How the count is made
The text is lowercased and split into words: a word here is a run of letters or digits, optionally followed by an apostrophe and more letters, so don't and team's each count as one word. Punctuation, line breaks and symbols are separators. Because matching is case-insensitive, Content, content and CONTENT are counted together as one entry - that is what you want for keyword analysis, though it does merge a proper noun with an ordinary word that happens to share its spelling.
Two filters shape the table. Minimum word length drops very short tokens; 3 is a good default because it removes most noise without hiding real words. Remove stop words excludes about 100 extremely common English function words - articles, pronouns, prepositions, auxiliary verbs like is, are, was, and conjunctions. Without that filter the top of any English text is always the, of, and, to, which tells you nothing. Two percentages are shown: share of the counted words (after filtering) and share of all words in the text, which is the figure SEO tools normally call keyword density.
Reading the results
In the sample text, content appears 6 times out of roughly 200 words, about 3% of everything and a clear signal of what the passage is about. Promotion, plan, piece and spend follow. That shape - one dominant term, a handful of supporting terms, then a long tail of words used once - is normal English, and it is a consequence of Zipf's law, which says a word's frequency is roughly inversely proportional to its rank: the second most common word appears about half as often as the first, the third about a third as often, and so on.
For SEO, a primary keyword at roughly 1-2% density in text that reads naturally is the usual guideline; nothing above 4-5% is worth chasing, and Google has said for years that there is no magic ratio - stuffing hurts more than it helps. The more useful use of this table is editorial: it exposes crutch words. If just, really, very, actually or leverage appear near the top, cut them. Seeing the same noun five times in a short paragraph is a sign to vary the wording or split the idea.
Other uses and limits
Teachers use frequency lists to check vocabulary range in student essays; a high ratio of words used only once suggests varied vocabulary, while a low one suggests repetition. Researchers use them for a quick read on survey free-text answers or support tickets - paste a few hundred responses and the top terms show what people actually complain about. Writers use them before submitting, since editors notice repeated words faster than authors do.
Limits to keep in mind: this counts exact word forms, so publish, publishes and published are three separate entries - no stemming or lemmatization is applied. Multi-word phrases are not detected, so a two-word keyword has to be read off its parts. Only English stop words are built in. And the whole analysis runs in your browser, so drafts, client documents and survey responses never leave your computer.