with Faisal TusherAnalytics · Visualisation
Data Visualisation
A chart is an argument. The right one makes the answer obvious in five seconds; the wrong one hides it behind decoration.
Try the interactive tool 
The building blocks
What goes into Data Visualisation
Chart Chooser
Pick what you want to show — the same demo dataset re-renders in the recommended chart.
How I do it
The process
- 1
Question
What decision is this for?
- 2
Data
Right grain and period.
- 3
Form
Chart type.
- 4
Polish
Labels, colour, annotation.
Principles I follow
- Write the question the chart answers as its title.
- Remove gridlines, 3D and legends where labels suffice.
- Design for management reading on a phone.
Measurement
The numbers I watch
Time to insight
Seconds for a reader to state the takeaway.
Tools I use
- LSGoogle Looker Studio
- G4Google Analytics (GA4)
What you get
- Chart style guide
- Dashboard layouts
- Annotated reports
Pie chart overload
A report uses ten-slice pies for channel share.
Replace with a sorted horizontal bar chart with direct labels.
Illustrative exampleAre donut charts bad?
Fine for 2–4 parts of a whole; poor for comparing many categories.
Related
Continue exploring
- Looker StudioBuilding Looker Studio marketing dashboards: data sources, blending, calculated fields, filters, page design and automated reporting, with a wireframe builder.Analytics
- Marketing DashboardInteractive marketing intelligence demo.Analytics
- KPI FrameworkA marketing KPI framework: objectives, outcome KPIs, driver metrics and diagnostics, targets and review cadence — with an interactive KPI tree.Analytics
- Reporting & Business IntelligenceMarketing reporting and BI: audiences, cadence, data pipelines, metric dictionaries and turning reports into decisions — with a reporting cadence matrix.Analytics