Word Cloud Generator

Turn survey answers, reviews or keyword lists into a word cloud sized by frequency, with a frequency table and SVG or PNG download.

Last updatedHow we build & check our tools
Results will appear here once you enter your values.

How the Word Cloud Is Built

The generator runs four steps, all in your browser. First it splits the text on spaces and dashes and trims punctuation from the ends of each word, so shipping, and shipping count as one word while e-commerce and don't stay whole. Second it drops words shorter than your minimum length and, if you leave the option on, common function words such as the, and, of and with.

  • Counting: every remaining word is counted, sorted by frequency (ties alphabetically) and cut to your maximum, 60 by default.
  • Sizing: the most frequent word gets 72 px and the least frequent 14 px. Square-root scaling keeps mid-frequency words readable; linear scaling makes the top words dominate.
  • Placement: words are placed largest first, starting at the centre and moving outward along an Archimedean spiral until a spot is found where the word's box doesn't touch any other word.

Placement is deterministic. The same text and settings always give the same picture, and Shuffle layout changes the starting angle of each spiral to try a different arrangement.

Where Word Clouds Help Marketers

A word cloud is a fast first look at a pile of unstructured text. It won't replace proper coding of responses, but it shows in seconds which words customers reach for.

  • Open-ended survey answers: paste 200 answers to “What almost stopped you from buying?” and words like pricing or shipping stand out.
  • Reviews and support tickets: compare a cloud of 5-star reviews with one of 1- and 2-star reviews to see which themes drive each.
  • Message research: the exact words customers use make strong headline and ad-copy candidates.
  • Presentations: an SVG export stays sharp at any size in slides and reports.

The frequency table under the cloud gives the exact counts, which are what you should quote in a report. The cloud's visual size is only an approximation of those numbers.

Common Mistakes to Avoid

  • Reading size as precise data: long words take up more space than short words with the same count. Use the table for comparisons.
  • Leaving brand or question words in: if every answer repeats your product name or the survey question, those words swamp the cloud. Delete them from the text first.
  • Ignoring negation: “not easy” is counted as easy, because the cloud counts single words. Read a sample of the source text before drawing conclusions.
  • Splitting one idea across spellings: price, prices and pricing are counted separately. Standardise them in the text if they mean the same thing.

If some words show “didn't fit” in the table, the canvas ran out of room. Lower the maximum words or shuffle the layout.

Tips for a Clearer Cloud

  • Keep it to 40–80 words: beyond that the smallest words become unreadable.
  • Use square-root scaling for survey data: it shows the second tier of themes instead of letting one word take over.
  • Choose a high-contrast palette: all three palettes meet at least 4.5:1 contrast against the white background, so words stay readable when printed.
  • Export SVG for print and PNG for chat or email: the PNG is rendered at twice the on-screen size (1600 × 1000 px).

Nothing you paste is uploaded or stored. Close the tab and the text is gone.

Frequently Asked Questions

Common questions about the Word Cloud Generator

Size is based on how many times the word appears after common words are removed. The most frequent word is drawn at 72 px and the least frequent at 14 px, with square-root or linear scaling in between.

Sources & References

Tag cloud

Background on word and tag clouds, sizing words by frequency, and their limits as a data display.

Stop word

Why common function words are filtered before counting word frequencies.

Archimedean spiral

The r = bθ spiral used to search outward from the centre for a free position for each word.

CanvasRenderingContext2D: measureText() method

How word widths are measured in the browser for collision-free placement.