AI-driven research workflows cut analyst turnaround time by up to 65%

A US-based asset manager sought our help to shorten the time gap between data release and investment research workflows. We enabled faster turnaround by developing a suite of artificial intelligence workflows covering the three core phases of research (earnings call interpretation, financial statement analysis and research note drafting).

• Manual transcript parsing to extract guidance and gauge management tone required repeated reading and cross-referencing
• Repetitive ratio, growth-driver and balance sheet analysis was taxing during earnings seasons
• Note structure and quality varied across analysts, with investment theses being compiled manually from scattered sources
• Tight post-earnings deadlines raised the risk of missing or misattributing guidance and red flags
• Lack of standardized logic led to heavy dependence on the analyst’s rationale, causing variations in output quality and turnaround time

We designed three purpose-built AI workflows that share common grounding and formatting standards:

• Earnings call interpretation: Extracts guidance figures and management tone, anchored in verbatim quotes, and works consistently on transcripts of any length or complexity
• Financial statement analysis tool: Calculates ratios, traces growth drivers and assesses balance-sheet strength
• Research note drafting: Generates a thesis covering positives, risks and valuation from the source material, ready for client use with minimal editing