Word Frequency Counter

Each row shows a rank, the word, how many times it appears, and its share of all the words counted. Paste The cat sat on the mat. The dog sat on the cat. and you get:

The sentence has 12 words, so 4 uses of the is 4 ÷ 12 = 33.3%. A few rules shape the list:

• Case is ignored: The, the and THE are one word.
• Words with the same count keep the order in which they first appear, which is why cat comes before sat.
• Only the top 25 words are shown. Percentages are still worked out against every word in the text, so the shown rows can add up to well under 100%.
• Common words such as the, and, of and to are not filtered out. They nearly always fill the top of the list, so look further down for the words that tell you something.

A word here is a run of the letters a to z, the digits 0 to 9, and the straight apostrophe '. Everything else acts as a separator. That gives some results that differ from a normal word count:

• Hyphens split words: well-known counts as well and known.
• Numbers count: 2026 appears as a word of its own.
• Email addresses and web addresses break into pieces: example.com adds one example and one com.
• Straight apostrophes stay inside the word, so don't and it's are single words. But a word inside straight single quotes, 'hello', is counted separately from hello, because the quote marks become part of it.

For a plain total of words, the Word Counter gives a different number, because it splits only on spaces.

Because only a to z are treated as letters, anything outside basic English letters gets split or dropped:

• Accented letters break words. café is counted as caf, résumé as r and sum, and naïve as na and ve.
• Other scripts are ignored. Words in Cyrillic, Greek, Arabic, Chinese or Hindi don't appear in the list at all, and they don't count toward the percentages.
• Curly apostrophes split words. Text from a word processor often uses ’, so don’t becomes don and t, separate from any straight-apostrophe don't in the same text.

For mixed sources, a quick fix is to replace curly apostrophes with straight ones before pasting. For languages with accents, treat the results as rough only.

Paste a draft, skip past the small function words, and look at the first content words in the list. If one appears far more than its neighbours, such as really, very or the same verb, that is where to start editing. For a page aimed at a search phrase, check that the main terms appear naturally, without one word swamping the rest.