Scatter & Bubble Charts

Scatter and bubble charts visualize the relationship between numeric variables. Each point represents a data row positioned by its X and Y values. Bubble charts add a third dimension by sizing each point according to a third metric.

Data Requirements

Optional: A second dimension enables grouping—each unique value becomes a color-coded series with its own legend entry.

Explorer Workflow (Standard Mode)

Steps

1. Add dimensions – Drag your label field (e.g., Product Name) to Rows. To group points by category, add a second dimension (e.g., Region).
2. Add metrics – Drag three numeric fields to Metrics: X value, Y value, and size value. For scatter charts, only the first two are used.
3. Select chart type – Open the chart selector and choose Scatter or Bubble.
4. Configure options – Use the Bubble/Scatter Configuration panel to fine-tune radius scaling, axis titles, and grouping behavior.

SQL Workflow (Dev Mode)

In SQL mode, Semaphor uses positional column mapping. Arrange your SELECT columns in this order:

Scatter Chart Example

SELECT
    sub_category,                   -- label
    AVG(sales) AS avg_sales,        -- x-axis
    AVG(profit) AS avg_profit       -- y-axis
FROM orders
GROUP BY sub_category

| sub_category | avg_sales | avg_profit |
| ------------ | --------- | ---------- |
| Tables       | 443.93    | -253.54    |
| Chairs       | 328.45    | 45.12      |
| Phones       | 321.25    | 35.59      |

Switch the card type to Scatter. Semaphor automatically maps sub_category as the label, avg_sales as X, and avg_profit as Y.

Bubble Chart Example

SELECT
    sub_category,                   -- label
    AVG(sales) AS avg_sales,        -- x-axis
    AVG(quantity) AS avg_quantity,  -- y-axis
    AVG(score) AS product_score     -- bubble size
FROM orders
GROUP BY sub_category

Switch the card type to Bubble. The fourth column (product_score) now drives the bubble radius.

Grouping (Color-Coded Series)

Add a second dimension to create grouped series. Each unique value gets its own color, and the legend displays automatically.

Explorer Mode

Drag two dimensions to Rows:

• First dimension = point label (shown in tooltip)
• Second dimension = group (creates separate colored series)

SQL Mode

When grouping is enabled, include the group column as the second column:

SELECT
    category,                       -- label
    region,                         -- group (legend)
    AVG(sales) AS avg_sales,        -- x-axis
    AVG(profit) AS avg_profit,      -- y-axis
    AVG(score) AS score             -- bubble size
FROM orders
GROUP BY category, region

| category        | region  | avg_sales | avg_profit | score |
| --------------- | ------- | --------- | ---------- | ----- |
| Technology      | West    | 420.69    | 73.96      | 17.68 |
| Technology      | East    | 495.28    | 88.71      | 18.53 |
| Office Supplies | West    | 116.42    | 27.73      | 18.65 |
| Furniture       | Central | 340.53    | -5.97      | 17.51 |

Enable grouping in Configuration → Bubble/Scatter → Enable Grouping and select the group column. The legend appears on the right by default.

Customization Reference

Tooltip Behavior

Tooltips show all relevant information for each point:

• Title: Point label (and group name if grouped)
• X value: Metric name with (x) annotation
• Y value: Metric name with (y) annotation
• Size value: Metric name with (size) annotation (bubble only)

Best Practices

• Choose scatter for relationships: Use scatter charts when you want to show correlation between two variables without emphasizing magnitude.
• Choose bubble for three dimensions: Use bubble charts when a third metric (like market size or volume) adds meaningful context.
• Keep point counts reasonable: Charts with 50–200 points are ideal. More than 500 points can become cluttered.
• Use grouping sparingly: 2–5 groups work well. Too many groups make the legend overwhelming and colors hard to distinguish.
• Enable axis titles: When sharing dashboards, axis titles help viewers understand what each axis represents without checking the data source.