How are URLs compared?

Both files are normalised before comparison: lowercased, query strings removed and trailing slashes stripped. Redirects that differ only in case, parameters or a trailing slash therefore count as matches, and mappings that intentionally differ only by query string cannot be validated with this tool.

What happens if a source URL appears twice in a file?

Each file is collapsed into a source to destination dictionary, so only the last row wins for a duplicated source URL. Deduplicate your mapping first if it contains alternative rules for the same source, or the earlier rows are silently ignored.

Which columns do I need, and what about redirect chains?

Any CSV or Excel with a source and a destination column in each file; you pick the columns in the UI. Chains are the main trap: if your spec lists final destinations but you map the crawl's first hop column, every chain shows as a mismatch. Map a final address column from your crawler where available.

How is the accuracy percentage calculated?

Matches divided by the total specified redirects (matches plus mismatches plus missing). Extra redirects found in the crawl but absent from your spec are reported in their own tab but do not affect the score.

Redirect Validator: free Technical SEO tool

#about

Offline dataset comparison (no live HTTP requests).

Compares crawled redirects export (Screaming Frog) against redirect mapping specification.

Four-category output: Matches (correct), Mismatches (wrong destination), Missing (not crawled), Extra (unspecified).

Column mapping UI for source/destination fields.

Streamlit app

Platform: Browser-based (no installation required)
Input: Crawled redirects file (CSV/Excel from Screaming Frog)Redirect mapping specification file (CSV/Excel)
Output: Excel with a Full Report sheet (matches listed within it) plus Mismatches (wrong destination), Missing (specified but not found), and Extra (found but not specified) sheets when non-empty, plus accuracy score.
Category: Technical SEO

Use cases

Post-migration redirect QA
Redirect implementation verification
Agency handoff validation
Technical SEO audits
Pre-launch redirect testing

#features

Dataset-based comparison (not live checking)
URL normalisation: lowercase, query strings and trailing slashes stripped
Four categories: Match/Mismatch/Missing/Extra
Flexible column mapping UI
CSV and Excel file support
Accuracy percentage calculation

#how-to-use

Upload your crawled redirects export
Upload your redirect mapping specification
Map source and destination columns from each file
Run validation comparison
Review categorised results in tabs
Download the full report as CSV, or Excel with Mismatches, Missing, and Extra sheets

#faq

How are URLs compared?: Both files are normalised before comparison: lowercased, query strings removed and trailing slashes stripped. Redirects that differ only in case, parameters or a trailing slash therefore count as matches, and mappings that intentionally differ only by query string cannot be validated with this tool.
What happens if a source URL appears twice in a file?: Each file is collapsed into a source to destination dictionary, so only the last row wins for a duplicated source URL. Deduplicate your mapping first if it contains alternative rules for the same source, or the earlier rows are silently ignored.
Which columns do I need, and what about redirect chains?: Any CSV or Excel with a source and a destination column in each file; you pick the columns in the UI. Chains are the main trap: if your spec lists final destinations but you map the crawl's first hop column, every chain shows as a mismatch. Map a final address column from your crawler where available.
How is the accuracy percentage calculated?: Matches divided by the total specified redirects (matches plus mismatches plus missing). Extra redirects found in the crawl but absent from your spec are reported in their own tab but do not affect the score.

/contact

Want me to run this for you? I run this tool as a managed service, or build something custom around your data. You get the insights without touching the code.

Send me a message →