Point the Content Refresher at a page you already have. It reads the live page, extracts the factual claims in it, researches what currently ranks for that topic, and rewrites the page to compete — checking its claims against what it extracted, so your facts survive the rewrite.
This is the tool for pages that used to rank and no longer do.
When to use it
Use it to refresh an old blog post that has slipped in rankings, update a page whose information has gone stale, or lift a thin article to match what now ranks on page one — without losing the facts the original got right. Reach for it when:
- A page has slipped and you don't know why.
- A page scores badly in the audit but the underlying content is sound.
- A page is factually right but structurally weak: no headings, no answer near the top, nothing an AI search engine can lift.
Use article generation → instead when the page doesn't exist yet, or when the existing page isn't worth keeping.
Refresh a page
To refresh a page:
- Open Content Refresher and paste the page's address. The target keyword is read from the page's title and H1; change it if it is wrong.
- Enter the URL. A target keyword helps but isn't required — the tool derives one when the page is already mapped.
- Leave
Light rewriteoff for the full rewrite: research, the fact check and before-and-after scores.Light rewriteis the older mode with none of those, and it still rewrites the page from scratch. - Run it, then compare the before and after scores.
Read the fact check
You get the rewritten article plus the content score. The score uses the same scoring engine as everywhere else, so the before and after numbers are directly comparable.
Facts extracted from the original are checked against the rewrite. Read that list — it's the part worth your attention, because a rewrite that improves the score while dropping a real claim isn't an improvement.
Limits and usage
- One run per page, and each run uses a usage credit.
- The page must be publicly reachable. Anything behind a login or a bot block can't be read.
- Pages that render entirely in JavaScript read as thin, because that's what the fetcher receives.
- It doesn't publish. Review, then publish through your integration →.
If something isn't working
"Could not read the page." Open the URL in a private browser window. If it needs a login or shows a bot check, the refresher can't see it either.
The rewrite lost something important. Paste the missing section back before publishing. The fact check lists any price or figure it could not find in the rewrite, but it is a safety net, not a guarantee.
The score barely moved. Usually the page is already structurally fine and the problem is elsewhere: intent mismatch, thin internal linking, or a topic your site has no authority on. The audit's Links category is the next thing to check.