OTTO Sessions
An ecommerce interaction benchmark for predicting what a shopper will click, add to cart, or order next.
What the resource contains
- About 12 million anonymized sessions
- Timestamped article interactions: clicks, carts and orders
- Approximately 220 million events involving 1.8 million article IDs
Potential uses
- Train next-item recommenders
- Model multi-event shopping journeys
- Evaluate session-based recommendations
Access & formats
JSONL. External dependencies. Follow the project’s instructions on GitHub; some resources require external files, registration or approval.
License & reuse
The README explicitly licenses the dataset under CC BY 4.0 and the code under MIT. Attribution is required for data reuse.
Important limitations
Article identifiers are anonymized. This is a historical benchmark without descriptive live product catalogs.
Source & attribution
otto-de — original GitHub project ↗
Documentation reviewed 2026-09-24. Review scope and licensing notes. Suggest a correction.
Related resources
Coveo SIGIR Ecommerce
coveooss
Search and browsing sessions for studying ecommerce recommendation and purchase intent.
CSV · Application / terms
Coveo Shopper Intent
coveooss
Session-level ecommerce events supporting research into early prediction of shopping intent.
CSV · Application / terms
Coveo Cross-shop Embeddings
coveooss
Two-shop session and item representations for evaluating transfer and recommendation across retail catalogs.
JSON / embedding files · Application / terms