Duplicate Title & Meta Finder
Paste your titles, one per line, optionally prefixed with the URL. The tool groups exact duplicates, flags near-duplicates that share 80 percent of their words or 85 percent of their characters, and tells you how many titles are actually unique.
Paste columns straight from Screaming Frog or a spreadsheet. Nothing is uploaded.
How duplicates are detected
Each line is normalized first: whitespace trimmed and collapsed, everything lowercased. Titles that are identical after normalization are grouped as exact duplicates and listed first. The remaining titles are then compared pairwise: a pair is flagged as a near-duplicate when its word sets overlap by 80 percent or more, or when the character-level edit distance puts the two titles at 85 percent similarity or higher. That second check catches singular and plural swaps, reordered words, and one-word edits that a pure word-overlap test would miss.
What to do with the results
Exact duplicates usually mean one of three things: a template bug stamping the same title across a section, true duplicate pages that should be canonicalized or merged, or pagination missing page numbers. Near-duplicates are subtler. If two pages genuinely serve different intents, rewrite the titles to lead with what distinguishes them. If they serve the same intent, the fix is consolidation, not a synonym swap, because two pages chasing one query split their ranking signals.
Working at scale
Pairwise comparison grows quickly, so the near-duplicate pass caps at 500 lines and shows a note when your list is larger. Exact-duplicate grouping has no practical limit. For very large sites, run sections through separately, or sort your crawl export by title first so duplicates land next to each other and chunks stay meaningful.
Frequently asked questions
Why do duplicate titles hurt?
Two ways. Google treats boilerplate and repeated titles as low-information and rewrites them with its own generated text, so you lose control of your SERP presentation. And pages with identical titles tend to target the same query, which means they cannibalize each other: the ranking signals split across them and neither page ranks as well as one consolidated page would.
How different is different enough?
Lead with the distinguishing term, not the shared brand boilerplate. 'Blue Widget Pricing | Acme' and 'Blue Widget Reviews | Acme' differ by one early word and that is fine, because the differentiator comes first and names what makes the page unique. Titles that only differ in their last few words after a long shared prefix are the ones Google collapses.
Where do I get all my titles from?
Three common sources: a Screaming Frog crawl exported as Internal HTML with the Title 1 column, the pages report in Google Search Console, or a quick crawl of your XML sitemap. Copy the URL and title columns out of the export and paste them here, tab-separated columns straight from a spreadsheet work as-is.
What threshold counts as a near-duplicate?
Two titles are flagged when their word sets overlap by 80 percent or more (Jaccard similarity), or their character-level similarity is 85 percent or more (normalized Levenshtein distance). Those thresholds catch reworded and lightly edited variants while leaving genuinely distinct titles alone.