gizmobench

Radar Chart Maker

Paste one row per axis, with a first row naming the series, and get a radar chart with the scale printed on it, a legend that separates series by colour and by dash, and a data table beside the drawing. A radar chart is only readable if you know what one ring is worth, which is the thing most makers leave out: they fit each axis to its own numbers without saying so, which quietly makes a wide shape on one axis mean something different from a wide shape on the next. Here the scale is part of the output. A common scale draws one set of rings and prints their values, and a per-axis scale labels every axis with its own range and says on the page, in the chart and to a screen reader that the axes cannot be compared with each other. Nothing is clamped: a value outside a scale you typed is named with its axis and its series instead of being pinned to the outer ring, and a blank cell opens a gap in the outline rather than dropping the line to zero. Up to 24 axes and 8 series, an SVG download, no sign-up, and nothing leaves your browser.

Chartscale stated on the chart
Axes
0
Series
0
Mode

Paste a row for each axis on the left to draw a radar chart.

  • Common scale2 series
    scale common 0 to 100, 4 axes
  • Per-axis scalesame data
    Speed 0 to 100, Power 0 to 100, and 2 more
  • Missing valuePower blank for B
    scale common 0 to 100, 4 axes, 1 blank

One row per axis: the label, then a number for each series, separated by tabs, commas or two or more spaces, so a block copied from a spreadsheet pastes straight in. The first row names the series. Leave a cell blank, or write a dash, for a value you do not have: the outline opens there instead of dropping to zero. Up to 24 axes and 8 series are drawn, the scale boxes stay empty to fit the scale to the data, and a value outside a scale you typed is reported rather than pinned to the edge. Your data and settings are kept in this browser only.

Accuracy. The geometry is exact: each point sits at its value on its own axis under the scale shown, and the scale is printed on the chart. A per-axis scale makes axes not comparable to each other, so the page says so whenever it is on, and a data table carries the same numbers.

Common questions

How do I enter the data?
One row per axis: the label, then one number for each series. Cells are separated by tabs, by commas, or by two or more spaces, so a block copied from a spreadsheet, a CSV row and a typed column all read the same, and an axis label of two words such as Customer support survives. A single space between the cells is enough when every axis label is one word, and a row whose cells do not add up is named rather than half read. The first row names the series when none of its cells after the first is a number; if it is all numbers it is read as data and the series are named Series 1, Series 2 and so on, so nothing is lost to a header that was not there. One thing to watch: if your series names are years such as 2024 and 2025, put a letter in front of them, otherwise that row reads as data.
Should I use a common scale or a per-axis scale?
Use a common scale when the axes are measured in the same units or on the same rating, for example five scores out of ten. One set of rings then covers every axis, the ring values are printed, and the shape means what it looks like it means. Use a per-axis scale when the axes are different quantities, for example price in pounds beside a rating out of five, since one shared ring set would flatten the small numbers into the centre. The cost is real, and the page states it: each axis is then measured against its own range, the same distance from the centre means a different amount on each axis, and the shape can no longer be compared across axes. That sentence appears under the chart, in the downloaded file and in the text a screen reader hears whenever the per-axis scale is on.
Where does the scale come from if I leave the boxes empty?
It is read from the data and rounded out to a round number. Values that are all positive are measured from zero, because a radar chart drawn from a floating baseline exaggerates small differences, and the top is the next round number at or above the largest value: 90 becomes 100, 3.2 becomes 4. Negative values round the other way to a round minimum. Type a number into either box to set that end yourself and leave the other empty to keep it fitted to the data. Under a per-axis scale each axis gets its own range by the same rule, and both the axis label and the copied table carry it.
What happens to a value that is outside the scale I typed?
It is reported, not drawn. The chart names the axis, the series and the number, says which end of your scale it broke, and tells you to raise that end or clear the box to fit the scale to the data. Nothing is pinned to the outer ring or squashed into the centre, because a point sitting on the edge of a chart that cannot hold it is a silent lie about the number. A minimum that is not below the maximum is refused the same way rather than quietly swapped, and a value exactly on either end of the scale is accepted: it sits at the centre or on the outer ring.
What if one value is missing?
Leave the cell blank, or write a dash, n/a or a question mark. That axis gets no point for that series and the outline opens there, so a gap in the shape reads as an absence instead of as a zero, which is what a line dropped to the centre would claim. The count of missing values is said in the description of the chart, the data table beside the drawing shows the word blank in that cell, and the copied table leaves the cell empty so a spreadsheet reads it as missing too. If every cell is blank the page says so rather than drawing an empty wheel.
How many axes and series can it draw?
Up to 24 axes and 8 series, and at least 3 axes, since two axes make a line rather than a shape. Past either limit the page says how many you gave and what the limit is instead of drawing part of the data and staying quiet about the rest. Each series has its own colour and its own dash pattern, so two series stay apart in a black and white print and for a reader who does not separate hues, and the legend shows both.
Can I download the chart or copy the numbers?
Both. Download SVG saves the drawing you are looking at as a vector file: it scales to any size without blurring and opens in Inkscape, Illustrator or Figma, where every ring, label and outline is a separate object you can restyle. It carries the dark background, the legend and the scale line you see on the page, and it contains no script. Copy table puts the grid on the clipboard as tab separated text that pastes into a spreadsheet as columns, with two extra columns for the scale each axis used when the per-axis scale is on.
Is my data sent anywhere?
No. The parsing, the geometry and the drawing all happen in your browser, and the SVG file is made there too and handed to the ordinary download. Nothing is uploaded and there is no account. Your grid, the scale boxes, the mode and the title are kept in this browser's local storage so the chart is still there when you come back, and Start over at the top of the page forgets them.
Is a radar chart the same as a spider chart?
Yes. Radar chart, spider chart, web chart and star chart all name the same picture: one spoke per axis, a point on each spoke at that axis value, and the points joined into a shape. The names carry no difference in method. Whatever you call it, the honest version states what one ring is worth, which is what this page prints on the chart itself.

The geometry is exact: each point sits at its value on its own axis under the scale shown, and the scale is printed on the chart. A per-axis scale makes axes not comparable to each other, so the page says so whenever it is on, and a data table carries the same numbers.