How toSchema turns supplier catalogs into reviewable product data

A catalog job begins with a PDF, XLS, XLSX, or CSV supplier file. Source-page and quote evidence remains available to the reviewer while product rows and field values are prepared for the catalog contract.

Documents that need OCR are processed before their text and layout can be interpreted as product information.

The catalog contract keeps product identity in named fields and supplier-specific specifications in typed attributes. Source page references, quotes, validation failures, and missing-evidence states stay with the relevant record.

A missing or unsupported source value can remain absent. It should not be replaced by a value inferred from another product or nearby table cell.

Reviewers can correct, approve, or reject catalog records. A saved mapping profile then selects and renames approved fields for CSV or JSON download.

The downloaded file is a handoff for downstream systems. Direct PIM or ERP writeback is not part of the current catalog workflow.

Readable source material matters: blurred scans, low-contrast tables, merged cells, cross-page headings, and footnotes can require manual review.

Source evidence helps a reviewer inspect a result, but it does not turn an extraction into a compliance, payment, or approval decision.