Paste one absolute HTTP or HTTPS URL per line — up to 1,000 lines and 100,000 characters. The studio finds exact duplicates after normalization, then lets you test how the inventory changes under comparison rules you choose. It runs locally and never fetches the URLs.
Normalization removes fragments and default ports through the browser URL parser, so https://example.com/guide/#top and https://example.com/guide/ are the same inventory entry. It deliberately retains path case, query parameters and trailing slashes, because each can identify a different resource: https://example.com/Guide/ is a distinct URL from https://example.com/guide/ until a server proves otherwise, and ?page=2 is not page 1.
The comparison rules exist for the next step of the audit — asking “what if we treated these as the same?” — without silently merging them. A near-duplicate group is a review list, not a verdict: check what each URL actually serves before consolidating.
Two different URLs can serve identical content, while two similar-looking URLs can be distinct resources. This tool does not crawl, inspect robots directives, check status codes, or determine which URL a search engine selected as canonical. Pair it with a crawl or Search Console data for those questions — and with the canonical inventory guide for the decision method.
No. No URLs are fetched. It only parses the text you provide. Use a crawl or Search Console to investigate HTTP status and indexing.
Your sitemap is the fastest source: the sitemap URL extractor turns sitemap XML into a clean list and can send it straight here. Crawl exports and CMS route lists work too — one URL per line.
A URL with ?utm_source= is usually the same resource wearing a campaign label — but “usually” is doing real work in that sentence. Some sites serve different content by parameter. The studio counts them and lets you preview the collapse; the decision stays yours.