What a word count actually measures

A word count is a convention, not a property hiding inside the letters. TextPilot follows Unicode word boundaries, the same family of rules a browser uses when it splits text into words. In ordinary English that usually matches a careful human count. The contraction in "I'm almost done" stays one word. The hyphen in "well-known" splits the compound into two words, because the hyphen is a boundary in that ruleset.

What the character rows mean

The character counter leads with grapheme clusters. That is the unit a reader sees as one mark. The letter in "café" is one character whether the file stores it as a single code point or as "e" plus a combining accent. A family emoji joined by a zero-width joiner is one character as well.

Two smaller units sit beside that headline number. Code points are the Unicode scalar values. UTF-16 code units are what older JavaScript fields often measure, and one emoji can occupy two of them. When a CMS disagrees with TextPilot, compare those rows before you rewrite a sentence to chase the wrong unit.

Sentences and paragraphs

Sentences start from Unicode sentence boundaries. Those rules are good at punctuation and imperfect with abbreviations, so TextPilot joins a short list of titles such as "Dr." and "Mr." back onto the following words. The list is small on purpose. It will not understand every abbreviation in every language. If a count looks wrong, open the sentence list and read the split.

Paragraphs are simpler. A blank line separates them. A single line break does not. The line count is available when you care about wrapping rather than structure.

Reading time is an assumption

Reading time divides the word count by a pace you choose. The default is 200 words per minute. That is a round figure for silent reading of general prose, not a measurement of your reader. A tutorial full of code, or a reader working in a second language, needs more time. The calculator shows 160 and 230 so you can see how easily the label moves.

Other writing systems

Chinese and Japanese text may be grouped into words such as 你好 rather than counted one character at a time. That is the segmenter doing its job, and it can surprise people who expected a character count. Use the character counter when the assignment is really about characters. Arabic text is counted with the same word boundaries, so short phrases stay intact.

Keep the draft local

None of these numbers require an account. The draft stays in the browser unless you explicitly save a project. Saving is the action that uploads that text, and the save control asks you to confirm it.

If a course or a client names a counting rule, match that rule instead of arguing with the tool. Some style sheets treat a hyphenated compound as one word. Some form fields count UTF-16 units. TextPilot shows its rules next to the numbers so you can choose the row that fits the brief.