Product Return Rate Config Estimation

Overview

The product_return_rate_config_estimation contains Return Rate (RR) estimations for the assortment that you offer. The estimations are based on various factors influencing return rates, such as country, brand, article type, price, sustainability, PDP content, delivery times, items sold per brand and per order, historical return rates, and benchmark return rates and are updated weekly. The estimations provided can be used as a replacement of actual Return Rate to steer the assortment (especially for new articles whose actual Return Rate is missing or for articles with few sales whose actual Return Rate fluctuates heavily).

The dataset only includes articles with at least one transaction in the past year. Return Rate estimations for new articles are published as follows: if the first transaction occurs before Wednesday, the estimation is published the following Monday; if it occurs after Wednesday, it's published the Monday after that (i.e., two weeks later).

Property	Description
Data granularity	Config SKU, daily, country level, merchant_id
History available	No historical data is available because the dataset is a snapshot
Update frequency	Weekly
Data retention	7 days
Primary keys	predicted_dt*, config_sku, country, merchant_id
SLOs	Weekly on a Monday at 02:00am UTC
Notification system for delays	See Versioning and Deprecation Policy
Support	Support available via partner-care@zalando.de; No 24/7 support available

*To join product_return_rate_config_estimation with other shares, you can use predicted_dt from this share and dt from other shares.

Table Reference

The table can be accessed using the following data reference from delta sharing client applications:

product_return_rate_config_estimation_share.direct_data_sharing.product_return_rate_config_estimation

Schema

Column name	Format	Description
created_at	timestamp (yyyy-MM-dd HH:mm:ss)	The datetime when the data was written to the table (reference timezone: UTC).
updated_at	timestamp (yyyy-MM-dd HH:mm:ss)	The datetime when the data was updated in the table (reference timezone: UTC).
predicted_dt	date (yyyy-MM-dd)	The calendar date the estimated return rate applies to. You can join this column with dt available in other shares.
model_snapshot_dt	date (yyyy-MM-dd)	The calendar date the model was executed. By design, all rows share the same value.
account_id	string	The Zalando Partner zDirect Account ID.
config_sku	string	Zalando's reference code for a SKU (Stock Keeping Unit) at the article and colour level. A config SKU might have multiple sizes (referred to as Simple SKUs).
merchant_sku	string	Partner's reference code for a SKU (Stock Keeping Unit) at the article and colour level.
legal_entity_name	string	The name of the specific legal entity associated with the merchant.
merchant_id	string	The merchant identifier. Also referred to as the Business Partner Identifier (BPID). For more information, see Merchant Identifier in the Developer Guide.
merchant_name	string	The name of the merchant.
country	string	The country code associated with the snapshot or event in ISO 3166-1 alpha-2 format (e.g. DE for Germany).
brand_name	string	The name of the brand.
estimated_return_rate	double	The Return Rate estimation for the given Config SKU, country, merchant_id and predicted_dt. By design, the estimated_return_rate value remains constant across all the predicted_dt for a specific Config SKU, country and merchant_id combination.

Changelog (non-breaking changes)

No non-breaking changelog entries have been documented yet.

Contact Support