Asset Management

Needl.ai Transforms Investment Firm’s Research Workflow and Decision Making

The customer is a prominent Indian boutique investment firm that focuses on direct investments in listed Indian equities. With close to US$1B AUM, analysts rely on a vast array of financial data sources to produce high-quality research reports and intelligence briefs.

26 Mar 2025
4 min read
Share this post

Challenge

The customer faced key operational inefficiencies in its research and intelligence processes, leading to increased workload, slower turnaround times, and inconsistent outputs.

Ineffective Research Workflows

Analysts spent excessive time manually filtering through news, reports & market data, leading to inefficiencies in report generation & delays in decision-making.

Excessive Noise

Existing news feeds contained a lot of irrelevant & generic market news, share price movements & trading volumes, making it difficult to extract relevant, valuable insights which were lost in the noise.

Lack of Customization & Control

Analysts wanted more flexibility in terms of report templates, keyword exclusions, and real-time refinement of AI-generated content.

Inefficient Learning Mechanisms

Their existing platform lacked an adaptive system that could learn from user feedback to improve content relevance and reduce manual adjustments.

Solution

The solution was implemented within 15 weeks, following four customer-defined milestones. Needl.ai deployed its solution on the customer’s Azure cloud, adhering to required standards and custom optimizations. It ingested stock exchange updates, company announcements, and high-impact market-making news on weather, outages, and regulatory changes from public and government sources. All public data was processed securely on the customer’s cloud and accessed via APIs with appropriate security measures.

Automated Information Extraction: Needl.ai parsed reports and key news alerts to extract relevant insights.
Intelligent Feed Curation: Filters reduced noise and prioritized relevant content.
Custom Report Generation: AI-generated reports were tailored to the customer’s specific templates, often condensing 500+ pages of information into concise 20-page reports containing only relevant insights with references and citations to source document(s).
Learning-Based Adaptation: The filtration and prioritization of insights improved with user feedback and movements, thus reducing manual effort.
Conversational Interface: RAG-powered interface allowed analysts to query data seamlessly for quick insights and decisions through natural conversation.

Outcome

Within the first year of deployment, Needl.ai delivered measurable improvements for the customer:

Increased Research Throughput

Analysts produced 144+ quick note reports annually, with each note being generated within 5 days, instead of ~10 days, improving efficiency by 50+%.

Increased Accuracy of Reports

Complex AI-generated custom reports maintained a 95% to 100%+ accuracy rate, due to reduced noise and  irrelevant financial data (industry standards using LLMs are today at 16% accuracy).

High User Satisfaction

The analysts reported an average customer satisfaction score of 8.7 on a scale of 10 (with 10 being best).

The ability to generate reliable, high-quality summaries and reports from extensive datasets allows us to focus on deeper analysis and quick decision-making to seize the right opportunities. The AI-driven learning mechanism ensures that our research is continuously improving – and Needl.ai has completely surpassed our initial expectations!

– Senior Analyst