Guide section 15 of 31
Chart Selection Guide
| User question | Data relationship | Recommended visual | Notes / alternatives |
|---|---|---|---|
| Trend over time | Ordered time | Line chart | Column for discrete periods; area only when magnitude emphasis matters. |
| Compare categories | Categorical comparison | Clustered bar/column | Bar is strong for long labels/ranking. |
| Part-to-whole | Composition | 100% stacked bar/column; stacked bar/column | Pie/donut only for very few simple shares. |
| Distribution | Spread/outliers | Scatter; binned column where appropriate | Custom box plot only if distribution statistics are essential and native options are insufficient. |
| Correlation | Relationship between measures | Scatter/bubble | Avoid implying causation. |
| Ranking | Ordered category comparison | Sorted bar | Ribbon only when rank changes over time are the question. |
| Variance | Actual vs expected/difference | Bar/column + variance; Waterfall for contribution | Keep baseline/reference explicit. |
| Target performance | Actual vs target | Card + linear bar/reference; KPI | Gauge only when the radial metaphor adds real value. |
| Contribution to change | Positive/negative drivers | Waterfall | Use ordered contributors and a clear start/end context. |
| Root cause | Drivers by hierarchy/dimension | Decomposition tree + waterfall/bar | Fixed hierarchy when path is known; decomposition tree for genuine flexible exploration. |
| Geographic distribution | Spatial relationship | Azure Maps | Avoid legacy Bing map visuals for new standards. |
| Process stages | Stage volume/conversion | Funnel or ordered bars | Funnel only when the stages truly represent a process. |
| Hierarchy/precise values | Exact values across levels | Matrix | Prefer table/matrix over chart when lookup and accuracy dominate. |
Charts to use cautiously
- Pie / donut: difficult precise angle/area comparison; keep to a few obvious categories.
- Gauge: space-inefficient for dense KPI monitoring.
- Treemap: useful for hierarchical area overview, weaker for precise ranking.
- Funnel: misleading if stages are not truly sequential.
- Stacked area: hard to compare non-baseline series.
- Dual-axis combo charts: only when measures are genuinely related and scales cannot mislead.
- Radar/spider, violin, Sankey, Gantt, box plot and similar custom visuals: use only for a real analytical need, governance approval and performance/accessibility validation.