Word Counter

How this word counter decides what counts as a word

The count you see is built the same way Microsoft Word and Google Docs build theirs: the text is split on whitespace, and every non-empty piece between two spaces counts as one word. That one rule explains almost every surprise people hit.

• “don't” is one word, there is no space inside it.
• “New York” is two words, because the space between them is a boundary.
• “state-of-the-art” is one word. The hyphens do not split it, so a phrase that reads as four ideas still counts as a single token.
• “hello, world” is two words. The comma is punctuation, not a separator, so it stays glued to the first word.

Numbers, dates and mixed strings follow the same rule. “2026-10-16” is one word, “3.5” is one word, and a lone em dash with spaces around it is not a word at all because the pieces on either side are empty. Paste a table straight out of a spreadsheet and the count will look inflated, every cell becomes a token.

If the number matters for a contract, a translation quote or a college submission, agree on the tool before you start. A freelancer billing by word and a client counting in Google Docs will usually land within a percent of each other, but footnotes, headers and hyphenated compounds are exactly where two counters disagree.

Characters with spaces vs without: which number do you need?

Two character totals are shown, and reaching for the wrong one is the most common mistake when you are filling in a form or trimming a headline.

• Characters (with spaces) is the raw length of everything you typed, including every space, tab and line break.
• Characters (no spaces) strips the whitespace out first. It is the figure that matters when a limit is about the visible text.

Meta descriptions, social captions, ad headlines, subject lines and SMS all count spaces, so use the “with spaces” number for those. A handful of translation and transcription contracts quote the no-spaces figure instead. When a platform says a meta description should be about 155 characters, it means characters with spaces, a space is a character like any other.

A quick worked example: the sentence “Free tools for everyday math.” is 29 characters with spaces and 24 without, five interior spaces make up the difference. So a caption sitting at 158 characters with spaces is three over a 155-character limit even when the no-spaces count looks comfortable. X adds its own twist: links are shortened and always count as 23 characters no matter how long the real URL is, and most emoji count as two. If you are writing for X specifically, trust its own counter over a generic one.

Reading time and speaking time: where 200 and 130 words per minute come from

Reading time here assumes 200 words per minute. That is deliberately middle-of-the-road: adults reading English silently for comprehension cluster around 200 to 250 wpm, faster when skimming and much slower through dense technical or legal text. Speaking time assumes 130 words per minute, a comfortable presentation pace, quick enough to sound confident, slow enough that an audience keeps up. Auctioneers and some podcast hosts run far faster; a nervous speaker at a podium often runs slower.

Both figures come from the same word count, so they scale linearly, double the words and you double the time.

So a 1,000-word article is about five minutes of reading and just under eight minutes read aloud. If you are scripting a five-minute talk, aim for roughly 650 words, 650 divided by 130 equals 5 minutes. Writers who hand in 1,000 words for a “five-minute” slot routinely run to nearly eight minutes once they add pauses, slides and audience laughter.

Worked example: what a 1,000-word article reports

Suppose you paste an article that is exactly 1,000 words. If the average word length is 5.1 characters, the tool reports about 5,100 characters before spaces and roughly 6,000 with them, because English prose runs about 15 to 20 percent whitespace. Split into sentences, a typical 1,000-word piece of web writing holds 55 to 70 sentences, so the average sentence lands at 15 to 18 words. That average is a readability signal on its own: once sentences pass about 25 words, readers start losing the thread.

Paragraph count depends entirely on your blank lines. A paragraph is any block separated by an empty line, so a solid wall of text is one paragraph no matter how long it is, while ten single-line items separated by blank lines count as ten. Most style guides suggest keeping web paragraphs to two to four sentences, and mobile readers prefer even shorter.

The unique-word count rounds out the picture. A 1,000-word article with 1,000 unique words is impossible, English repeats function words constantly, but a healthy ratio for plain prose is roughly 40 to 55 percent unique tokens. Drop below about a third and the draft is leaning on the same vocabulary, which usually reads as thin and is worth a rewrite.

Sentences and paragraphs: why the count sometimes looks off

The sentence counter looks for runs of text that end in a period, question mark or exclamation point. That is close enough for ordinary writing, but a few patterns inflate it:

• Abbreviations: “Dr. Smith arrived at 9 a.m.” can read as three sentences because of the periods in “Dr.” and “a.m.”
• Decimals and version numbers: “version 2.1 shipped” can look like a sentence break if a capital letter follows.
• Initials: “J. R. R. Tolkien” can register as three sentences.

None of these touch your word or character counts, which are exact, only the sentence figure, and occasionally the readability numbers, are affected. If precision matters, read the sentence count with that in mind.

Paragraphs are simpler and more literal: the tool splits on blank lines. A single line break inside a paragraph does not start a new one; you need an empty line between blocks. That matches how most content management systems and Markdown renderers define a paragraph, so the number usually lines up with what a reader sees on the page.

How long should your content be? Working targets by format

There is no universally correct length, but there are useful working targets for common formats in the US market. Treat these as starting points, not rules, the right length is the shortest one that answers the reader's question completely.

Two caveats. Longer is not automatically better for search, a padded 3,000-word page performs worse than a tight 1,200-word one that answers the query. And length should follow intent. Someone asking how many cups are in a gallon wants an instant answer; someone researching a mortgage wants depth. Match the format to the question, then use the counter to check you did not overshoot or underdeliver.

Unique words, repetition and keyword density

The tool also lists your ten most frequent words after removing common stop words, the, and, of, to and roughly ninety others, and ignoring anything shorter than three letters. What is left is the vocabulary that carries meaning, so you can see at a glance whether you have repeated yourself.

This doubles as a rough keyword-density check. If a target keyword appears 15 times in a 1,000-word article, that is a density of 1.5 percent, which is on the high side; readable copy usually sits around 0.5 to 1 percent. Density is not a ranking factor by itself, but a phrase that turns up in every paragraph reads as spam to humans, and humans are the audience you are writing for.

Average word length is a second, quieter signal. Everyday English averages about 4.7 to 5.2 characters per word. Push that average toward six and you are probably reaching for long Latinate words where short ones would do, “utilize” instead of “use”, “commence” instead of “start”. Neither the frequency list nor the average is a hard rule, but together they are two of the fastest ways to spot a draft that needs tightening before you publish.