Word & Character Counter

Count Words, Characters, Sentences & Paragraphs.

What Counts as a Word?

Two tools can count the same text and disagree, and neither is broken. There is no single definition of a word, so every counter makes choices — and the choices show up most on exactly the text people care about counting.

Most counters, including this one, split on whitespace, which treats hyphenated compounds and contractions as single words. That matches how people usually read, and it is why a document can come out a few percent shorter here than in a tool that splits more aggressively.

The practical rule is that if a specific count matters — a submission limit, a contractual deliverable — count it in the tool that will be used to judge it, and treat every other figure as an estimate.

Characters With and Without Spaces

The two figures differ by roughly the word count, since most words are followed by one space. Which one applies depends entirely on who is asking.

• Social platforms and SMS count characters including spaces, because every character occupies transmission space.
• Translation and copywriting are frequently priced per character excluding spaces, since spaces are not work.
• Database fields and form limits count everything, including spaces and line breaks.

The difference is not trivial. A 1,000-character piece with spaces is roughly 830 without, which changes an invoice noticeably.

Why Some Characters Count as More Than One

A character limit is often not counting what you see. Many systems count storage units rather than visible characters, and the two diverge outside plain English.

An emoji can occupy several units while appearing as one symbol, so a few emoji can exhaust a 20-character field that looks nearly empty. Accented letters and non-Latin scripts can also take more than one unit each. Some emoji are built from several joined components, which is why deleting one sometimes leaves a fragment behind.

If a field rejects text that appears to be within the limit, this is almost always why. Removing emoji is usually enough to bring it back under.

Reading Time Is an Assumption, Not a Measurement

Estimated reading time is word count divided by an assumed speed, and the assumption does most of the work. Typical figures sit somewhere around 200 to 250 words per minute for silent reading of ordinary prose, which is a wide enough band that the same article can honestly be labelled a five or a seven minute read.

The estimate also ignores everything that changes actual reading speed: technical vocabulary, code samples that get studied rather than read, images and tables that interrupt the flow, and whether the reader is skimming. Speaking is slower still — a presentation script usually runs closer to 130 to 150 words per minute, which is why a ten-minute talk needs far fewer words than a ten-minute read.

Treat the number as a rough signal of length rather than a promise about time.

Hitting a Limit Without Damaging the Writing

Cutting to a word count is a skill, and the order in which you cut matters.

• Remove qualifiers first — very, quite, really, somewhat. They almost always weaken the sentence they modify.
• Replace phrases with words: "in order to" becomes "to", "due to the fact that" becomes "because".
• Cut whole sentences before trimming words from every sentence. Removing one paragraph usually reads better than shaving ten percent off all of them.
• Leave the opening and closing alone until last. They carry disproportionate weight, and cutting them to save words is a poor trade.

To normalise capitalisation across a piece before submitting it, the text case converter handles that, and the random text generator produces text at an exact word count if you need to test a limit.

Counting Questions