What Is a Bar Chart?
A bar chart compares numerical values for separate categories. Each category has a label; its bar extends from zero to the relevant value. Longer positive bars indicate larger values. A negative value extends in the opposite direction from zero.
For example, a shop could compare the units sold for five products. The product names identify categories, while the numerical axis shows how many units were sold. The position of a product label does not itself represent a quantity. It's the bar length, measured against the shared scale, that carries the comparison.
A bar graph and bar chart are usually the same thing. Vertical bar charts are also called column charts. A data visualization workspace can help when you need several chart types together; this free maker handles one focused comparison in your browser.
How to Make a Bar Chart Online
- Enter or import data. Edit the category/value table directly, paste CSV or spreadsheet cells, or choose a local CSV file. Imported rows are previewed before replacing your current entries.
- Choose the type. Use a simple vertical or horizontal chart for one series. Add another numeric series for grouped or stacked bars.
- Set labels and formatting. Give the chart a descriptive title, specify measurement units, choose an order and palette, and adjust the numeric display.
- Review the plot and summary. Check the maximum, minimum, totals and original values. If the tool flags a blank or invalid number, correct that cell before exporting.
- Download the result. Use PNG for slides or documents, SVG for scalable graphics, or CSV for the underlying data. Compatible browsers can also copy a PNG directly.
For long category labels, start with Horizontal bar. If order matters—such as months of the year—keep Original order rather than sorting by value.
Types of Bar Charts
Vertical bar chart
Each category appears along the horizontal axis, with values shown as upward or downward columns. A vertical layout is straightforward for a modest number of products, classes or departments. As names grow longer, labels require extra room.
Horizontal bar chart
Categories run down the left-hand side. Their values extend across the page. This is often the clearer option for survey answers, country names or service descriptions because labels need less rotation or abbreviation.
Grouped bar chart
Two or more series appear side by side within each category. For example, show sales in 2025 and 2026 for each product. This makes series comparisons easier but becomes crowded when many series and categories appear together.
Stacked bar chart
Each category's series share one bar. Positive parts accumulate above zero; negative parts accumulate below it. Stacking helps show component values and combined amounts, though middle segments are harder to compare directly because their start points differ.
100% stacked bar chart
The components within each category are divided by their total and scaled to 100%. This answers a composition question such as “What percentage of responses selected each option?” It does not show differences in category totals. This maker requires nonnegative values and positive totals so the percentages are defined.
Bar Chart Example With Data
Suppose these are illustrative unit sales for five products during one reporting period:
| Product | Units sold |
|---|
| Product A | 120 |
| Product B | 180 |
| Product C | 95 |
| Product D | 210 |
| Product E | 150 |
Product D has the highest count at 210 units. Product C has the lowest count at 95. Their difference is 115 units. Across all five products, the total is 755 units and the average is 151 units per product. Product D represents about 27.8% of the total (210 ÷ 755 × 100), assuming the categories are disjoint and measure the same thing.
Load Product sales (example) in the maker to reproduce this chart. A simple horizontal view with bars sorted from largest to smallest makes the ranking easy to scan. Do not describe these made-up figures as a real company's performance.
How to Read and Interpret a Bar Chart
Read the category labels first, then check the units and the numerical axis. A value of 40 “responses” cannot be compared meaningfully with 40 “dollars” on the same axis. Compare how far each bar reaches from the zero line, not how much colored area appears on the page.
Find the largest and smallest categories and note their actual difference. Sorting may reveal rankings, but it can conceal a natural sequence, such as January through December. When a chart has grouped series, compare bars of the same color across categories or different colors inside one category. For stacks, compare total heights and segment sizes with care.
Why the zero baseline matters: A bar uses length to represent magnitude. Starting the scale at 80 rather than zero would exaggerate the visible difference between values of 90 and 100. This maker always includes zero, including when the data contain negative values.
Bar Chart vs. Histogram
Bar charts work with distinct named categories. Histograms divide measurements from a continuous numeric variable into intervals called bins. The histogram's bars normally touch because neighboring bins cover adjacent stretches of the numeric scale.
Use a bar chart to compare departments or product types. Use the Histogram Maker to examine the distribution of measurements such as commute times, heights or test scores. If you want to compare the spread and outliers of several numeric groups instead of displaying one value per group, try the Box Plot Maker.
Bar Chart vs. Line Graph
A line graph joins observations in order, suggesting meaningful progression between successive positions. It is often preferable for a dense daily or monthly time series. A bar chart is usually clearer for discrete category comparisons or small sets of periods where the exact values are the focus.
If time is represented as categories in a bar chart, preserve chronological order. Sorting the months alphabetically or by sales changes the story of the sequence.
Common Bar Chart Mistakes to Avoid
- Truncating the numerical axis. For bar lengths, always start comparisons at zero.
- Mixing units. Don't put revenue and customer counts on the same numeric scale.
- Plotting too many categories. Use a shorter selection or break the comparison into several charts.
- Using unreadable category names. Choose horizontal bars for longer labels.
- Using too many colors. Reuse a consistent color for a single series and distinguish multiple series through a restrained palette and legend.
- Leaving units unclear. Specify whether values are dollars, percentages, responses or another measure.
- Misreading a 100% stack. Equal bar lengths there do not mean equal absolute totals.
- Overstating an observation. A difference between bars does not establish why it happened or whether it is statistically significant.
- Choosing a bar chart for a distribution. Use a histogram or box plot when the question concerns variation within numeric samples.
Our guide to common data visualization mistakes covers other problems to check before sharing a chart.
Best Practices for Clear Bar Charts
Give the title a specific subject and period: “Customer support tickets by channel, Q2” is more informative than “Results.” Show the measurement unit close to the axis. For percentages, verify whether the values already add up to 100 or whether normalization is needed.
Use one bar color for a single series unless a difference has a defined meaning. For multiple series, keep each series color consistent across categories. Try a horizontal layout when the category names would otherwise become difficult to read.
In research and education, preserve a table of exact source values alongside the graphic. In business presentations, sort categories when rank is the point, but keep periods in sequence. For surveys, distinguish the number of people from the percentage of respondents, and provide sample sizes when discussing rates.
Before exporting, compare at least two bars with the data table and read the chart at its final viewing size. A narrow mobile chart may need horizontal scrolling. If you are undecided about the format, consult how to choose the right chart type.