Word Counter with Reading Time

Word count is the unit almost every writing brief is denominated in, which is odd, because it is a measure of length and briefs are usually about depth. It survives because it is easy to check and roughly correlates with effort - and because editors, clients, and content management systems all speak it.

This counter reports eight figures from whatever you paste: words, characters with and without spaces, sentences, paragraphs, and an estimated reading time, plus a standing reminder of the 60- and 160-character limits that meta titles and descriptions live under.

Each figure comes from a specific rule about where text divides, and each rule disagrees with human judgement somewhere. This page sets out where, so you can tell a genuine result from an artefact of the splitting.

For search, length is a symptom rather than a cause. Longer pages tend to rank better on competitive queries because covering a subject properly usually takes more words, not because word count is rewarded - which is why padding a 700-word answer to 2,000 words reliably makes the page worse. The useful question is whether every section earns its place, and the counter cannot answer it.

Where the number does have real force is as a constraint. A 300-word product description, a 90-word app store blurb, a 1,200-word column: these are container sizes, and writing to fit them is a legitimate craft problem. The counter is at its most useful when you have a container and need to know how far over you are.

Reading time is the figure to be most careful with. It is computed at 200 words per minute, a reasonable average for prose and a poor fit for anything else - dense technical writing reads far slower, a scannable listicle far faster, and neither is captured by dividing words by a constant. Treat the minute figure as a rough courtesy label, not a measurement.

Sentences are found by cutting the text at every run of full stops, question marks, and exclamation marks, then discarding empty fragments. That is fast and it over-counts in three predictable places: abbreviations like "e.g." or "Ltd.", decimal numbers such as 3.5, and ellipses. A paragraph containing two prices and an abbreviation can report five sentences where a reader sees three.

It also under-counts. A sentence that ends with a colon before a list, or a line in a bulleted list with no terminal punctuation, is not a boundary the splitter can see - so a page built mostly from bullets reports very few sentences regardless of how many statements it makes.

Paragraphs are counted by splitting on runs of two or more newlines, which means the figure depends entirely on how you separated them. Text pasted from a source that uses single newlines between paragraphs reports one paragraph, however long it is. If the paragraph count looks wrong, that is almost always the reason, and the fix is to check the source formatting rather than the tool.

Words: the text is trimmed and split on any run of whitespace, and the resulting segments are counted. An empty or whitespace-only input returns zero. A hyphenated compound is one word, a spaced em dash is one word, and a standalone numeral is one word.

Characters: the with-spaces figure is the raw string length in UTF-16 code units; the without-spaces figure removes every whitespace character first. Sentences: the text is split on runs of terminal punctuation and non-empty pieces are counted. Paragraphs: the text is split on runs of two or more newlines and non-empty pieces are counted.

Reading time is the word count divided by 200, rounded up, with a floor of one minute - so any non-empty text reports at least "1 min". The two meta-limit tiles in the second row are static reference values, not measurements of your text; they are there as a reminder of the 60-character title and 160-character description conventions, and the counter does not compare your input against them.