This tutorial demonstrates how to use Haystack-based semantic question answering systems to extract information from a collection of financial statements. It covers the benefits of automating information extraction, including reduced manual effort and improved accuracy. The tutorial provides a step-by-step guide on building an information extraction pipeline using Haystack's NLP framework, including preprocessing documents, indexing them into a database, and constructing a question answering pipeline with retriever and reader components. The example shows how to use the pipeline to extract answers from a 300-page document in a matter of seconds. It also highlights the importance of metadata filtering to prevent processing duplicate documents and improve efficiency. Overall, the tutorial provides a practical example of using Haystack for information extraction and demonstrates its potential to streamline data analysis and processing tasks.