How investors can improve at expert calls and AI with AlphaSense's Ryan Fennerty

Episode Summary

Executive Summary: Andrew Walker and Ryan Fennerty explore how expert calls and AI are reshaping investment research. The central message: the best expert calls test a thesis with multiple credible viewpoints, while AI is rapidly becoming the force that synthesizes transcripts, earnings, internal research, and external data into faster, more disciplined investment workflows. The conversation emphasizes triangulation, bias management, and the shift from information gathering to judgment and differentiated interpretation.

Main Topics: How to use expert calls better (Priority: 5/5): The strongest expert calls start with a clear hypothesis or thesis, then use structured-but-flexible questioning to validate or invalidate it rather than hunt for confirmation. Evolution of expert networks and libraries (Priority: 5/5): Expert-call libraries have expanded access dramatically, lowering costs and moving much of the basic triage and up-to-speed work away from live calls and into searchable archives. Bias in expert calls and transcript reading (Priority: 5/5): Both investors and experts bring bias; the key is triangulation across multiple sources, checking an expert’s true role, and using open-ended questions to expose hidden risks and second-order effects. AI as an investment workflow accelerator (Priority: 5/5): AI is strongest where investors need rapid synthesis across many sources, especially earnings season, portfolio monitoring, and comparing management, sell-side, expert, and internal views. How AI changes concentrated investing (Priority: 4/5): For long-term public and private investors, AI compresses the time needed to get from raw information to a credible view, increasing conviction on fewer deals while expanding the number of opportunities screened. Future skill sets for investors (Priority: 4/5): Fennerty argues that analytical production is becoming commoditized; future edge comes more from pattern recognition, judgment, architecture of workflows, and the ability to ask better questions and act decisively. Expert calls, AI, and corporate users (Priority: 3/5): AlphaSense is increasingly used not only by investors but also by corporate strategy, corporate development, and IR teams, who use the same insight to inform decision-making and strategy.

Key Arguments: Expert calls work best when used to test a hypothesis with a credible thought partner, not to force an expert to confirm a preconceived conclusion. The market has moved much of the basic triangulation and first-pass learning from live calls into expert-call libraries, which are now the starting point for deeper work. Bias is unavoidable in management commentary, sell-side research, financial data, expert views, and investor prompts; the answer is not eliminating bias but triangulating across biased sources. Investors should verify an expert’s role, use multiple experts, and include “burner” or open-ended questions to reveal hidden risks and tone shifts. AI is most valuable where humans need to synthesize many inputs quickly, such as earnings analysis, portfolio monitoring, and comparing management guidance against peers and experts. For concentrated investors and PE firms, AI does not necessarily increase the number of deals, but it increases the speed and quality of diligence and conviction on selected opportunities. The biggest long-term skill shift is from manual analysis toward judgment, pattern recognition, and workflow design; the human edge remains in knowing what makes sense and what does not. AI and expert calls are complementary: AI can scale transcript review, summarize vast research sets, and even power AI-led interviews, while human judgment still determines what matters.

Data Points: Average expert call price before disruption: $1,500 to $2,000 per call - Fennerty described the pre-Tegas market as highly expensive and selective. Typical number of expert calls per year by host: 25 to 50 - Walker estimated his own annual expert-call usage. Breakdown of Walker’s expert calls: 10% awesome, 50% good, 40% okay, 10% bad - His framing for improving the quality of expert calls. Typical investor workflow reduction: Two weeks to one day - Fennerty said expert libraries now compress the time needed to get up to speed on a topic. Private-market adoption lag: 18 months - He said private-market investors have been mimicking public-market expert-call behavior after an approximately 18-month lag. Corporate user share at AlphaSense: Almost 40% - AlphaSense’s business from CorpDev, CorpStrat, and IR teams. Private-market deal count example: 3 to 5 deals per year - Example of funds that remain focused on a small number of concentrated bets despite AI improvements. Portfolio monitoring cadence example: Every Friday - Example of an AI-generated recurring report format for portfolio trend monitoring.

Pivotal Quotes: "I am testing a hypothesis or a thesis, and I want a thought partner who's credible to think through that." — Ryan Fennerty: His core advice on how to frame expert calls effectively. "The number one thing is to frame expert calls are really well utilized for humans who can help you think through a hypothesis you have and really help test your thinking on that." — Ryan Fennerty: Reinforcing that expert calls should validate or invalidate an investment thesis. "AI is actually one of the biggest places where we're early." — Ryan Fennerty: On how AI is beginning to reshape expert calls and related workflows.

Implications: Investors who combine better thesis-driven expert calls with AI-based synthesis will likely outcompete those relying on manual research alone. The edge shifts from collecting information to filtering it, triangulating bias, and making faster, higher-conviction judgments.