# Stacked and Composed Charts Stacked charts answer how a total divides into contributions. They work best when the total and a small number of stable components both matter. Interior layers do not share a baseline, so their individual values are harder to compare than the first layer or the total. Use a normalized stack when proportion matters more than magnitude. Use a mosaic when both column width and internal height carry part-to-whole meaning. Use a streamgraph only when changing shape is the primary story and exact values remain available elsewhere. ## Choose the composition | Reader question | Start with | | --------------------------------------------------------------- | --------------------------------------- | | How do several series contribute to a changing total? | Stacked area | | How does proportional mix change independently of the total? | Normalized 100% stack | | How does the overall shape of many positive series evolve? | Streamgraph | | How do two categorical part-to-whole dimensions interact? | Marimekko or mosaic | | Which subgroup values must be compared precisely across groups? | Grouped bars or aligned small multiples | | Do contributions extend in positive and negative directions? | Diverging stack around an explicit zero | Prepare explicit interval endpoints for every stacked row. The application owns series order, offset, normalization, and the denominator behind every proportion. [Scales and D3](../concepts/scales-and-d3.md) defines that preparation boundary. ## Preserve totals with a stacked area A stacked area combines a shared ordered x domain with explicit lower and upper boundaries for every series. The top boundary carries the total; the thickness of each layer carries its contribution. Keep series order stable across updates. Reordering layers can make unchanged values appear to move substantially and breaks the reader's spatial memory. When one series needs precise comparison, place it on the shared baseline or give it a separate aligned view. The prepared rows should retain original values for tooltips and selection, even when marks consume `y1` and `y2` interval channels. See [Data and Channels](../concepts/data-and-channels.md). ## Compare proportional mix A normalized stack gives every x position the same total height. It answers which series gained or lost share, but deliberately removes the original total magnitude. Format the quantitative guide as a percentage and state the denominator. Keep raw totals available in a tooltip, table, or companion view when the reader may otherwise mistake stable share for stable volume. Normalize within each x group before rendering. Do not normalize the complete dataset once and then divide it into time slices. ## Emphasize changing shape A streamgraph offsets and orders layers to reduce visible oscillation around a central baseline. It is effective for the broad shape of many positive series, but the displaced baseline makes precise values and totals difficult to read. Treat offset and layer order as analytical parameters. Keep them deterministic across revisions, preserve stable series colors, and provide exact values through [Tooltips and Focus](../guides/tooltips-and-focus.md). Use an ordinary stacked area when totals or baselines are part of the question. ## Encode two part-to-whole dimensions A Marimekko chart uses column width for one categorical total and vertical composition for a second. Each cell is an explicit rectangle with both horizontal and vertical interval endpoints. The two dimensions have independent denominators: response totals determine each question's column width, while response-category shares determine height within that question. Keep those calculations separate and label both meanings. Small cells may need a tooltip or adjacent table rather than unreadable direct text. Rectangle endpoint semantics are defined in [Bar and Rect Marks](../reference/marks/bar-and-rect.md). ## Production checks - State whether the chart preserves totals or normalizes every group. - Keep series order, category order, colors, and keys stable. - Include zero for ordinary positive stacks and an explicit zero rule for diverging stacks. - Preserve raw values alongside derived endpoints and proportions. - Avoid too many layers; group minor categories only when the aggregation is defensible and disclosed. - Use a legend or direct labels that remain meaningful in light and dark themes. See [Legends and Color](../guides/legends-and-color.md). - Verify keyboard focus and exact-value access with [Accessibility](../guides/accessibility.md). [Marks and Layering](../concepts/marks-and-layering.md) explains how the interval areas, rules, labels, and highlights compose into one chart.