Identifiers
Applies toSurface-agnostic
EDITED tags every record with stable identifiers. They're what you use to follow one option's documents over time, and to join EDITED's data to your own catalog. The same identifiers appear whether you read the data over MCP, the Query API, or the Jobs API.
This page is about the IDs that come back in the data — the keys you link and join on. Turning a name into the slug or ID you put in a filter (resolving "Nike" to a brand slug, say) is a separate step — see Resolving values and How to query.
Option, SKU, and product
The identifiers make more sense once the three levels of granularity are clear:
- An option is a product at style + colour level (a specific colourway of a style).
- A SKU is an option at a specific size — one option has many SKUs.
- A product is the thing that groups an option's linked SKUs together.
In the apparel and beauty verticals the data is presented at the option level; homeware is available at both the option and SKU level.
The three stable IDs
| Identifier | What it identifies | Use it to… |
|---|---|---|
option_id | One option (a style + colour) | Link an option's documents across dates — it never changes |
product_hash | The product that connects an option's linked SKUs | Tie an option together with its SKUs |
retailer_product_id | The retailer's own product code (usually style level) | Join EDITED's data to a retailer's catalog or your internal data |
The key property is stability: option_id and retailer_product_id are
immutable over time, so either one is a reliable key for linking the same
option's documents from different dates. Each field is defined in full in the
data dictionary.
Resolving a retailer's product ID to EDITED's
If you hold a retailer's own product IDs and want EDITED's product_hash for
them, the Query API's retailer_id_to_product_hash endpoint does the mapping —
including for products EDITED hasn't seen yet, since it computes the hash from
the retailer and ID. It's the clean way to join a retailer's catalog to EDITED's
data. See the API reference.
Related
- Data dictionary — the full definition of each identifier field.
- Filters → Resolving values — turning names into filter slugs/IDs.
- How to query — the identifier-resolution pattern over MCP.