Free toolRuns on our server; one fetch per checkProtocol v1.0Updated

Does your product page tell one story?

A shopping agent reads your price from more than one place. Paste a product page address and see what each place says, side by side, with the exact strings. Where two places disagree, you see both.

One product page per check. Free, no sign up. Results in under 20 seconds.

Last reading
None published. The reading is the one you run; results are not collected into a figure.
Protocol
Fact Split 1.0 (lib/factsplit.ts): three public surfaces, six fields, raw strings kept. A disagreement is declared only when normalized values differ; a field on fewer than two surfaces gets no verdict.
Verified
27 of 27 unit tests passed at build . How this is verified.
Failure state
A page that cannot be fetched shows the fetch error. No partial comparison is shown.
Receipt
/api/fact-split?url= returns the same comparison as JSON.
Method
Every normalization, limit and rule

Three public surfaces, six fields, no score

  1. 01

    Fetch once

    We read the page's robots.txt, then fetch the one page you gave us without running JavaScript, the way many AI crawlers do. Shopify product paths also get their public product JSON.

  2. 02

    Read each surface

    Structured data, visible text and the Shopify JSON each give us price, currency, availability, SKU, GTIN and variant id, exactly as written.

  3. 03

    Compare, carefully

    "$1,299.00" and "1299" are the same price. A disagreement is shown only when values differ after formatting is ignored, and a field stated on fewer than two surfaces gets no verdict. Read the method.

Light on the store

One page and at most one small JSON file per check, each limited to a few seconds. Our fetcher follows robots.txt and identifies itself. Results are kept in memory for ten minutes. What we store. Need a fix? OptiVis Marketing does this work for stores.