Excel Chart Types: Which One Should You Use?
The right Excel chart type depends on what question you're answering: use a column or bar chart to compare categories, a line chart to show a trend over time, a scatter chart to show a relationship between two numeric variables, and a pie chart only for a part-to-whole breakdown with five or fewer slices. Picking the wrong type doesn't just look bad โ it actively hides the pattern in the data.
Quick answer: Comparing categories โ column/bar chart. Trend over time โ line chart. Relationship between two numbers โ scatter (XY) chart. Part of a whole โ pie chart (5 or fewer slices) or a 100% stacked bar. Two measures with different units โ combo chart with a secondary axis. Avoid 3D charts and pie charts with more than 5-6 slices; both distort the reader's judgment of size.
Which chart type should I use to compare categories?
Use a clustered column chart for comparing values across categories (regions, products, reps) when you have up to about 10-15 categories; switch to a horizontal bar chart once category labels get long, since bars give the labels room to run left to right instead of being squeezed or rotated.
Column chart: Insert > Charts > Insert Column or Bar Chart > Clustered Column
Bar chart: Insert > Charts > Insert Column or Bar Chart > Clustered Bar
Both must start their value axis at zero. Truncating the axis to "zoom in" on small differences is the single most common way Excel charts mislead a reader โ a bar that looks 3x taller than another should mean the value is roughly 3x larger, not 1.1x larger with a cropped axis.
Which chart type should I use to show a trend over time?
Use a line chart whenever the x-axis is a continuous timeline (dates, months, quarters) and you have more than about 8-10 points; the connecting line is what makes the trajectory readable. For a handful of time periods, a column chart works just as well and avoids implying interpolation between points that don't really connect.
Line chart: Insert > Charts > Insert Line or Area Chart > Line
Best for: daily/weekly/monthly series, 8+ data points, one or a few series
For a compact, inline version of the same idea โ a tiny trend indicator inside a cell rather than a full chart โ see Excel sparklines.
Which chart type shows a relationship between two numeric variables?
Use a scatter (XY) chart, not a line chart, whenever both axes are numeric and you're testing whether two measures move together โ price vs. demand, ad spend vs. conversions, hours studied vs. score. A line chart on numeric x-values wrongly implies the x-axis is a sequence; a scatter chart plots each point independently and lets you add a trendline to see the relationship.
Scatter chart: Insert > Charts > Insert Scatter (X, Y) or Bubble Chart > Scatter
Add a trendline: right-click any point > Add Trendline > Linear (tick "Display R-squared value")
When is a pie chart appropriate in Excel?
A pie chart works only when you have one category field, values that sum to a meaningful whole (100% of something), and five or fewer slices; beyond that, humans can't compare wedge angles accurately and the chart becomes a legend with decoration. If you have more categories or need to compare across several groups, use a 100% stacked bar or a simple ranked bar chart instead โ see pie chart alternatives for the reasoning.
Which chart type should I use for two measures with different units?
Use a combo chart โ one series as columns, the other as a line on a secondary axis โ when you're pairing measures like revenue (dollars) and order count (a count), where a shared axis would flatten one series to a nearly flat line. The secondary axis needs its own scale and its own axis title, or readers won't know which line reads on which axis.
Full setup steps, including when a secondary axis quietly misleads the reader, are in the combo chart guide; the general risk of any second axis is covered in dual axis charts: when two y-axes mislead.
How do I pick a chart type in Excel quickly?
Select your data and use Excel's Recommended Charts (Insert > Charts > Recommended Charts), which suggests types based on the shape of the selected range โ it's a reasonable starting point, but it doesn't know your intent, so treat it as a shortlist rather than the final answer.
| Chart type | Best for | Avoid when |
|---|---|---|
| Clustered column/bar | Comparing values across categories | More than ~15 categories |
| Line | Trend over a continuous timeline | Fewer than ~8 points, categorical x-axis |
| Scatter (XY) | Relationship between two numeric variables | One axis is categorical |
| Pie / doughnut | Part-to-whole, 5 or fewer slices | More than ~6 slices, comparing multiple pies |
| 100% stacked bar | Part-to-whole compared across several groups | Reader needs to compare exact segment sizes |
| Combo (bar + line) | Two related measures in different units | Both series share the same unit and scale |
| Waterfall | Cumulative effect of sequential positive/negative changes | Values aren't sequential contributions to a total |
Which Excel chart types should I avoid?
Avoid any 3D chart and pie charts with more than five or six slices โ both distort how accurately a reader can compare sizes, and Excel's own chart gallery still ships them because they were popular in the 2000s, not because they communicate well.
- 3D column/pie/bar: perspective distorts which bars or slices look bigger, purely based on their position in the depth field.
- Pie charts with 8+ slices: the smallest wedges become unreadable; group the tail into "Other" or switch to a bar chart.
- Dual-axis line charts with a manipulated scale: a secondary axis with a cherry-picked range can make two unrelated series appear to move together.
- Radar charts for more than 2-3 series: overlapping polygons become unreadable fast; a small-multiples grid of bar charts usually communicates the same comparison better.
Pro Tip: Right-click any chart and choose "Change Chart Type" to try alternatives on your actual data before committing โ the visual difference between a line chart and a scatter chart on the same numbers is often more convincing than any rule of thumb.
โ Back to Excel Tips