Line chart

Updated
View as text
Open in Confluence
Open in ChatGPT Open in Claude

Line charts are for time. The connecting line implies continuity, which is exactly why they are wrong for categories: connecting Billing to Access implies something that is not true.

Multiple series are fine, up to about five. Beyond that use small multiples.

#When this is the right chart

  • Trends over time.

  • Comparing a few series on the same scale.

  • Before and after, where the shape of the change matters.

#What the data needs to look like

Field

Type

Time

Temporal: a date or timestamp

Value

Quantitative

Series

Nominal, optional: one line each

#A worked example

Two series over eight weeks:

Line chart

The source, which you can paste into a new diagram and edit:

{
  "$schema": "https://vega.github.io/schema/vega-lite/v5.json",
  "description": "Weekly active users by plan",
  "width": 380,
  "height": 220,
  "data": {"values": [
    {"week": "2026-07-06", "plan": "Free", "users": 1210},
    {"week": "2026-07-13", "plan": "Free", "users": 1305},
    {"week": "2026-07-20", "plan": "Free", "users": 1288},
    {"week": "2026-07-27", "plan": "Free", "users": 1402},
    {"week": "2026-07-06", "plan": "Team", "users": 420},
    {"week": "2026-07-13", "plan": "Team", "users": 468},
    {"week": "2026-07-20", "plan": "Team", "users": 515},
    {"week": "2026-07-27", "plan": "Team", "users": 559}
  ]},
  "mark": {"type": "line", "point": true},
  "encoding": {
    "x": {"field": "week", "type": "temporal", "title": "Week"},
    "y": {"field": "users", "type": "quantitative", "title": "Active users"},
    "color": {"field": "plan", "type": "nominal", "title": "Plan"}
  }
}

#A few things that catch people out

  • Set the field type to temporal, or dates are treated as unordered categories.

  • Do not connect categories with a line. That is a bar chart wearing a disguise.

  • Label the series directly if you can; legends cost the reader a lookup.

Pick the chart for the question, not for the look.