Free chart tool
A histogram is a chart that groups numeric data into adjacent, equal-width ranges, called bins, and shows how many values fall in each. Paste your numbers or upload a CSV below to get automatic bins, a frequency table, summary statistics, and group comparisons, free and without signing up.
Runs entirely in your browser. Your data is never uploaded.
Step 1
Examples (illustrative, generated data)
Separate numbers with commas, spaces, tabs, semicolons, or new lines, or paste a spreadsheet column. Negatives, decimals, and scientific notation such as 2.5e3 work. Up to 50,000 values.
CSV, TSV, or TXT up to 5 MB. The file is read in your browser and never uploaded.
300 values read.
Step 2
Step 3
| Bin | Count |
|---|---|
| [0, 1) | 16 |
| [1, 2) | 74 |
| [2, 3) | 62 |
| [3, 4) | 39 |
| [4, 5) | 34 |
| [5, 6) | 23 |
| [6, 7) | 17 |
| [7, 8) | 7 |
| [8, 9) | 6 |
| [9, 10) | 5 |
| [10, 11) | 6 |
| [11, 12) | 2 |
| [12, 13) | 1 |
| [13, 14) | 0 |
| [14, 15) | 2 |
| [15, 16) | 0 |
| [16, 17) | 1 |
| [17, 18) | 0 |
| [18, 19) | 1 |
| [19, 20) | 0 |
| [20, 21) | 0 |
| [21, 22) | 2 |
| [22, 23) | 0 |
| [23, 24) | 0 |
| [24, 25) | 0 |
| [25, 26) | 1 |
| [26, 27) | 0 |
| [27, 28) | 0 |
| [28, 29) | 0 |
| [29, 30) | 0 |
| [30, 31) | 0 |
| [31, 32) | 0 |
| [32, 33) | 0 |
| [33, 34) | 0 |
| [34, 35) | 0 |
| [35, 36) | 0 |
| [36, 37) | 0 |
| [37, 38) | 0 |
| [38, 39) | 0 |
| [39, 40) | 0 |
| [40, 41) | 0 |
| [41, 42) | 0 |
| [42, 43) | 0 |
| [43, 44] | 1 |
44 bins of width 1, from 0 to 44
Freedman-Diaconis: h = 2 x IQR x n^(-1/3) = 2 x 3.225 / 6.6943 = 0.9635; k = ceil(range / h) = ceil(42.8 / 0.9635) = 45; nice width h = 1
Boundary rule: each bin is [a, b), so it includes its lower edge and excludes its upper edge. The last bin is [a, b] so the maximum is counted. A value on an interior edge goes to the upper bin.
Images are 1200 px wide (PNG at 2x) on a white background. The CSV is safe to open in Excel or Google Sheets.
Privacy: your data never leaves this browser. It is saved only in this browser's local storage so it is still here next time.
The distribution looks strongly right-skewed (long tail toward higher values) (skewness 4.552). Skewness between -0.5 and 0.5 reads as roughly symmetric.
| Statistic | All values |
|---|---|
| n | 300 |
| Min | 0.3 |
| Q1 | 1.7 |
| Median | 2.9 |
| Q3 | 4.925 |
| Max | 43.1 |
| Mean | 4.026 |
| Standard deviation (sample) | 4.054 |
| IQR | 3.225 |
| Skewness | 4.552 |
| Outliers (1.5 x IQR) | 20 |
Quartiles and the median use linear interpolation between sorted values (type 7, the same as Excel QUARTILE.INC and R). Skewness is the adjusted Fisher-Pearson coefficient (Excel SKEW). Outliers fall below Q1 - 1.5 x IQR or above Q3 + 1.5 x IQR.
| Bin | Midpoint | Count | Percent | Cum. count | Cum. percent | Density |
|---|---|---|---|---|---|---|
| [0, 1) | 0.5 | 16 | 5.3% | 16 | 5.3% | 0.05333 |
| [1, 2) | 1.5 | 74 | 24.7% | 90 | 30.0% | 0.2467 |
| [2, 3) | 2.5 | 62 | 20.7% | 152 | 50.7% | 0.2067 |
| [3, 4) | 3.5 | 39 | 13.0% | 191 | 63.7% | 0.13 |
| [4, 5) | 4.5 | 34 | 11.3% | 225 | 75.0% | 0.1133 |
| [5, 6) | 5.5 | 23 | 7.7% | 248 | 82.7% | 0.07667 |
| [6, 7) | 6.5 | 17 | 5.7% | 265 | 88.3% | 0.05667 |
| [7, 8) | 7.5 | 7 | 2.3% | 272 | 90.7% | 0.02333 |
| [8, 9) | 8.5 | 6 | 2.0% | 278 | 92.7% | 0.02 |
| [9, 10) | 9.5 | 5 | 1.7% | 283 | 94.3% | 0.01667 |
| [10, 11) | 10.5 | 6 | 2.0% | 289 | 96.3% | 0.02 |
| [11, 12) | 11.5 | 2 | 0.7% | 291 | 97.0% | 0.006667 |
| [12, 13) | 12.5 | 1 | 0.3% | 292 | 97.3% | 0.003333 |
| [13, 14) | 13.5 | 0 | 0.0% | 292 | 97.3% | 0 |
| [14, 15) | 14.5 | 2 | 0.7% | 294 | 98.0% | 0.006667 |
| [15, 16) | 15.5 | 0 | 0.0% | 294 | 98.0% | 0 |
| [16, 17) | 16.5 | 1 | 0.3% | 295 | 98.3% | 0.003333 |
| [17, 18) | 17.5 | 0 | 0.0% | 295 | 98.3% | 0 |
| [18, 19) | 18.5 | 1 | 0.3% | 296 | 98.7% | 0.003333 |
| [19, 20) | 19.5 | 0 | 0.0% | 296 | 98.7% | 0 |
| [20, 21) | 20.5 | 0 | 0.0% | 296 | 98.7% | 0 |
| [21, 22) | 21.5 | 2 | 0.7% | 298 | 99.3% | 0.006667 |
| [22, 23) | 22.5 | 0 | 0.0% | 298 | 99.3% | 0 |
| [23, 24) | 23.5 | 0 | 0.0% | 298 | 99.3% | 0 |
| [24, 25) | 24.5 | 0 | 0.0% | 298 | 99.3% | 0 |
| [25, 26) | 25.5 | 1 | 0.3% | 299 | 99.7% | 0.003333 |
| [26, 27) | 26.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [27, 28) | 27.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [28, 29) | 28.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [29, 30) | 29.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [30, 31) | 30.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [31, 32) | 31.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [32, 33) | 32.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [33, 34) | 33.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [34, 35) | 34.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [35, 36) | 35.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [36, 37) | 36.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [37, 38) | 37.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [38, 39) | 38.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [39, 40) | 39.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [40, 41) | 40.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [41, 42) | 41.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [42, 43) | 42.5 | 0 | 0.0% | 299 | 99.7% | 0 |
| [43, 44] | 43.5 | 1 | 0.3% | 300 | 100.0% | 0.003333 |
Drag the slider to redraw your data with more or fewer equal-width bins. The data stays the same; only the bins change.
10 bins of width 4.28. Too few bins hide the shape of your data; too many show random noise. Aim for the fewest bins that still show the peaks and tails.
Step 1
Paste numbers separated by commas, spaces, tabs, or new lines, or upload a CSV and pick the numeric column. Add an optional group column, such as plan, to compare segments.
Step 2
Keep the default Freedman-Diaconis rule or pick Sturges, square root, Rice, Scott, a custom bin count, or a custom width and start. The tool shows the formula, bin count, and bin width.
Step 3
Check where values cluster, how wide they spread, and which way the tail points. Switch between count, percent, and density, and add a cumulative line or a reference normal curve.
Step 4
Download the histogram as SVG or PNG, download the frequency table as CSV, or copy the table as Markdown or TSV and the summary statistics as text.
n is the number of values, k the number of bins, h the bin width, SD the sample standard deviation, and IQR the interquartile range. Width rules turn h into k = ceil((max - min) / h). Every bin is [a, b), including its lower edge, except the last bin, which is [a, b] so the maximum is counted.
| Rule | Formula | Best for |
|---|---|---|
| Freedman-Diaconis | h = 2 x IQR x n^(-1/3) | Default. Robust to outliers and skewed data such as response times. |
| Sturges | k = ceil(log2(n)) + 1 | Small, roughly bell-shaped samples. Too few bins for large n. |
| Square root | k = ceil(sqrt(n)) | A quick, simple starting point. |
| Rice | k = ceil(2 x n^(1/3)) | A middle ground between Sturges and square root for larger samples. |
| Scott | h = 3.49 x SD x n^(-1/3) | Data that is close to normal. Excel's histogram chart uses it by default. |
Response times are almost always right-skewed. A histogram shows the long tail of slow replies that an average hides, so report the median and the share of tickets beyond your target.
Two products can share the same average score while one is polarized and the other is lukewarm. A histogram of 0 to 10 or 1 to 5 ratings shows which, and an NPS calculator turns the same ratings into one number.
Plot minutes of use or events per account, split by plan or cohort on the same bins, to see whether a feature has a core of power users or broad light use.
A histogram is a chart that groups numeric data into adjacent, equal-width ranges called bins and shows how many values fall in each bin. The bar heights reveal the shape of the data: where values cluster, how spread out they are, whether the distribution is skewed, and whether there are outliers.
A histogram shows the distribution of one numeric variable, so its bars touch and sit on a continuous number line where each bar covers a range such as 10 to 20. A bar graph compares separate categories, such as features or survey answers, so its bars have gaps and can be reordered. Use a histogram for measurements like response times and a bar graph for counts per category.
There is no single correct number. Common rules are Sturges (k = ceil(log2 n) + 1), the square root rule (k = ceil(sqrt n)), Rice (k = ceil(2 x n^(1/3))), Scott (bin width 3.49 x SD x n^(-1/3)), and Freedman-Diaconis (bin width 2 x IQR x n^(-1/3)). Freedman-Diaconis is robust to outliers and skew, so this tool uses it by default and falls back to Sturges when the IQR is 0. Try a few bin counts: too few hide the shape and too many show noise.
Look at the center (where the tallest bars are), the spread (how wide the bars extend), and the tails. If the long tail stretches to the right toward higher values, the data is right-skewed and the mean is usually above the median, which is typical of response times and revenue. A long left tail means left-skewed. This tool reports sample skewness and calls the shape roughly symmetric when it is between -0.5 and 0.5.
In Excel 2016 and later, select your numbers and choose Insert, Insert Statistic Chart, Histogram; by default Excel picks the bin width with Scott's rule, and you can set the bin width or count under Format Axis. In Google Sheets, select the column, choose Insert, Chart, and pick Histogram chart as the chart type, then adjust the bucket size. Excel's histogram chart puts a value that sits exactly on an edge in the lower bin, while this tool puts it in the upper bin, so edge counts can differ. This tool also adds a frequency table, summary statistics, and group comparisons without a spreadsheet.
Chart counts per category, such as votes per feature, with gaps between bars.
Open toolCheck whether a difference between two groups is real or noise.
Open toolCompare the means of three or more groups at once.
Open toolCompare spread across data sets with different scales.
Open toolBuild balanced rating scales whose answers you can chart here.
Open toolTurn 0 to 10 ratings into a Net Promoter Score.
Open toolFeatureVote gives you a public board where customers suggest ideas and vote on them, so you can see which requests have broad support and which come from a vocal few, then tell voters when features ship.
Try FeatureVote free