A group of new fund houses is adopting a data-led approach rather than traditional active fund management to differentiate themselves in India's mutual fund industry. According to reports from Mint, India's mutual fund industry currently has 55 fund houses managing over 1,900 open-ended schemes, including more than 500 actively managed equity funds. These emerging asset managers are deploying multi-factor algorithms, alternative datasets, and market-regime rules to eliminate human emotion from investment decisions. The shift toward algorithmic decision-making reflects a broader trend where AI and advanced analytics introduce qualitatively new evaluation capabilities by processing data volumes and detecting patterns beyond human cognitive capacity. As per recent AI decision-making research, AI thrives on data but humans are essential for providing context, creating a powerful partnership where humans and AI systems can form a collaboration that influences how decisions are shaped.
Capitalmind Mutual Fund scores its investable universe on five factors—momentum, low risk, quality, profitability and value, but the weight each factor carries shifts with market-regime classification. As reported by Mint, Anoop Vijaykumar, head of equity at Capitalmind Mutual Fund, explained that this regime interpretation is entirely data-driven, evaluating metrics such as market breadth, performance of an equal-weighted index relative to a market cap-weighted index, and how wide factor valuation spreads compare to historical trends. The dynamic factor weights yield a composite score for every stock, which the model then uses to generate a portfolio. This approach aligns with the Integrated Personnel Evaluation Model (IPEM), which advocates for balanced indicators capturing input (capabilities), process (quality/collaboration), and output (tangible results) metrics to reduce gaming incentives and improve evaluation validity. The AI-driven approach allows for faster, smarter decisions by providing up-to-date insights that can guide actions, enabling sales reps to prepare for client meetings with curated views of customer preferences and buying patterns rather than digging through numerous reports.
Jio BlackRock Mutual Fund describes its investment strategy as 'systematic active', evaluating approximately 400 data signals per Indian stock and actively deploying 30 to 50 metrics per fund. According to Mint, Rishi Kohli, chief investment officer at Jio BlackRock Mutual Fund, explained that the sentiment bucket focuses on analyst sentiment, while the fund also taps alternative data such as job postings. AlphaGrep Mutual Fund combines a multi-factor, adaptive model with alternative data sources including analyst estimates, earnings revisions, news and social sentiment, order-book and flow data, and analyst call transcripts. This aligns with the AI and advanced analytics integration that enables descriptive, predictive, and prescriptive insights far beyond human cognitive capacity while maintaining human oversight to prevent algorithmic systems from reproducing historical biases. The AI-powered approach can handle massive datasets and make complex tasks more manageable, with AI-powered dynamic pricing in e-commerce adjusting prices based on demand, inventory, and competitor activity to maximize profitability without manual intervention.
While these fund houses aim to deliver more resilient, tech-driven models by drawing on broader datasets, their real-world performance across full Indian market cycles remains to be proven. As reported by Mint, Vishal Dhawan, founder of Plan Ahead Wealth Advisors, noted that while several rule-based strategies have been widely tested globally and demonstrated performance, these strategies have been back-tested on historical data only. The challenge lies in testing these strategies over longer periods in Indian markets to validate their effectiveness in real-world conditions. Research by Scullen and colleagues demonstrates that performance ratings partition into three variance components: actual job performance, rater-specific effects, and measurement error, with rater effects often explaining variance proportions comparable to or exceeding true performance variance. This underscores the need for algorithmic governance structures ensuring fairness, transparency, and employee rights protection to build evaluation capabilities that are simultaneously more objective, more predictive, and more developmental than legacy approaches. The AI-driven approach can eliminate bottlenecks by automating repetitive processes, with customer support AI agents answering questions and guiding customers to solutions while saving operational costs, but businesses must address challenges including incomplete, outdated, or biased data that can lead to flawed predictions.
