Illustration of a head silhouette with gears, hands interacting with a data analytics dashboard, and various data icons, symbolizing data engineering and analytics for a fitness startup's growth.
SUCCESS STORY
Stacks of coins with an overlay of a fluctuating financial graph, symbolizing investment analysis and market trends related to the Multi-Agent Investment Research Tool.

Multi-Agent Investment Research Tool

A leading private equity (PE) funding firm partnered with ThirdEye Data to organize its investment research process. Focused on identifying growth-stage IT companies, the firm sought to overcome the inefficiencies of manual research workflows that limited its ability to scale. Traditionally, every investment analysis required 8–12 hours of manual evaluation per company, making it impossible to assess hundreds of potential opportunities effectively.

ThirdEye Data designed and implemented a Multi-Agent Investment Research Tool, a Copilot-based assistant that automates the end-to-end process of investment discovery, data collection, analysis, and reporting. By combining Microsoft’s Power Platform, Azure AI, and multi-agent orchestration, the solution enables the firm to evaluate hundreds of companies in parallel, achieving deeper insights at a fraction of the time and cost.

THE CUSTOMER

The customer is a leading growth-focused private equity firm in India, managing assets across multiple sectors and supporting mid-sized, high-potential companies that align with India’s growth story. The firm’s investment philosophy revolves around scalability, operational excellence, and early identification of promising businesses.

With a portfolio spanning IT services, infrastructure, and digital transformation ventures, the firm aims to leverage technology to accelerate its decision-making and enhance its research-driven investment approach.

BUSINESS GOALS OR CHALLENGES

Business Goals

  • Automate the discovery and evaluation of 600–800 growth-stage IT companies across India.
  • Replace manual analyst research (8–12 hours per company) with AI-driven automation.
  • Enable interactive, natural language–based investment research through Microsoft Copilot.
  • Create a centralized intelligence system integrating data from multiple web and document sources.
  • Improve decision-making accuracy while maintaining human-in-the-loop validation for final recommendations.

Understanding the Challenges:

  • Research relied heavily on analyst judgment, making it slow and difficult to scale.
  • Data scattered across PDFs, websites, and reports required manual collation and interpretation.
  • Lack of integration between data sources prevented unified company profiles.
  • Limited bandwidth meant several promising opportunities were missed due to time constraints.
  • Decision-making lacked structured intelligence, making performance benchmarking difficult.

Prerequisites and Preconditions:

  • Clear definition of investment criteria and prioritization filters.
  • Access to company datasets, APIs, and relevant financial reports.
  • Secure Azure environment with Power Platform integration.
  • Defined validation workflow involving analysts, investment managers, and IT.
  • Proof-of-concept (PoC) phase covering top 100–1000 IT companies before full-scale rollout.

THE SOLUTION

ThirdEye Data built a Multi-Agent Investment Research Tool powered by Microsoft Copilot Studio and Azure OpenAI. The system enables users to conduct conversational investment research through natural language queries, orchestrating multiple AI agents to discover, parse, analyze, and rank companies in real time.

The Copilot interface acts as a “research orchestrator”, interpreting investor intent, triggering automated workflows via Power Automate, and returning structured, ranked insights from Dataverse and Power BI dashboards. This system dramatically reduces analyst workload and improves the precision and speed of research-based decisions.

Solution Highlights

  • Natural Language Interaction via Copilot Studio:
    Investors can ask questions such as “Show me top 20 IT service companies with revenue growth >25% and strong customer sentiment.” The AI orchestrator understands, retrieves, and ranks relevant companies automatically.

  • Automated Data Ingestion & Processing:
    Power Automate and AI Builder scrape and ingest structured and unstructured data from multiple sources, including PDFs, financial reports, and websites.

  • Advanced Analysis & Insights:
    Azure OpenAI performs reasoning, summarization, and ranking. AI Builder extracts entities, classifications, and sentiment from qualitative data to generate a holistic company view.

  • Decision Support & Reporting:
    Power BI delivers dynamic dashboards with financial ratios, performance overlays, and sentiment insights. Copilot in Power BI enables Q&A-based exploration.

  • Human-in-the-Loop Validation:
    A Power Apps interface allows investment managers to review, annotate, and refine AI-generated recommendations. Adjustments feed back into the system for continuous learning and improved ranking accuracy.

Supported Use Cases

  • Automated screening and shortlisting of 600–800 companies.

  • Financial performance correlation and sentiment analysis.

  • Instant generation of investment-grade company summaries.

  • Dynamic company ranking and opportunity scoring.

  • Interactive dashboards for deal evaluation and validation.

  • Continuous feedback loop for improving AI-driven recommendations.

Technologies Used

  • Copilot Studio:
    Conversational AI interface to interpret investor queries and orchestrate agent actions.
  • Azure OpenAI:
    Advanced reasoning, summarization, and ranking for investment intelligence.
  • Power Automate:
    Multi-agent orchestration, ETL workflows, and Azure service integration.
  • AI Builder:
    Entity extraction, classification, and sentiment analysis from qualitative data.
  • Azure Cognitive Search:
    Large-scale search and retrieval of company and industry data.
  • Azure Document Intelligence:
    Parsing and structuring data from PDFs, reports, and scanned images.
  • Dataverse:
    Centralized knowledge repository for company profiles, KPIs, and scores.
  • Power BI + Copilot:
    Interactive visualizations, financial dashboards, and Q&A-driven insights.
  • Power Apps: Review, annotation, and validation interface for human decision-makers.

VALUE CREATED

Since implementation, the Multi-Agent Investment Research Tool has demonstrated transformative value across research operations:

  • 70–90% Reduction in Analyst Research Time: Reduced average research effort from 8 to 12 hours to less than 2 hours per company.
  • 1,000 Analyst Hours Saved per 100 Companies: Equivalent to ₹30–50 lakh in opportunity cost savings.
  • 5–10x Efficiency ROI: Benchmarked against McKinsey and Deloitte studies on AI-driven research productivity.
  • Faster Decision-Making: Enabled earlier identification of promising mid-cap IT companies, improving investment timing and valuation entry.
  • Scalable Research Without Headcount Increase: Added 5–10x research capacity with existing teams.
  • Human-AI Synergy: Investment managers can override, annotate, and retrain models for continuously improving decision logic.
  • Financial Impact: For firms managing ₹2,000–₹3,000 Cr AUM, even a 0.5% improvement in research-driven investment accuracy equates to ₹10–₹15 Cr+ in value gain.
ThirdEye Data logo featuring a stylized elephant in white on a green background, representing data analytics and AI solutions.

Transforming Enterprises with
Data & AI Services & Solutions.

ThirdEye delivers Data and AI services & solutions for enterprises worldwide by
leveraging state-of-the-art Data & AI technologies.

Talk to ThirdEye