, users would have seen data that appeared live but was actually stale, eroding trust in the product.
The UX decision it drove: We redesigned the data refresh model to make the 15-minute cadence transparent showing a "last updated" timestamp on every data point and designing visualisations that reflected meaningful change intervals rather than minute-by-minute noise. Users could track trends throughout the day without being misled by false precision.
This cross-functional involvement sitting with hardware engineers, understanding sensor behaviour, and translating it into interface decisions was one of the most valuable parts of this project.
Challenge 1: Data visualisation for non-technical users Facility managers are not data analysts. Charts that make sense to a data scientist can be overwhelming or misleading to someone who just needs to know "is my building generating more waste this month?"
Challenge 2: Balancing depth with simplicity
The product needed to serve both a cleaning supervisor who wants one thing (which bins to empty today) and a sustainability officer who wants aggregate trends across 10 buildings. These are very different needs on the same platform.
UI Decisions
Dashboard
The redesigned dashboard delivers: Aggregate analytics view waste generation across all buildings, filterable by time period, floor, and waste type
Cleaning planner Scheduling interface showing predicted fill times based on trend data, allowing managers to assign staff before bins overflow
ESG report builder One-click export of waste reports in PDF and CSV, with date range selection and automatic data population
Before
The original dashboard showed individual bin fill levels in a list format. No trends. No filtering. No multi-building view. No way to export data. Users described it as "a monitor, not a tool."
After
The redesigned dashboard surfaces what matters immediately total waste generated, trends over time, bins approaching capacity and puts reporting and scheduling one click away.
Before
