Stacked vs Grouped Bar Chart
COMPAREBoth show several series at once. One lines them up for comparison; the other stacks them into a total.
The difference: a grouped bar chart (also called a clustered bar chart) places each series' bar side by side within a category, all starting from zero, so you compare series directly. A stacked bar chart piles the series on top of one another, so each full bar shows the category total and its parts.
When you have more than one series to plot against the same set of categories, a plain bar chart is not enough — you need to decide how to arrange the extra bars. The two standard answers are the grouped bar chart, where series sit side by side, and the stacked bar chart, where series pile on top of each other. The choice hinges on a single question: do you care more about comparing the series or about seeing the total?
Use a grouped bar chart when you want to compare the series directly — every bar starts from the same zero baseline, so the lengths are easy to read against each other. Use a stacked bar chart when the total per category matters and you also want to show its parts — the full bar height is the total, and each segment is one series' contribution.
The core difference
In a grouped chart, every bar in a category begins at zero. Because they all share that common baseline, comparing them is effortless — the eye is excellent at judging lengths that start from the same line. The cost is that the chart never shows a combined total; you would have to add the bars up yourself.
In a stacked chart, the series are laid one on top of the next, so the height of the whole bar is the sum of its parts. This makes the total instantly visible and is perfect for "parts of a whole" within each category. But there is a catch that trips many people up: only the bottom segment sits on the shared zero baseline. Every segment above it floats at a different starting height in each bar, so comparing those middle segments across categories is genuinely hard. You can see the total clearly and the bottom series clearly — the rest, much less so.
A worked example with real numbers
The figure above uses this small dataset — sales (in units) of two products over two quarters:
| Quarter | Product A | Product B | Total |
|---|---|---|---|
| Q1 | 70 | 50 | 120 |
| Q2 | 90 | 40 | 130 |
Reading the grouped version: in each quarter the two bars stand next to each other on the same baseline. You can see at once that Product A beat Product B in both quarters and that the gap widened in Q2 (20 units, then 50). What you cannot see directly is that total sales rose from 120 to 130 — you would have to add the bars in your head.
Reading the stacked version: each quarter is a single bar 120 and 130 units tall, so the growth in the total is obvious. Product A sits at the bottom, so its rise from 70 to 90 is easy to judge. Product B floats on top of A, starting at 70 in Q1 and at 90 in Q2; to see that B fell from 50 to 40 you have to compare two segments with different starting heights — exactly the weakness described above.
Side-by-side comparison
| Grouped bar chart | Stacked bar chart | |
|---|---|---|
| Series arranged | Side by side | On top of each other |
| Shared baseline | Every bar starts at zero | Only the bottom segment |
| Shows the total | No | Yes (full bar height) |
| Compare series directly | Easy | Hard above the bottom |
| Parts of a whole | Not shown | Shown per category |
| Crowding | Grows with more series | Stays compact |
| Best for | Comparing few series | Totals split into parts |
Stacked vs 100% stacked bar chart
A 100% stacked bar chart is a stacked bar chart in which every bar is stretched to the same height, representing 100%. Each segment shows its percentage share of that bar's total rather than its raw value. In the example above, Product A would be 70 ÷ 120 = 58% of Q1 and 90 ÷ 130 = 69% of Q2, while Product B would be 42% and 31%.
- Regular stacked bars answer "how big is each total, and what is it made of?" Bar heights differ because totals differ.
- 100% stacked bars answer "how does the mix compare across categories?" Totals are hidden on purpose, so a small category and a huge one get bars of equal height.
Use the 100% version when the share is the story — for example, the proportion of survey answers by age group, or how the product mix changed from year to year. Add the totals as labels if readers might assume every bar represents the same amount. The stacked bar chart guide covers both variants in more depth.
Is a clustered bar chart different from a bar chart?
A clustered bar chart is still a bar chart — it is the multi-series version of one. A plain bar chart has one bar per category. A clustered (grouped) chart has a cluster of bars per category, one for each series, and uses colour plus a legend to tell the series apart. "Grouped" and "clustered" are two names for the same thing; Excel, for instance, calls it "Clustered Bar" or "Clustered Column". If you only have one series, there is nothing to cluster and a regular bar chart is the right choice. See the grouped bar chart guide for how to read one.
How the two are built in common tools
The data for both charts is identical — a table with categories in rows and one column per series. Only the chart setting changes:
- Excel and Google Sheets: pick "Clustered Column/Bar" for grouped, "Stacked Column/Bar" for stacked, and "100% Stacked" for the percentage version. You can switch between them without touching the data.
- Matplotlib (Python): grouped bars are drawn by shifting each series' x positions by a bar width; stacked bars are drawn at the same x positions with the
bottomargument set to the running total of the series below. - Chart.js: bar datasets are grouped by default; setting
stacked: trueon both the x and y scales stacks them.
Because the data layout is the same, it costs nothing to try both and keep the one that answers your reader's question.
When the grouped bar chart wins
Group the bars when the comparison between series is the headline:
- You want a direct, fair comparison. Every bar starts at zero, so readers can judge which series is larger in each category at a glance.
- You have only a few series. Two or three per category stay readable; more than that and each group gets cluttered.
- The total is not the point. You care about how the individual series stack up against each other, not about their sum.
When the stacked bar chart wins
Stack the bars when the total and its composition both matter:
- The whole is meaningful. Total revenue, total responses, total hours — and you want that total visible per category.
- You are showing parts of a whole. Each segment is a slice of the bar's total, like categories that make up overall sales.
- You have many series that would crowd a grouped chart; stacking keeps each category to a single compact bar.
Using a stacked bar chart when the reader actually needs to compare the middle series across categories. Those floating segments do not share a baseline, so the comparison is unreliable — readers will misjudge them. If comparing every series matters more than the total, switch to grouped bars, where everything starts from zero.
The decision rule
Ask what the reader must do. If they need to compare the series against each other, use grouped bars so every value shares a zero baseline. If they need to see a total and how it breaks down, use stacked bars — and order your segments so the most important series sits on the bottom, where it is easiest to read.
Frequently asked questions
What is the difference between a grouped bar chart and a stacked bar chart?
In a grouped bar chart the series sit side by side, each starting from the same zero baseline, so you can compare them directly. In a stacked bar chart the series are placed on top of one another, so the full bar shows a total and each segment shows that series' share of it.
What is the difference between a bar chart and a clustered bar chart?
A bar chart shows one bar per category for a single series. A clustered bar chart, also called a grouped bar chart, shows a cluster of bars per category, one for each series, so you can compare several series within and across categories.
What's the difference between a stacked bar chart and a 100% stacked bar chart?
A stacked bar chart shows raw values, so bar heights differ and the total is visible. A 100% stacked bar chart scales every bar to the same full height and shows each segment as a percentage of its bar, which makes the mix easy to compare but hides the totals.
When should I use a stacked bar chart?
Use a stacked bar chart when the total matters and you also want to show what makes it up, such as total sales split by product. It works best with a few segments and when readers mainly care about the total and the bottom segment, since only the bottom segment shares a common baseline.
Why is it hard to compare middle segments in a stacked bar chart?
Because every segment except the bottom one starts at a different height in each bar, the eye has no common baseline to line them up against. Lengths floating at different starting points are much harder to compare than lengths that all begin at zero, which is why grouped bars are better for direct comparison.
What is the difference in syntax between a grouped bar and a stacked bar chart?
The data is the same; only one setting changes. In Excel or Google Sheets you choose Clustered versus Stacked as the chart type. In Matplotlib you offset the x positions for grouped bars or pass a bottom value for stacked bars. In Chart.js you set stacked to true on the axes.
Build either one
The stacked bar chart maker supports both layouts — toggle between stacked and grouped to see which fits your data. It runs in your browser and exports PNG or SVG with no signup.