Scatter Plot Maker starts with one strict numeric x,y pair per non-empty line; generate the plot, verify the reported counts and axis ranges, then download the same SVG shown in the preview.

Each valid pair becomes one blue circle: x controls its horizontal position, while y controls its vertical position. This makes the tool suitable for paired observations, small experiments, calibration samples, classroom demonstrations, and quick visual checks of a relationship.

If your numbers begin in two spreadsheet columns, this guide to prepare two spreadsheet columns as x,y pairs can help with the conversion before pasting. Keep the original dataset because the chart is a visualization rather than a precision-preserving analysis file.

how do i get started when i need to make scatter plot when using scatter plot maker
How to Get Started With a Scatter Plot Maker?

Prepare the x and y measurements

Before opening the maker, copy the measurements into a simple working list. Each non-empty line must contain exactly two comma-separated finite numbers in x,y order. For example, 2.5,-1 is one accepted point. Signed values, decimals, zero, and scientific notation are accepted, and negative zero is normalized to zero.

  • Do not include a header row, because nonnumeric text is invalid.
  • Remove units such as centimeters or kilograms; unit text is not accepted.
  • Avoid empty components, extra commas, and other separators.
  • Do not use hexadecimal values, Infinity, or NaN; each is invalid.
  • Keep every value at or below an absolute magnitude of 1e12; larger magnitudes are invalid.

Blank lines are ignored, so they do not become points. The pasted input is limited to 50,000 UTF-16 code units. Enter the optional title in its own field rather than adding it to the data rows.

Input condition How it is handled
First 200 non-empty rows Each row is checked as a strict x,y pair and contributes to either the valid-point or invalid-row count.
Rows after the first 200 non-empty rows Each additional row is counted as an over-limit skip and is not parsed.
Blank rows Ignored; they are not plotted or included in the three summary counts.
Pasted input above 50,000 UTF-16 code units The parser rejects the input.
No valid point remains after parsing Generation reports an error and clears any older chart.

Create the first scatter plot

  1. Open Scatter Plot Maker and enter an optional title. The title is included in the generated chart and downloadable SVG.
  2. Paste the prepared rows, placing exactly one numeric x,y pair on each non-empty line. Confirm that the first value is the horizontal measurement and the second is the vertical measurement.
  3. Generate the plot. If the data contains no valid point, the maker reports an error rather than displaying a misleading chart.
  4. Read the summary before evaluating the preview. It separates valid points, invalid rows, and over-limit rows, including when a count is zero.
  5. Inspect the displayed axis ranges and several representative points. Confirm that the valid-point count matches the number of observations you intended to plot.
  6. Download the standalone SVG when the preview and counts look correct. The file is generated from the same SVG string used for the on-page preview.

If you change the title or data after generating the chart, generate it again before downloading. Editing either input clears the previous generated SVG and disables the old download.

Read the three row totals

The three summary numbers describe different problems. A valid point was parsed successfully and plotted. An invalid row was checked but rejected, so it is skipped rather than converted to a zero point or allowed to move an axis. An over-limit row was deliberately not parsed at all, so it is not mixed into the invalid count.

This separation makes it possible to see whether the chart represents the entire pasted dataset. Nothing is silently truncated. Review the source for malformed rows when the invalid count is above zero. When the over-limit count is above zero, decide whether using the first 200 non-empty rows is appropriate for your intended chart.

Check the displayed axis ranges

Both axes use independent linear scales. The smallest valid x value maps to the left plot boundary, the largest x value maps to the right, the smallest y value maps to the bottom, and the largest y value maps to the top. Position depends on numeric distance rather than row order, so unequal numeric gaps occupy unequal amounts of space. Five reference lines and numeric labels expose each displayed range.

Do not assume that an axis starts at zero. The maker uses the data’s minimum and maximum unless a constant value requires a special domain, so read the labels before judging the apparent height, spread, or distance of a cluster.

Axis condition Displayed behavior
x values vary The x axis uses a linear scale from its smallest to largest valid value.
y values vary The y axis independently uses its own linear minimum and maximum.
All x values are identical The x domain expands symmetrically around the constant by 10% of its absolute value or one unit, whichever is larger.
All y values are identical The y domain receives the same independent symmetric expansion.
Only one valid point exists The point is centered horizontally and vertically.

Axis labels use practical significant digits or scientific notation for very large and small values. Coordinates in the SVG are rounded to three decimal places, so the visualization is not suitable for preserving exact numerical values.

Inspect points before export

Check representative points in the preview, including points near the displayed minimum and maximum and one from the middle of the dataset. Each generated circle contains a text title with its formatted x and y values for SVG-aware viewers. Comparing those values with the source is a practical way to catch an incorrect row order or an unexpected axis interpretation.

Overlapping points can hide one another, particularly when the same pair appears more than once. A visible circle count may therefore be smaller than the valid-point count. For dense datasets, inspect the underlying rows rather than assuming every point is visible.

Download the matching SVG

The downloaded file is a standalone 800 by 500 SVG with a white background, title, grid, numeric axis labels, and point markers. The preview and download are not reconstructed separately: both use the same percent-encoded SVG data URI derived from one generated string. Because SVG is vector-based, its lines, text, and circles remain sharp when resized.

User-entered titles are XML-escaped before entering the SVG text or accessible label. Script-like text remains visible text rather than creating SVG elements, and code points forbidden by XML are replaced. For background on the file format, MDN’s SVG tutorial explains how its vector elements and XML structure work. Parsing, layout, SVG creation, preview preparation, and download preparation all happen locally; no points or title are uploaded.

Know what the visualization can establish

The chart does not calculate correlation, regression, a trend line, confidence intervals, error bars, density, clusters, or statistical significance. An upward or downward visual pattern is therefore an observation for further review, not a statistical conclusion. Retain the source rows for uncertainty analysis, reproducibility, and exact values.

Use a dedicated plotting package when you need dense-data handling, categorical colors, accessible labeling, logarithmic scales, publication styling, or formal statistical analysis. For smaller paired datasets and immediate visual inspection, the verified input, count, preview, and SVG workflow provides a straightforward starting point.

Related reading: Bar Chart Maker Safe: Local Data and SVG Export.