Free Private Chart Makers

Paste CSV to build line, bar, area, pie, waterfall, combo, scatter, histogram, funnel, and Sankey charts entirely in your browser. Every page explains what the format shows and where it misleads.

Your spreadsheet never leaves the page

Every chart here is built in the browser. CSV is parsed by PapaParse and rendered by Apache ECharts, both running as JavaScript in the page, and the PNG export is produced from the canvas locally. There is no request to a server at any point, which you can confirm by opening the network tab while you paste.

That matters because of what people chart. A revenue bridge contains figures that are not public. A funnel contains conversion rates a competitor would like to know. A scatter plot of experiment results is unpublished research. Pasting any of that into a hosted chart builder makes it a row in someone else's database, and the usual reassurance is a privacy policy rather than an architecture.

Pasting directly from Excel or Sheets works because the clipboard carries tab-separated values, so there is no file export step and the data never touches disk on the way in.

Choosing the format is most of the work

Rendering a chart is the easy part. The decisions that determine whether it communicates anything are the choice of format, the axis, and how much you put on it, and those are what each page here is written about.

Some of it is settled by perception research rather than taste. People judge position on a common scale most accurately, then length from a shared baseline, then angle, then area — which is why bars are read more reliably than pie slices, and why a 23 percent slice and a 26 percent slice are indistinguishable while the equivalent bars are not. Some of it is about what an encoding claims: connecting two points with a line asserts that intermediate values exist, which is true for months and false for product names.

And some of it is about the parameters that quietly change the conclusion. Bin width can make a histogram look unimodal or bimodal from the same data. A truncated bar axis turns a 4 percent spread into an apparent threefold difference. A second axis on a combo chart lets whoever sets the ranges decide where the series appear to cross. Each page states the ones that apply to it rather than leaving them as defaults nobody examined.

Guides by question

Four longer articles cover the decisions that span several chart types: what a connecting line claims and what stacking costs, why bars are recommended so consistently and where a pie chart is still right, how bin width and overplotting change what a distribution looks like, and what a waterfall or Sankey shows that a pair of bars cannot.

Change over time

Charting change over time — Connecting two points asserts that the values between them exist. That assertion is the whole difference between a line chart and a bar chart.

Comparison & composition

Comparison and composition — People judge length against a common baseline far better than they judge angle or area. Most charting advice follows from that one fact.

Distribution & correlation

Distribution and correlation — Summary statistics hide things that a plot of every observation does not. That is the entire argument for these two chart types.

Flow & change

Flow and change — Two bars tell you a number changed. These formats tell you what changed it, which is usually the question.