HTML to Text
Extract the visible text of pasted HTML or one public URL, with character, word and line counts. Scripts, styles and markup are dropped.
The words, without the markup
Word counts, readability checks and quotes all start from the same thing: the text a reader actually sees. This page extracts it from pasted markup or from one public URL and reports the character, word and line counts beside it.
How to use it
Paste the markup and press Extract the text, or enter a public https:// URL instead. Pasted markup is read as-is and nothing is fetched, so a page behind a login or a draft that is not deployed can still be read. Exactly one of the two inputs is required; when both are given the pasted markup wins.
What counts as visible
Body copy counts. Scripts, styles, tags, comments and head metadata do not, and entities are decoded, so & reads as &. A URL is read from its raw HTTP response without running JavaScript, and a non-HTML answer reports that instead of an empty page.
One owner, three readers
This extractor is the same code the SEO checker and the link checker read their visible text through. Paste a page here and audit it there, and the word count is the same number, because there is only one definition of visible text to drift.
A real capture
This is the recorded output for pasted markup carrying one heading and one sentence: 21 characters, 4 words, 1 line — Rates How rates work.
Privacy and safety
Pasted markup never leaves the request that carries it. A fetched URL goes through the same guarded path as the other audit tools: private addresses refused, redirects capped, timeouts set, the read capped, nothing written to disk or logged.
Paste markup or enter one public URL and this page returns the visible text — the body copy a reader actually sees — with the character, word and line counts beside it.
Scripts, styles, tags, comments and head metadata contribute nothing, and entities are decoded. Pasted markup is read as-is and nothing is fetched, so a page behind a login or a draft that is not deployed can still be read. Exactly one of the two inputs is required.
This is the same extractor the SEO checker and the link checker read their visible text through, so the word count here is the word count there: one definition, three readers.
Pasted markup or a fetched page
Pasting is the path for anything the server cannot reach: drafts, staging hosts, pages behind a login. The markup travels with the request and is parsed in memory; no fetch is made and no client is constructed.
A URL is fetched through the same guarded path as the audit tools and read from its raw response without running JavaScript. A non-HTML answer reports that instead of an empty page.
- Exactly one input required; pasted markup wins when both are given
- Input past the shared read cap is refused, not truncated
- Entities decoded, whitespace collapsed
What the counts mean
Characters count the extracted string, words split on whitespace, lines split on line breaks — the same three numbers the audit payloads carry, so a count quoted from one surface matches the other.
Nothing is stored. The text is returned in the response and never written to disk or logged.
How to use the two inputs
Paste the markup for anything the server cannot reach, drafts, staging hosts, login-walled pages, or enter one public URL for a live page. Exactly one input is required, and pasted markup wins when both are given.
Run the tool and read the visible text with its character, word and line counts, the same three numbers the audit tools quote. Scripts, styles, tags and head metadata contribute nothing; nothing is stored.
What to Expect
Best for
- Getting a word count from markup without rendering it
- Quoting a page's visible text for a readability check
- Reading a draft or login-walled page you can copy
Not ideal for
- Text that only exists after JavaScript runs
- Preserving layout, tables or formatting — output is plain text
- Pages on private network addresses, which are refused
What this tool keeps
- The body copy, in document order
- Decoded entities and collapsed whitespace
- Character, word and line counts with the text
What may need cleanup
- Navigation and footer copy, which is visible text too
- Line breaks, which follow the markup rather than the render
- An empty result on a script-rendered shell
Common errors
- Expecting table layout or headings markup in plain-text output
- Pasting a URL into the markup field instead of the URL field
- Assuming hidden text is excluded — display:none is still markup text
Example use cases
- Word count for an article draft
- Source text for a readability or keyword pass
- Quoting a page without its navigation chrome
Sample input
Paste a page's HTML, or enter its public URL.
Sample output
The visible text with 21 characters, 4 words and 1 line for a heading plus one sentence.
Who this is for
- Writers counting words in a draft
- Editors preparing copy for a readability check
- Anyone who needs a page's words without its markup
Frequently asked questions
What counts as visible text?
Body copy with scripts, styles, tags and comments dropped and entities decoded, plus character, word and line counts.
Can it read a page behind a login or an undeployed draft?
Yes. Pasted markup is read as-is and nothing is fetched, so any markup you can copy can be read.
Is this the same text the audit tools count?
Yes. The SEO checker and the link checker read their visible text through this same extractor, so the word count is one number everywhere.
Build custom tools for your store
Join the SDK program to create useful ecommerce utilities for your store.