The reverse feed — pulling Amazon's data back out.

The reverse feed is the underrated workhorse of Amazon catalogue work. It is a flat-file Amazon generates from your live data on demand — the same column shape as the upload templates from Episode 10, but populated with the values Amazon currently has stored on each ASIN. If the upload template is "what you want Amazon to know", the reverse feed is "what Amazon actually knows right now". Every serious catalogue audit, every cross-marketplace migration, every translation handoff, and every diff between what you submitted and what Amazon stored starts with a reverse feed export.

Where to get it

Seller Central → Catalog → Inventory Reports → Category Listings Report. Vendor Central has an equivalent under Items → Reports → Catalog Listings. Request the report once; Amazon emails a download link or surfaces it in the Reports inbox within 5–30 minutes depending on catalogue size. The download is a UTF-8 TSV file structurally identical to the upload template for the corresponding category. SP-API integrations can request the same data programmatically through the GET_MERCHANT_LISTINGS_ALL_DATA and GET_FLAT_FILE_OPEN_LISTINGS_DATA report types, with category-specific extensions through GET_XML_BROWSE_TREE_DATA and the SP-API Catalog Items API.

The two flavours

• All Listings Report. The lighter version. Per-SKU offer-level data — ASIN, SKU, price, quantity, fulfilment channel, status, item-condition. Generated continuously by Amazon and available on demand without a fresh request. The right tool for offer-level diagnostics ("which SKUs are out of stock?", "which SKUs are below my reserve price?") and for repricer integration. Lacks attribute-level data — no title, no bullets, no images.
• Category Listings Report. The full reverse feed. Per-SKU attribute-level data for one specified category: title, bullets, description, images, keywords, dimensions, weight, materials, certifications, the variation theme and parent-child relationships, every category-specific attribute. The right tool for audits, migrations, translations, and any operation that touches more than offer-level fields. Has to be requested per category — a catalogue spanning five categories requires five separate report requests.

The workflows the reverse feed unlocks

• Catalogue audit. Pull the Category Listings Report, diff it against your source-of-truth content master (the spreadsheet or DAM where the brand canonical copy lives), and surface every divergence. Common findings: Style-Guide rewrites you didn't notice, contributor overrides from competing sellers, translations that drifted from the source language, A+ Content assignments that don't match the latest brand spec.
• Cross-account migration. Moving SKUs from one Seller account to another (acquisition, restructure, agency change). Pull the reverse feed on the source account, re-upload through the new account's flat-file pipeline. The shape compatibility makes this a one-day operation rather than a weeks-long re-keying project.
• Translation source-export. Pull the reverse feed in the source marketplace, send to translators with the column meaning documented, receive the translated TSV back, upload to the target marketplace. Far more reliable than re-keying titles from a Word document.
• Bulk edits. Edit the reverse feed in Excel, change the update_delete column to PartialUpdate (Episode 10), re-upload. The reverse feed is the natural source for any bulk edit because it pre-populates every required column with the current valid value, leaving you to change only the cells that need changing.
• Suppression diagnostics. When a listing is suppressed, the reverse feed shows the stored values — frequently revealing the Style Guide rewrite or the missing attribute that triggered the suppression in the first place.
• Variation-family verification. Confirm the parent-child relationships are intact and the children's variation-theme attributes are populated correctly. A common audit finding: children whose variation-theme attribute drifted to blank after a partial-update upload, breaking the picker on the live PDP without producing a visible error.
• Compliance-attribute backfill. After a regulatory change (new EU labelling requirement, new energy-label rule), pull the reverse feed, identify SKUs missing the now-required attribute, populate the column, re-upload as PartialUpdate.

The mental model: reverse feed as Amazon's source-of-truth, not yours

The most important thing to understand about the reverse feed is that it shows you what Amazon thinks is true about your catalogue, not what you think is true. The two diverge constantly:

• Brand Registry edits override Seller-Central submissions — the reverse feed shows the Brand Registry value.
• Style Guide rewrites silently change the stored value — the reverse feed shows the rewritten value.
• SP-API feeds fire periodically and overwrite manual edits — the reverse feed shows the most recent feed value.
• Catalogue-team interventions modify attributes without notification — the reverse feed is the first place these surface.
• Variation-family contributors with higher trust scores reassert different values — the reverse feed shows the winning value.

The diff between your content master and the reverse feed is the most honest catalogue audit you can run. It tells you not what should be on Amazon, but what actually is.

Operational rhythm — how often to pull

• Weekly for active catalogues with multiple contributors (multiple Seller accounts, agencies, automated feeds). The diff against last week's reverse feed surfaces unauthorised edits within seven days.
• Monthly for stable catalogues with a single content owner. Sufficient to catch Style Guide updates and catalogue-team interventions.
• Before every major upload. Always export current state before pushing a flat-file update — both as a rollback artefact and as the basis for the PartialUpdate diff.
• After every catalogue policy change announcement. Amazon updates Style Guides and attribute schemas with little advance notice. A reverse feed pulled the week of an announced change surfaces every SKU the change affects.

Gotchas in the reverse feed format

• Excel mangles GTIN columns. 13-digit EANs get stored in scientific notation or lose leading zeros when Excel auto-types the column. Open the TSV in a text editor first to verify the raw values, or import into Excel with the GTIN column explicitly set to Text.
• Multi-value attributes split across columns. Bullet points populate bullet_point1, bullet_point2, etc. — five separate columns rather than one. Same pattern for images, keywords, and several attribute fields.
• HTML entities in description. The description column can contain HTML entities (&, ") that need decoding before display. Round-tripping these without re-encoding can introduce bugs on re-upload.
• UTF-8 BOM in some downloads. Some reverse-feed exports include a UTF-8 BOM that breaks naive TSV parsers. Strip the BOM before processing.
• Locale-specific decimal separators. EU-marketplace exports may use comma as decimal separator in price fields; US uses period. Normalise before bulk-edit operations.

Seller vs Vendor — reverse feed access differs

Sellers access the reverse feed through the standard Category Listings Report described above, with self-service request and download. Vendors have access through Vendor Central → Items → Reports, with a similar but more limited dataset — Vendor reverse feeds typically exclude offer-level data that Vendors don't control (price, inventory, fulfilment) and include richer cost and packaging fields that Vendors do control. Hybrid brands running both account types maintain two separate reverse-feed pipelines and need to reconcile across them periodically.

What to take into the next episode

The reverse feed tells you what is stored. The next episode covers Amazon's own quality scorecard for what is stored — the Listing Quality Dashboard, which scores every ASIN against attribute completeness, content presence, and several other factors and gives you a prioritised list of fixes ranked by likely impact.