Word Counter

What It Does

Paste or type text on the left. Stats update live on the right.

• Headline number — big “X,XXX words · Y,YYY characters” at the top, the same way most counter sites lead
• Full stats panel — words, characters (with/without spaces), sentences, paragraphs, reading time, speaking time, reading level (Flesch-Kincaid grade), average word length, average sentence length, unique words, longest word
• Top-10 most-used words — collapsible list with counts. Spot repetition before your editor does
• Word goal — slider from 50 to 5000, live progress bar that turns green when you hit it
• Case converter — UPPER, lower, Title, Sentence, aLtErNaTiNg
• Native spell check — right-click any underlined word for suggestions. Works offline, in your locale, free

How the numbers are calculated

Metric	How
Words	Split text on whitespace runs (/\s+/), count non-empty
Characters	text.length including spaces and newlines
Characters (no spaces)	text.replace(/\s/g, '').length
Sentences	Count terminal-punctuation groups: ., !, ?, …
Paragraphs	Split on blank-line runs (/\n\s*\n/), count non-empty
Reading time	words / 265 minutes (Brysbaert 2019 silent-reading meta-analysis)
Speaking time	words / 150 minutes (average conversational speech rate)
Reading level	Flesch-Kincaid grade: 0.39 × (words/sentences) + 11.8 × (syllables/words) - 15.59
Syllables	Vowel-group counting with silent-e / -ed / -es stripping; minimum 1 per word
Unique words	Lowercase, strip non-letter/digit chars, count distinct
Longest word	Iterate cleaned words, keep the max
Top 10	Frequency map → sort desc → slice 10

The Flesch-Kincaid label maps grade ranges to plain English: <6 = Elementary, 6-8 = Middle school, 9-12 = High school, 13-15 = College, 16+ = Graduate.

Why no server-side grammar checker

Other counter sites bolt a paid “Check” button onto their UI that pings a remote LanguageTool-style backend. We don’t.

• Your text never leaves your browser.
• The browser’s native spell check (the spellcheck="true" attribute on the textarea) already catches misspellings and offers corrections via right-click. It uses your OS dictionary, your locale, and is offline.
• A real grammar checker (subject-verb agreement, passive voice, comma splices) needs ~5-10 MB of WASM or a server round-trip. Both would break the 100%-client-side promise. If you want that level, LanguageTool runs locally as a Java server and there are good browser extensions; that’s the right home for it, not a side feature in a counter.

When To Use It

• Drafting a blog post, essay, LinkedIn update, or newsletter and need a live word count
• Checking the reading level of marketing copy or technical docs
• Spotting overused words (“just”, “really”, “very”) before your editor flags them
• Quick character count for tweet/title/meta-description length checks
• Estimating reading time for a piece of long-form content

What’s NOT Included

• Grammar check — see above, deliberate scope choice
• Persistent state — reload = blank slate, by design (every other Gekro app in this batch follows the same rule)
• Thesaurus / synonyms — would need a 1-2 MB word database; out of scope
• Multi-document tabs — one textarea, one counter; if you need multiple drafts use separate browser tabs
• History / undo button — the <textarea> has native Ctrl+Z / Ctrl+Y, no JS button needed

Related Tools

• Markdown Visualizer - Markdown editor with live HTML preview and side word count
• Prompt Token Counter - Token estimation and API cost for LLM prompts
• Tokenizer Visualizer - See exactly how an LLM splits your text into tokens
• Text Diff - Compare two versions of any text
• Rich Text to Markdown - Convert pasted formatted text to clean Markdown