How to Check Word Count in Google Docs (and What to Do When It Doesn't Match)

Where to Find Word Count in Google Docs

Google Docs shows word count in two places:

• Menu path: Tools → Word count (or the keyboard shortcut Ctrl+Shift+C on Windows/ChromeOS, Cmd+Shift+C on Mac)
• Live counter: In the Word count dialog, checking "Display word count while typing" pins a small counter to the bottom-left of the document that updates continuously as you type

The dialog itself shows four numbers: page count, word count, character count, and character count excluding spaces. On mobile, the same information is available through the three-dot menu in the Docs app, under Word count.

Selection vs. Whole Document

The most common source of confusion is not realizing that Google Docs counts differently depending on whether text is selected. With nothing selected, the word count dialog shows the count for the entire document. With a portion of text selected, the same dialog shows the count for just that selection — and this is easy to miss, because the dialog looks nearly identical in both cases, with only a small distinction in the labeling.

This becomes a practical problem when someone means to check the word count of a specific section — an introduction, a single chapter — but forgets whether text was selected beforehand, and ends up comparing the wrong number against a target. Before checking, it's worth clicking once in the document (to deselect anything) if the intent is to see the whole document's count, or deliberately selecting the exact range needed if the intent is a partial count.

Why the Count Doesn't Match What You Expect

Headers, footers, and footnotes are excluded from the main count. Google Docs' word count only counts the body text by default. A document with substantial footnotes — common in academic writing — will show a lower word count than the document actually contains when footnote content is factored in. There's no built-in option to include footnotes in the count; if that content needs to be counted, it has to be copied out and counted separately.

Hyperlinked text counts as words, but the URL doesn't. A hyperlink displayed as "click here" counts as two words; the underlying URL isn't counted at all, regardless of its length. This is generally the expected behavior but occasionally surprises people who assumed a long link was inflating their count.

Numbers and standalone symbols count as words. "2024" counts as one word. A standalone dash or ampersand used as a separator, depending on how it's spaced, may or may not count as a word — Google Docs' word-boundary logic treats whitespace-separated tokens as words regardless of whether they're alphabetic.

Comments and suggested edits are excluded. Text in the comments panel or in suggestions that haven't been accepted yet doesn't count toward the document's word count. This matters for collaborative documents mid-review — the count reflects the accepted, current body text, not any pending suggested additions.

Tables are counted, but formatting can make it feel inconsistent. Text inside table cells is included in the word count, but because tables visually separate content, it's easy for a writer to mentally undercount how much text a table actually contains relative to the surrounding paragraphs.

None of these behaviors are bugs — they're consistent, defined rules. The confusion comes from not knowing the rule, not from the count being wrong. Once the rule is known, the number stops being surprising.

Why Different Tools Show Different Numbers for the Same Text

Pasting the same document into Microsoft Word, Google Docs, and a third-party word counter can produce three slightly different word counts, even though nothing about the text has changed. This happens because "word" isn't a perfectly standardized unit — different tools make different decisions about edge cases:

• Whether a hyphenated term ("well-known") counts as one word or two
• Whether an em dash with no surrounding spaces splits or joins the words on either side of it
• Whether numbers, currency symbols, and standalone punctuation count as words
• Whether headers, footnotes, and captions are included

For most documents, these edge cases produce small differences — a handful of words out of several thousand, rarely enough to matter. For documents with heavy use of hyphenation, numbers, or footnotes, the difference can be more noticeable, and it's worth using the same tool consistently rather than switching between Google Docs, Word, and a third-party counter and expecting identical numbers.

Setting a Live Word Count Target

For writing against a specific target — an assignment with a minimum, an article with a maximum — Google Docs supports setting a target and displaying progress toward it. In the Word count dialog, checking "Set a word count target" adds a small progress indicator alongside the live counter, useful for essays and assignments where hitting a specific range matters, covered in more detail in our post on how to accurately count words for a school assignment.

When a Quick, Separate Count Is Faster

For a fast check that doesn't require opening the full document — verifying the length of a paragraph before pasting it somewhere else, checking a draft that exists outside Google Docs, or comparing several candidate versions of a short piece of text side by side — pasting into a standalone word counter is often quicker than navigating the Tools menu each time. It also avoids any ambiguity about selection state, since the entire pasted text is what gets counted, with no partial-selection confusion possible.

This is particularly useful when the text in question isn't in Google Docs at all — an email draft, a social media post, a Slack message — where opening a full document editor just to check a count adds unnecessary steps.

---

Google Docs' word count is accurate and consistent — it just follows specific rules that aren't always obvious until they explain an unexpected number. Selection state, excluded content types, and the standard edge cases around hyphens and numbers account for nearly every case where the count seems wrong. It usually isn't; it's just counting something slightly different than what was expected.