Dispatchers spent the first hour of every shift hunting through 60 charts for problems.
Demo case study. The story, team and numbers are placeholders that show how a real case study will read.
(Case study 02) SaaS dashboard
An operations dashboard for a logistics company running 400 delivery vans. Dispatchers had 60 charts and no idea where to look. I redesigned the dashboard around the daily decisions they actually make.
3 min read · 2025
Placeholder artwork
The 30-second version
For busy readersRebuilt the dashboard around a ranked list of today’s issues, with the charts one click behind it.
Morning triage dropped from 52 minutes to 14, and support tickets fell by 27%.
- 14 minmorning triage
- 27%fewer support tickets
- 90%of teams use it weekly
Sample numbers
Everything was on the screen, so nothing stood out
The old dashboard tried to show every metric at once. Each chart was accurate, but together they hid the late van, the empty warehouse slot and the driver who had not checked in.
Dispatchers had built their own workaround: a spreadsheet they updated by hand every morning.
I designed it, built it and shipped it.
- 01 · design
Shadowed 12 dispatchers on shift, and designed the issue feed and the chart system.
- 02 · build
Prototyped the feed in React with live data so we could test it on real shifts.
- 03 · ship
Rolled it out depot by depot, with a feedback button on every issue card.
Team: 1 PM, 4 engineers, 1 data analyst
Dispatchers were not reading charts. They were looking for exceptions
Watching people work showed that they scanned for anything unusual, then dug in. The charts were only ever the second step.
- Insight 01
Exceptions first
Nine out of twelve dispatchers started their shift by looking for what was wrong.
- Insight 02
The spreadsheet was the spec
Their hand-made sheet listed exactly the five issues the dashboard should have surfaced.
- Insight 03
Shared screens, shared language
Teams talked about problems out loud, so issues needed names, not chart titles.
The calls that shaped the product
Why the home screen is a list
Stakeholders asked for “a better dashboard”. Users needed a to-do list.
Shipped
Today’s problems, sorted by cost, each with the chart that explains it.
Why only five colours
The old charts used 23 colours, so nothing could mean “urgent”.
Shipped
If something is coloured, it needs you. Everything else steps back.
Prototyped with live data, not lorem ipsum
Fake data hides the real problems, so I built the feed as a working React prototype on top of the live API. We tested it on real shifts in week three.
Drag the slider to compare the design with the shipped dashboard.
FigmaProduction
- Chart components shared with engineering
- Keyboard first, for busy dispatchers
- Loads in under a second on depot wifi
From first beta to live
→ git log --releases
- PilotJun 2025Two depots, live data
- v1.0Aug 2025Issue feed for every depot
- v1.2Nov 2025Shared views for team stand-ups
✓ Used every day by 90% of dispatch teams.
Less hunting, more fixing
- 14 minmorning triage
- 27%fewer support tickets
- 90%of teams use it weekly
Sample numbers
The spreadsheet disappeared within a month, which was the result I cared about most.
What I would do next
Sitting next to people on a real shift taught me more in a day than a month of stakeholder meetings. I now start every data product that way.
Where it's at: Used every day by 90% of dispatch teams.
Next case study