Case Study: ZEIT ONLINE improves personalized news discovery with deepset's Haystack Enterprise Platform

How major German news platform Zeit Online is using LLM technology to improve content discovery

ZEIT ONLINE, a major German news platform, sought to improve reader engagement by making its vast archive of content more discoverable and personalized. The challenge was to move beyond traditional news delivery and leverage generative AI to create a more interactive and intuitive user experience for its readers. They partnered with deepset, utilizing the Haystack Enterprise Platform to build a solution.

deepset provided a solution using a retrieval augmented generation (RAG) setup powered by their Haystack Enterprise Platform. This allowed ZEIT ONLINE to create "Ask Zeit Online," a tool where readers ask questions and receive summarized answers with citations directly from the news archive. The result was a highly successful product that improved content discovery, received positive user feedback, and became a staple on the website's homepage after a beta test with one thousand users.

ZEIT ONLINE

Sebastian Horn

Deputy Editor-in-Chief