DUMBBELL

two dots + connector per category: before and after. Needs: categories with a value at two moments; the movement is the message.

deviationchange-over-timecomparison

DL-01 AGRAW

Use when

before/after per category: backlog W24 vs W28, score last year vs now
spotting the row that moved against the trend

Do not use when

more than two moments (CH-TIM-01)
two unrelated measures (CH-DST-02)

See instead

more than 2 time points: CH-TIM-01

Alternatives

CH-TIM-05 (few categories, slope reads better)

Rules

before = dot_a muted, after = dot_b primary; the connector is the story
labels state both values: '52 > 63'
highlight the row moving against the trend

Never do this

arrows instead of connectors (double-encodes direction)
legends
sorting that hides the counter-mover

QA before delivering

both values labeled per row
series labeled on first row
highlight <= 1
source present

Data shape

title	action title naming the mover with numbers
source	bottom-left always
unit	on first label
series_labels	[before_name, after_name]
categories	array of {label, values: [before, after], highlight?}

Sample data (golden)

{
  "title": "System D is the only system where punch grew, 52 to 63",
  "eyebrow": "OPEN PUNCH . W24 VS W28",
  "source": "Source: sample dataset v1, 2026-07-12",
  "unit": "items",
  "series_labels": [
    "W24",
    "W28"
  ],
  "categories": [
    {
      "label": "System A",
      "values": [
        28,
        12
      ]
    },
    {
      "label": "System B",
      "values": [
        31,
        18
      ]
    },
    {
      "label": "System C",
      "values": [
        39,
        25
      ]
    },
    {
      "label": "System D",
      "values": [
        52,
        63
      ],
      "highlight": true
    },
    {
      "label": "System E",
      "values": [
        36,
        31
      ]
    }
  ]
}