Does my data get uploaded when I make a chart?

No. All seven charts are drawn in your browser from the numbers you type or paste. Nothing is sent to a server, nothing is logged and nothing is stored, so internal figures never leave your machine. You can confirm it by disconnecting from the network once a page has loaded — the charts keep generating.

How do I save a chart as an image?

Hover over any finished chart and a small toolbar appears in the top-right corner. The camera icon downloads a PNG. Six of the seven render a 1000 by 600 frame at 2x scale, so the file is roughly 2000 by 1200 pixels — sharp enough for a slide or a printed report. The dot plot is the exception: its height grows with the number of rows, so a thirty-category chart exports taller rather than more cramped.

How many categories is too many for one chart?

It depends on the chart, because each one fails differently. A pie chart breaks around five or six slices, once wedges get too thin to label and similar sizes become impossible to rank. Vertical bars crowd past about twelve; horizontal bars stay readable to roughly twenty-five. A dot plot handles thirty or forty rows comfortably. Past that, group the tail into an 'Other' row or split the data across two charts.

Can I put two series on the same chart?

Not on these tools — each draws one series at a time. The workaround is small multiples: generate one chart per series with the same options and the same axis range, then place the PNGs side by side. Above about four or five series that layout is easier to read than a single chart with every line or bar stacked into it. For a before-and-after comparison, plotting the change itself is usually clearer than plotting both states.

I have a single column of numbers and no category labels. Which tool takes that?

The histogram maker or the box plot maker. Both read one list of values and describe how those values are spread out, so neither asks for labels — there are no categories to name. Every other tool here needs a second input: names paired with values for the bar, pie and dot plot makers, or an X list paired with a Y list for the line and scatter makers. If a tool is asking you for something you do not have, that is usually the signal you have opened the wrong one.

Chart Makers

Seven chart makers with one workflow: paste your values, adjust a few controls, download a PNG. There is no sign-up and no upload — every chart here is drawn in your browser from the numbers you type, so internal figures stay internal. Each tool's own page carries a worked example. The hard part is rarely the tool; it is choosing the right chart, which is what follows.

Bar Chart Maker

Create a bar chart from labels and values to compare categories.

Box Plot Maker

Create a box plot from your data and view the five-number summary.

Dot Plot Maker

Create a Cleveland dot plot from labels and values for ranked comparisons.

Histogram Maker

Create a histogram from your data to visualize frequency distributions.

Line Chart Maker

Create a line chart from X and Y data to visualize trends over time.

Pie Chart Maker

Create a pie or donut chart from labels and values to show proportions.

Scatter Plot Maker

Create a scatter plot from X and Y data to visualize relationships.

Which Chart Should You Use?

Chart	Your data	The question it answers
Bar chart	One value per named category	Which category is biggest?
Dot plot	One value per named category, many rows or long labels	How do these rank, when the values sit close together?
Pie chart	Parts of one mutually exclusive whole, five or six of them	How does a single total divide up?
Line chart	Paired X and Y along an ordered, evenly spaced axis	How did this quantity move?
Scatter plot	Two measurements taken on each of the same subjects	Do these two things move together?
Histogram	One column of continuous measurements	What shape is this distribution?
Box plot	One column of measurements, ten or more	Where is the centre, the spread and the extremes?

Start from the data you have, not the picture you want.

One value per named category — revenue by region, seconds by landing page — is the commonest shape, and three charts compete for it. A bar chart is the default and the one every audience already reads. A pie chart is only defensible when those categories are parts of one whole, every unit belongs to exactly one of them, and the total is a number someone cares about. A "select all that apply" question fails the second test; two years of sales fail the first, being two wholes rather than two parts. A dot plot takes over as the list grows. Twenty vertical bars are a wall of ink that forces long labels to be rotated; the bar chart maker's horizontal mode solves exactly that, and is the right answer when magnitude is the message. What it cannot solve is the axis. Because a dot encodes value as position rather than length, its axis is free to skip the empty space below your data — so eight values between 148 and 182 fill the plot, where eight zero-based bars would each stretch past four-fifths of the width and leave the real differences squeezed into what is left.

A column of raw measurements — 200 page-load times, every order value from last week — is a distribution, not a set of categories, and needs the histogram maker or the box plot maker. The histogram shows the shape: how many peaks, which way the tail runs, whether anything sits out on its own. The box plot compresses that shape into five numbers. On one dataset that is a loss. The payoff arrives when several groups have to share an axis, since each takes up almost no width — though this tool draws one dataset at a time, so comparing groups means generating each in turn and reading its five-number summary.

Confusing those two families is the commonest charting mistake, and one question separates them: would reordering the rows destroy anything? Support tickets by channel can sit in any sequence and the chart still says what it said — those are categories, and the tool is a bar chart. Salary bands cannot move, because each band's neighbours are fixed by the arithmetic that produced them — those are bins, and the tool is a histogram.

The remaining two charts need paired data. Use the scatter plot maker when you have two measurements on the same subjects and want to know whether they move together, and the line chart maker when the X axis is time or another ordered sequence. A line between two points claims the quantity travelled continuously from one to the other, so joining unordered categories draws a slope that means nothing.

Reading Your Data First

Four questions settle the choice before you open a tool.

How many variables are you plotting? One measured variable is a distribution; one measure across named categories is a comparison; two measured variables is a relationship. Those three words pick your row in the table above.

Is each variable continuous or categorical? Continuous means the values in between exist: durations, prices, temperatures. Categorical means it names a group. Continuous data needs binning or a number-line axis; categorical data needs a label.

Do the categories have an order of their own? Months, age bands, sizes S/M/L and a scale from "strongly disagree" to "strongly agree" all carry meaning in their sequence, and sorting them by value destroys it. Everything else should be sorted largest to smallest, so the message appears without the reader working for it. Alphabetical order is almost never right — it arranges your data by the spelling of the labels.

How many categories, and how far apart are the values? A short list of well-separated values suits bars; a long list, or values packed into a narrow band far above zero, suits a dot plot.

One last check before you paste: exactly one value per category. A label appearing twice gives two marks competing to represent the same thing, and no tool will notice — aggregate first. Unequal box lengths are caught, and reported as two counts rather than guessed at.