Word Frequency Counter

Most free word frequency tools botch the count before you even read it. They split don't into don and a stray t, count cat, separately from cat, and chop mother-in-law into three pieces. So the number you came for scatters across near-duplicate rows, and the word you actually overuse hides in fragments that never add up. This one counts the way a reader would. A contraction is one word, a hyphenated compound is one word, and the comma stuck to the end of a word is not part of the word. Paste your text and you get a clean frequency table: every distinct word, how many times it shows up, and the share of the total it takes, ranked from most-used down.

That table answers a different question than a plain word counter does. A counter gives you a total, how many words, how many characters. This gives you the breakdown underneath it, which words you used and how hard you leaned on each. If you have ever finished a draft suspecting you keep reaching for the same handful of verbs, or you want to check whether a page mentions its topic enough to read as being about it, the per-word count is what you need. And it all runs in your browser. The text never leaves your machine, which matters when the draft is a client's document or something you would rather not paste into a random website.

1. Paste your text into the box. The table updates as you type, there is no analyze button to press.
2. Set the options. Ignore case folds The, the, and THE into one row. Ignore common words drops filler like the, and, of so the words that carry meaning rise to the top. Min length hides words shorter than the number you pick. Top N trims the table to the N most frequent (set it to 0 to see every word).
3. Read the table. Each row gives the rank, the word, its raw count, and a bar showing its percentage share of the words that survived your filters.
4. Watch the three stats up top. Total words is how many got counted after filtering, Unique words is how many distinct ones, and Shown is how many rows are on screen once Top N trims the list.
5. Export it. Copy table drops a tab-separated version on your clipboard that pastes straight into a spreadsheet. Copy CSV and Download CSV hand you a comma-separated file for Excel, Sheets, or a script.

This is the use worth the tool. Every writer has crutch words, the ones you reach for without noticing until an editor circles them. A frequency table finds them for free, and it finds them in ten seconds, an editor takes years to do the same thing by feel.

Paste a finished draft, turn on Ignore case, turn on Ignore common words, and set the minimum length to about 4 so the small connectors drop out, then read down the top twenty. Those are the words a reader feels you leaning on, whether or not they could name them. Some belong there. If you wrote about budgets, budget near the top is just the topic doing its job. But when a limp adverb or a generic adjective like very, really, or important is riding high, that is your edit list.

One honest thing about Ignore common words. The stop-word list is crude, around 125 of the usual suspects, and it strips not right along with the filler. In a line like not good that not is doing real work, so the toggle is great for reading the shape of a text and worth turning off when every word matters.

Then vary the ones you overuse, but not mechanically. Repetition can be deliberate, even good rhythm. When the same modifier turns up eight times in six hundred words and none of them are doing real work, swapping half of them tightens the whole piece. Once the table says actually appears eleven times, Find and Replace is the fast way to hunt each one down. It works on more than essays too. A resume where every bullet opens with Managed, a product page with a fifth premium creeping in.

The counting rule is exactly where cheap frequency scripts fall apart. A word here is a run of letters and digits, plus any apostrophe or hyphen sitting between letters. That keeps don't whole instead of breaking it into don and a lone t, and state-of-the-art as one entry instead of four. Quotes and brackets are boundaries and get stripped, and so is any comma or period that is not sitting between two digits, which is why cat, cat, and cat! all collapse into one word, count 3. Numbers count too, so 2024 is a word like any other and $1,500 stays on one row instead of splitting into a 1 and a 500.

This matters because punctuation splitting is the single most common way a naive counter lies to you. It files design and design, as two separate entries, eight of one and five of the other, so a real count of forty ends up scattered across rows that each look minor. You would never spot the word you overuse. Here the count stays whole, so the top of the list reflects what you actually wrote.

Case folding does the same for capitals. Start three sentences with "Marketing" and use "marketing" a dozen more times mid-sentence, and to a computer those are two rows, not one. Flip on Ignore case and they merge into the honest total. Turn it off only when capitals carry meaning, a proper noun kept apart from the common word, or code like getUser versus getuser.

If you write for search, keyword density comes up constantly, how often a target phrase shows up relative to the total. A frequency table is the clean way to see it, the raw count and the percentage side by side.

The old advice was to hit a magic ratio, with one to two percent the number everyone repeated. That advice is dated and was never that precise. Modern search engines do not count a phrase and reward a fixed ratio, they read for meaning and related terms, and stuffing a keyword to hit a target tends to get you demoted instead of ranked. The table is genuinely good for a sanity check. Did you mention your main topic enough that a reader skimming the page can tell what it is about? Twelve hundred words on cold brew coffee where the exact phrase turns up twice reads like a page that wandered off its subject. It catches the opposite just as fast, the word that crept in nine times because you ran out of synonyms.