Here is the short answer, because you probably came here to settle an argument or a slide: use a bar chart to compare distinct things, and a line chart to show how one thing changes over time. Sales by region is bars. Revenue by month is a line. If that resolves it, you can stop reading and go build the thing.
But you are probably still here, which means you have hit one of the cases where the rule blurs, and those cases are where most charts go wrong. Revenue by month sounds like a line, but with only four data points a bar chart is often clearer. A line chart connecting product categories is one of the most common chart mistakes there is, and it looks subtly wrong to everyone who sees it even when they cannot say why.
This is a practical guide to making the call with confidence, including the genuinely hard overlap, and one factor almost no one considers: that an animated bar and an animated line tell different stories even from the same numbers.
In this article
- The one rule that settles most cases
- Bar chart vs line chart at a glance
- When to use a bar chart
- When to use a line chart
- The hard case: data that changes over time
- How animation changes the choice
- Common mistakes with both
- A three-question checklist
The one rule that settles most cases
The choice comes down to what your horizontal axis represents.
If the x-axis holds distinct categories with no inherent order between them, regions, products, plans, teams, channels, you want a bar chart. Each bar is a separate thing, and the eye compares their lengths instantly. There is no implied connection between "North America" and "Europe," so nothing should visually connect them.
If the x-axis holds a continuous progression, usually time, months, quarters, days, you want a line chart. The line's whole purpose is to connect points and show the path between them, so the reader follows the trend, the slope, the direction of travel.
That single distinction, categories versus continuum, decides the large majority of charts correctly. The rest of this guide is about the cases where it is not quite enough.
Bar chart vs line chart at a glance
| Bar chart | Line chart | |
|---|---|---|
| Best for | Comparing values across categories | Showing change over a continuous range |
| X-axis holds | Distinct categories (regions, plans) | A continuum, usually time |
| The eye reads | Length, which value is biggest | Slope and direction, the trend |
| Ideal data points | A handful (roughly 3 to 12) | Many (often 10+) |
| Implies connection between points? | No, each bar stands alone | Yes, points are part of one path |
| Baseline | Must start at zero | Can be context-dependent |
| Typical question it answers | "Which is biggest?" | "Where is this heading?" |
Keep this table in mind as a gut check, but the reasoning behind it is what lets you handle the edge cases, so here is each side in full.
When to use a bar chart
Reach for a bar chart whenever the core question is "which one is bigger?" Bars encode value as length, and length is the single easiest visual quantity for the human eye to compare accurately, far easier than angle (the problem with pie charts) or area. When you want a comparison to be unambiguous, bars win.
Use a bar chart when:
- You are comparing distinct categories. Revenue by region, signups by channel, headcount by department. Sort the bars from largest to smallest so the ranking is obvious, unless the categories have a natural order you need to preserve.
- You are ranking things. "Our top five products by revenue" is a bar chart every time. If the labels are long, switch to a horizontal bar chart so the text has room to breathe.
- You are comparing a few values across two dimensions. Revenue by region and by quarter, for instance, works as a grouped bar chart when you want to compare within each group, or a stacked bar chart when the parts add up to a meaningful whole.
The one rule you cannot break with bars: start the axis at zero. Because bars communicate through length, truncating the baseline distorts the comparison directly, a bar twice as tall genuinely looks like twice the value, so if the axis starts at a non-zero number, every comparison the chart makes is a lie. Lines have more latitude here; bars have none.
When to use a line chart
Reach for a line chart whenever the core question is "where is this heading?" The connecting line is not decoration, it is the message: it tells the reader these points belong to a single, continuous story, and it makes the trend, the rise, the dip, the plateau, the thing you actually see.
Use a line chart when:
- You are showing change over time. Revenue by month, active users by week, page load time by day. Time is the classic continuous axis, and the line lets the reader trace the trajectory rather than compare isolated values.
- You have a lot of data points. Twelve months, fifty-two weeks, a year of daily numbers, this would be an unreadable fence of bars but a clean, legible line. The more points you have, the more a line outperforms bars.
- You are comparing a few trends against each other. Two or three lines on the same axis, this year versus last, product A versus B, let the reader compare trajectories. Keep it to a handful; five tangled lines is a dashboard, not a chart.
- You want a single growing metric to feel substantial. Fill the space under the line and you have an area chart, which gives weight to a metric like MRR or cumulative users.
Lines have more freedom with the baseline than bars, because they communicate through slope rather than length, so a non-zero axis can be legitimate when you are emphasising the shape of a change. But that freedom is also where lines get abused, which we will come back to.
The hard case: data that changes over time
This is the overlap that sends people looking for an answer, because time-series data can legitimately go either way, and the rule "time means line" is not quite right.
The deciding factor is how many periods you are plotting.
With many periods, ten, twelve, fifty, a line chart is almost always correct. The points are dense enough that the connecting line reveals a trend, and that many bars would be cramped and hard to read. A year of monthly revenue, weekly signups over a quarter: line.
With just a few periods, four quarters, three years, a bar chart is often the better choice even though the axis is time. Four bars growing from zero make each discrete value easy to read and compare directly, and the "biggest one last" effect lands the headline cleanly. This is exactly why a revenue growth chart by quarter usually works better as bars than as a line, something we dug into in the animated revenue growth chart guide.
So the refined rule for time-series is: many points lean line, few points lean bar. If you can count the periods on one hand, try bars first. If you would run out of fingers, use a line.
One more nuance. Even with time on the axis, ask whether your real message is the trend or the individual values. "Our growth is accelerating" is a trend, that is a line. "We did $42K, then $58K, then $76K, then $119K" is a set of values you want read individually, that is bars. Same data, different message, different chart.
How animation changes the choice
Here is the factor almost no comparison of these two charts mentions: once a chart moves, the bar and the line tell genuinely different stories, and that can tip a close call.
An animated bar chart grows each bar up from zero, often in sequence. That sequencing creates suspense and emphasis, the final, tallest bar arriving last reads as a punchline. Animation makes bars feel like a reveal, which is perfect when you have a small number of values and want the biggest one to land as the headline.
An animated line chart draws itself on from left to right, and the effect is momentum. The eye follows the line as it is created, and a steepening slope reads as acceleration in a way a finished static line never quite captures. Animation makes a line feel like a journey unfolding, which is ideal when the story is direction and pace rather than individual figures.
So in a genuine toss-up, let the story you want decide: if you want a single value to feel like a climax, animated bars; if you want to convey building momentum, an animated line. This matters most when the chart is going somewhere attention is scarce, a deck, an investor update, or a social feed, which is the entire subject of our guide on making an animated chart for LinkedIn.
Common mistakes with both
Most bad charts are not the wrong type so much as the right type used carelessly. The recurring errors:
- A line chart for categories. Connecting "North America, Europe, APAC" with a line implies a progression between them that does not exist. It is the single most common misuse, and it looks wrong to everyone even if they cannot name why. Categories are bars.
- A bar chart with thirty bars. Too many bars become an illegible fence. If you have that many points and they are time-ordered, you wanted a line. If they are categories, show the top handful and bucket the rest into "Other."
- A truncated bar baseline. Starting bars above zero to exaggerate a difference. Because bars are read by length, this distorts the comparison outright. Bars start at zero, always.
- Too many lines. Five or six series crossing over each other is unreadable. Limit it to two or three, or split into small multiples.
- A dual axis to force two series together. Putting revenue and conversion rate on one chart with two different y-axes lets you imply a relationship that may not exist, and the reader cannot tell which line belongs to which scale. Usually two charts are more honest than one clever one.
- Ignoring the message. Picking the chart by habit instead of by what you are trying to say. Always start from the sentence you want the reader to walk away with.
A three-question checklist
When you are not sure, answer these in order and the choice resolves itself:
- Is the x-axis categories or a continuum? Categories with no natural order, bars. A continuum like time, lean line, then ask question two.
- How many data points? Many (roughly ten or more), line. Just a few, bars are often clearer even for time.
- Is my message the trend or the individual values? Trend and direction, line. Comparing specific values, bars.
Get those three answers and you will pick correctly far more often than not. And when it is genuinely a tie, the answer is rarely "agonise harder", it is to build both in a few seconds and look at them side by side. You can do exactly that with the bar chart generator and the line chart generator: paste your numbers once, try each, and keep whichever tells your story more clearly.