Data AnalysisInternal OnlyVerified

Reviewed and published by trentmaziarz, March 23, 2026. Discovered and drafted by our automated research pipeline.

Pinterest uses an AI system to classify each user's active interests into named 'journeys,' such as planning a nursery or exploring Japanese cuisine, and applies those journey labels to personalize the timing and content of email and push notifications. The system was described in an October 2025 Pinterest Engineering blog post. In initial experiments covering April-May 2025, journey-aware notifications produced an 88% higher email click rate and a 32% higher push open rate compared to the prior interest-based notification system.

Details

The system extracts keywords from a user's recent searches, saves, and boards, then uses hierarchical clustering to group those keywords into journey candidates, which are then named, ranked, and classified by stage. Pinterest used GPT to generate ground-truth labels for fine-tuning Qwen, a smaller open-source model, so that journey classification could run at scale with lower cost and latency. The journey labels are matched to relevant Pin content and used to personalize email and push notification timing and content. Pinterest also uses an LLM-based evaluation pipeline to score the relevance of predicted journeys, which the engineering team reports correlates closely with human assessments.

Products affected

Pinterest email notificationsPinterest push notificationsPinterest recommendations

Sources & Evidence

News Article

https://www.zenml.io/llmops-database/user-journey-identification-using-llms-for-personalized-recommendations

News Article

https://medium.com/pinterest-engineering/nep-notification-system-and-relevance-a7fff21986c7

Company Disclosure

Company DisclosureVerified

Identify User Journeys at Pinterest

Pinterest Engineering Blog·Oct 2025

https://medium.com/pinterest-engineering/identify-user-journeys-at-pinterest-b517f6275b42

Cite this record

Trace Foundation. (2026). Pinterest: Pinterest uses an AI system to classify each user's active interests into named 'journeys,' such as planning a nursery or exploring Japanese cuisine, and applies those journey labels to personalize the timing and content of email and push notifications. The system was described in an October 2025 Pinterest Engineering blog post. In initial experiments covering April-May 2025, journey-aware notifications produced an 88% higher email click rate and a 32% higher push open rate compared to the prior interest-based notification system (data as of 2026-06-06) [Data set record]. AI Trace. https://www.aitrace.org/r/practice/59dc3dc3-f821-4154-9395-73ef646522e5. Accessed October 14, 2026.

Stable link: https://www.aitrace.org/r/practice/59dc3dc3-f821-4154-9395-73ef646522e5
Data as of: June 6, 2026
Last verified: March 23, 2026

How to cite AI Trace

Other practices by Pinterest

OtherPinterest maintains an internal program called Responsible AI that covers three areas: building inclusive AI features (the skin tone, body type, and hair pattern tools described in Practice 21), testing its AI systems for unfair bias (using internal tools that measure whether models perform equally well across different groups of users), and developing guidelines for the safe use of generative AI. The program includes red teaming — structured attempts to find flaws or harmful outputs in AI systems before they launch.OtherPinterest is testing AI-powered upgrades to its Boards feature that go beyond simple user-organized collections. In October 2025, Pinterest launched an experiment in the U.S. and Canada where the platform automatically generates outfit ideas from a user's saved fashion Pins ("Styled for you"), creates entirely AI-curated boards based on trending styles ("Boards made for you"), and suggests related products based on what a user has already saved ("Make It Yours").OtherPinterest uses AI to identify diverse skin tones, body types, and hair patterns across billions of images on its platform, then uses those signals to ensure that search results and recommendations reflect a broader range of human appearances. Users can filter fashion search results by body type ranges, skin tone ranges, and hair patterns. The body type technology launched in September 2023, with public user filters rolling out in March 2024; skin tone ranges have been available since 2018.

Have evidence about Pinterest's AI practices? Submit a report.

Report a Sighting →