Summary
Key results
Publication lead times reduced by several dozen percent
Real-time extraction of financial metrics and data
Manual processes threatened competitive advantage
Infront’s newsrooms processed thousands of items daily: company financial reports, news stories, calendar data (e.g., earnings releases), market data, offerings, and analyst recommendations. Manual preparation had become a bottleneck – every minute of delay meant losing the information advantage to competitors. The lack of automation led to:
- Heavy editor workload on repetitive data analysis
- Risk of inconsistencies and errors in reports
- Longer publication times – critical for financial-market clients
Infront needed a solution that would accelerate publication, standardize data, and still preserve full editorial control.
Integrated GenAI editorial environment
Sii Poland delivered a comprehensive system using Large Language Model (LLM) and Natural Language Processing (NLP) models that unifies automatic summarization, KPI extraction, and document classification in a single editorial tool.
The scope included:
- Fine-tuning GPT models on Infront’s historical datasets
- Automated extraction of key financial indicators (e.g., EBITDA, EPS, revenue)
- Context-based document classification (e.g., report, brief, analysis)
- Interactive summaries with edit and translation capabilities
- Integration with Microsoft Azure and the internal editorial platform
- Quality testing and model calibration for financial-publishing requirements
The rollout modernized newsroom workflows, enabling faster, more precise output. The architecture is scalable – new data sources and publication languages can be added with minimal code changes. In practice, the environment is continually extended with additional sources and output languages.
Faster decisions, better data, greater efficiency
With Sii Poland’s solution, Infront achieved a substantial productivity lift – cutting publication times by several dozen percent and eliminating many manual, repetitive tasks. Automated KPI extraction and summarization let teams focus on interpretation and content quality instead of time-consuming document review.
Centralizing processes in a single AI-integrated environment ensured data consistency and standardized communications – Infront’s clients receive reports faster, in a consistent format, without risking omission of key information. As a result, the platform reinforced its position as one of the market’s most trusted financial information sources, increasing user loyalty and competitive advantage. The initiative has become a cornerstone of Infront’s newsroom digitalization – proving that AI can materially accelerate editorial work while preserving high quality and editorial control, and creating a solid foundation for future AI initiatives across the organization.