A histogram shows how the individual values of a single numeric measure are distributed, by splitting the full range of values into consecutive buckets (bins) and drawing a bar for how many records fall into each bin. Unlike a bar chart, a histogram never aggregates data per category - it always works on the raw, unaggregated value of every record, because the shape of the distribution is the entire point.

When should I use a histogram instead of a bar chart?

Use a histogram when the question is about the shape of a single measure's distribution - is it skewed, clustered, normal, bimodal, or full of outliers. Use a bar chart when the question is about comparing a measure across distinct categories. A histogram's bins are computed ranges of one continuous variable; a bar chart's categories are discrete, named groups.

What does the optional dimension field do on a histogram?

Assigning a categorical field to Column splits the histogram into one overlapping distribution per category, all binned against the same shared ranges so they can be fairly compared - for example, order value distribution by region. Every category's fill is drawn with reduced opacity so overlaps remain visible, and a legend identifies each one. Leaving the dimension empty produces a single pooled distribution across every record.

Why does the histogram require the measure to be unaggregated?

A histogram bins individual record values - if the measure were pre-aggregated (for example, summed per category), there would only be one number per category left to bin, and the chart would show a meaningless handful of bars instead of a real distribution. For this reason, the Rows field on a histogram always uses raw values and has no aggregation dropdown to change - it reads (Default) because there is no other valid choice.

What is the difference between the four Bin Methods?

Sturges (the default) sets bin count based on the log of the sample size - a reasonable general-purpose choice for moderately sized, roughly symmetric data. Square Root sets bin count to the square root of the sample size - simpler, often produces more bins on large datasets. Scott's rule and the Freedman-Diaconis rule both compute bin width from the data's spread (standard deviation for Scott, interquartile range for Freedman-Diaconis) and are more robust when the data has outliers or is not symmetric, since Freedman-Diaconis in particular is less sensitive to extreme values.

Why does my histogram show one giant bar and many tiny ones?

This happens when the measure has a long tail - a small number of extreme outlier values stretch the full axis range across every bin, while most of the real data clusters in the first bin. Common with prices, revenue, or any right-skewed business metric. Pre-filter extreme outliers in Transform, or switch Bin Method to Freedman-Diaconis, which is less sensitive to outliers when computing bin width.

Can I set the number of bins manually?

Yes. Bin Count overrides whatever the selected Bin Method would have computed automatically. Leave it empty for an automatic bin count based on Bin Method, or set a specific value from 1 to 200 if you need a fixed number of bins regardless of sample size.

How many bins should a histogram have?

Most distributions read well with somewhere between 8 and 20 bins - too few bins hides real structure in the data by lumping different patterns together, while too many bins makes the chart noisy and turns genuine patterns into visual static. The default Bin Method (Sturges) targets a reasonable count automatically based on sample size, so manual adjustment is usually only needed for side-by-side comparison or when a dataset's specific shape calls for finer or coarser detail.

Does the histogram show a gap between bars?

No, by default. Bins represent a continuous numeric range, so bars touch each other with no gap, unlike a regular bar chart where categories are visually separated. This gap is controlled by Bar Gap in Bar Styles if you want to add spacing.

Histogram | Edilitics Docs

Show how a single measure's values are distributed across automatically computed ranges. Best for spotting skew, clusters, and outliers in raw numeric data.

A histogram - also called a frequency distribution chart - shows how the individual values of a single numeric measure are distributed, by splitting the full range of values into consecutive bins and drawing a bar for how many records fall into each bin. Drop one numeric field into Row - no aggregation, since the chart needs every record's raw value, not a pre-aggregated number. Optionally drop one categorical field into Column to compare multiple distributions side by side. Use it when the question is about the shape of the data itself - skew, clustering, outliers - not a comparison across named categories.

When to Use

A histogram answers a fundamentally different question than every other chart in this module. A bar chart compares a measure across discrete categories you already know. A histogram reveals the shape of one continuous measure you don't yet understand - whether order values cluster near a typical price with a few expensive outliers, whether response times are mostly fast with a long slow tail, or whether a measure is roughly symmetric, skewed, or has two separate peaks.

The measure must stay unaggregated. A histogram bins individual record values - it has nothing to bin if the measure has already been collapsed into one number per group. This is why the Rows field never shows an aggregation choice: there is no legitimate alternative to raw values for this chart type. If you need to compare a summed or averaged measure across categories, that is a bar chart question, not a histogram question.

Switch to a different chart when:

The question is a comparison across named categories, not a distribution - use Horizontal Bar
You need the distribution summarized as quartiles and outliers rather than a full bin-by-bin shape - use a Box Plot
You want to see the relationship between two measures, not the shape of one - use Scatter Plot
Your dimension is time and the question is about trend - use Line Chart

Scenario	Dimension (optional)	Measure
Distribution of order values	None	Order value (raw)
Order value distribution by region	Region name	Order value (raw)
Distribution of page load times	None	Load time in seconds (raw)
Customer age distribution by acquisition channel	Acquisition channel	Customer age (raw)
Distribution of support ticket resolution time	None	Resolution time in hours (raw)

Required Inputs

Field	Type	Count
Dimension	Categorical	0 to 1
Measure	Numeric	Exactly 1

The measure field always uses raw, unaggregated values - the (Default) label confirms this, with no dropdown to change it. The optional dimension, when assigned, always groups the distribution by category - confirmed by the (Group By) label.

For step-by-step build instructions, see Build Your First Chart.

Formatting Options

The Format tab unlocks after the measure field is assigned. Once your chart is rendering, use these controls to define how it looks and how viewers interact with it.

Style

Use the chart title to state what measure the distribution represents - viewers need this before reading bin labels on the X-axis.

Control	What it does
Show Chart Title	Shows or hides the title. The text is preserved when hidden so you can toggle it back without re-entering.
Enter Chart Title	Title text. Maximum 50 characters.
Font family	Font applied to the title.
Font size	5 to 30.
Bold / Italic	Weight and style.
Alignment	Left, center, or right within the chart container.

Histogram controls determine how the measure's range is split into bins - the single most important decision on this chart type, since it directly shapes what pattern viewers see.

Control	What it does
Bin Method	Sturges (default): bin count based on the log of the sample size, a reasonable general-purpose choice. Square Root: bin count is the square root of the sample size, simpler and often produces more bins on large datasets. Scott: bin width computed from standard deviation, a good fit for roughly symmetric data. Freedman-Diaconis: bin width computed from the interquartile range, more robust when outliers are present since it does not let extreme values dominate the width calculation.
Bin Count	Overrides Bin Method's automatic calculation with a fixed number of bins, 1 to 200. Leave empty for automatic. Set this manually when you need a consistent bin count across multiple histograms for visual comparison.

Control

What it does

Bin Method

Sturges (default): bin count based on the log of the sample size, a reasonable general-purpose choice. Square Root: bin count is the square root of the sample size, simpler and often produces more bins on large datasets. Scott: bin width computed from standard deviation, a good fit for roughly symmetric data. Freedman-Diaconis: bin width computed from the interquartile range, more robust when outliers are present since it does not let extreme values dominate the width calculation.

Bin Count

Overrides Bin Method's automatic calculation with a fixed number of bins, 1 to 200. Leave empty for automatic. Set this manually when you need a consistent bin count across multiple histograms for visual comparison.

Bar Styles controls the visual appearance of the bins - color, spacing, border, and corner rounding. With a dimension assigned, each category gets its own color from this panel's gradient; the gap and radius controls still apply uniformly across all categories.

Control	What it does
Bar Color	Fill color of the bars. Used directly when no dimension is assigned.
Bar Gap	Space between bins, 0 to 5. Default is 0 - bins touch, since they represent a continuous range rather than separate categories. Increase only if you want visual separation between bins.
Border Color	Color of the bar outline.
Border Width	Thickness of the bar outline, 0 to 3. At 0, no outline is drawn.
Corner Radius (top-left, bottom-left, top-right, bottom-right)	Rounds bar corners independently, 0 to 10 each. Do not push corners high on narrow bins - it distorts how the bin width reads visually.

When a dimension is assigned, every category's bars overlap fully at each bin with reduced opacity, so multiple distributions can be visually compared at once - this overlap is fixed and not adjustable. Bar Gap and Corner Radius from this panel still apply to every category's bars uniformly.

Data labels print the count above each bin. On a histogram with many bins, labels can get crowded - use them when the bin count is small enough that every label has room.

Control	What it does
Show Data Labels	Shows or hides count labels above each bin.
Font family	Font applied to data labels.
Font size	Maximum 18. Keep small with many bins.
Bold / Italic / Color	Text styling.
Position	Vertical and horizontal alignment of the label relative to the bar.
Number Type	Default, Scientific, Decimal, or Percentage.
Display Unit	None, Thousand, Million, or Billion.
Decimal Places	0 to 6. Available when Number Type is not Default.

Axes

The X-axis carries the bin ranges of the measure; the Y-axis is always the count of records in each bin. Name the X-axis with the measure's unit so viewers know what range each bin represents.

Select X or Y before making changes. Settings apply to the selected axis only.

Control	What it does
Show Axis Name	Shows or hides the axis name label.
Axis name text	Label text. Maximum 20 characters. Defaults to the field name.
Font family	Font applied to the axis name.
Font size	5 to 30.
Bold / Italic	Weight and style.
Offset (%)	Distance between the axis name and the axis line. Increase if the name overlaps the axis labels.
Alignment	Start, center, or end along the axis.

Leave these at their defaults unless you are building a minimal dashboard layout.

Select X or Y before making changes.

Control	What it does
Show Axis Line	Shows or hides the axis line. Hiding it also hides ticks.
Line Color	Color of the axis line.
Line Width	Thickness of the axis line, 0 to 5.
Line Type	Solid, Dashed, or Dotted.
Show Ticks	Shows or hides tick marks on the axis line.
Tick Color	Color of the tick marks.
Tick Length	Length of the tick marks, 5 to 10.

X-axis labels show each bin's range (for example, "0-12"). With many bins, these labels can overlap - the chart automatically thins overlapping labels, but reducing font size or bin count also helps.

Select X or Y before making changes. Hiding axis labels also hides the axis name.

Control	What it does
Show Axis Label	Shows or hides labels along the axis.
Font family	Font applied to axis labels.
Font size	Label font size.
Bold / Italic / Color	Text styling.
Rotation	Angle of label text. Useful on the X-axis when bin range labels are long.
Display Unit	Auto, None, Thousand, Million, or Billion. Available on the Y-axis (count) only.

Y-axis grid lines help viewers read bin counts against the scale. X-axis grid lines running through bins add visual noise and are rarely needed.

Select X or Y before making changes.

Control	What it does
Show Gridlines	Shows or hides grid lines for the selected axis.
Grid Color	Color of the grid lines.
Left / Right / Top / Bottom	Plot area margins as a percentage. Increase Top when data labels above the tallest bin are clipped.

Interactivity

The tooltip shows the bin's range and its count. With a dimension assigned, every overlapping category's count for the hovered bin appears together in one tooltip, identified by name.

Control	What it does
Show Tooltip	Shows or hides the tooltip on hover.
Background Color	Tooltip background color.
Headers / Values tabs	Separate styling for the bin range row (Headers) and the count rows (Values). The Values tab includes Number Type, Display Unit, and Decimal Places.

Keep animation on for presentations - the bars building up draws attention to the overall shape as it forms. Turn it off on dashboards that auto-refresh.

Control	What it does
Enable Animation	Turns the build animation on or off.
Duration	How long the animation runs, 0 to 3000ms. Default is 1000ms.
Delay	Time before the animation starts after the chart loads, 0 to 2000ms.
Easing Function	The motion curve. Cubic Out (the default) gives a natural deceleration.

Enable Data Zoom when a high bin count makes the chart wide and hard to read fully without scrolling.

Control	What it does
Show Data Zoom	Shows or hides the zoom control.
Zoom Type	Slider: a draggable range bar below the chart. Inside: scroll-to-zoom on the plot area. Use Slider for dashboards.
Slider Size (px)	Height of the slider bar, 10 to 100.
Position	Position of the slider relative to the chart.
Alignment	Alignment of the slider control.
Orientation	Horizontal for a histogram - scrolls through the bin range.
Show Detail Label	Shows the bin range values at the slider handles.
Background Color	Slider background.
Filler Color	Color of the selected range inside the slider.
Border Color	Slider border.
Handle Color	Color of the drag handles.
Brush Selection	Enables click-and-drag to zoom into a range.

With a dimension assigned, Series focus highlights one category's bars across every bin, useful for tracing a single distribution among several overlapping ones.

Control	What it does
Show Emphasis	Enables or disables the hover effect.
Focus Type	Item: highlights the hovered bar only. Series: highlights the entire hovered category across all bins - use this when comparing multiple distributions. None: no visual change on hover.
Enable Scale	Scales the hovered bar slightly larger.
Scale Size	How much the bar scales. Keep at 1.2 or below.

Enable the Toolbox when viewers need to export the chart or inspect the exact bin counts behind the shape.

Control	What it does
Show Toolbox	Shows or hides the toolbox icon bar.
Save as Image	Adds a download icon that saves the chart as a PNG.
Data View	Adds an icon that opens the underlying data table in the chart area.

Best Practices

Never aggregate the measure before assigning it. The Rows field always reads raw values for a histogram - there is nothing to configure here, but understand why: a histogram needs the full set of individual record values to compute a meaningful distribution. A sum or average per category leaves nothing to bin.

Watch for long-tail outliers squashing the chart. A handful of extreme values - one very expensive order, one very slow request - can stretch the axis range so far that almost everything else lands in the first bin. If your histogram looks like one giant bar and a flat line afterward, this is almost always the cause. Pre-filter extreme outliers in Transform, or switch Bin Method to Freedman-Diaconis, which resists this distortion better than Sturges or Square Root.

Use the optional dimension to compare distributions, not to categorize a total. Splitting a histogram by category answers "does this distribution look different across groups" - it is not a substitute for a bar chart's per-category totals. If the real question is "which category has the highest total," that's a bar chart question.

Keep grouped histograms to 3 or 4 categories. Beyond that, overlapping fills become hard to distinguish even with reduced opacity and a legend. Pre-filter to the most relevant categories in Transform if you have more.

Set Bin Count manually when comparing histograms side by side. If you are placing two histograms next to each other on a dashboard and want their bin structures to align visually, set the same fixed Bin Count on both rather than relying on automatic calculation, which can produce different bin counts for different sample sizes.

FAQs

Need help? Email support@edilitics.com with your workspace, job ID, and context. We reply within one business day.

Histogram

What is a histogram?