Research

My research examines how organizations identify problems worth solving, generate possible solutions, and evaluate uncertain ideas. Using field experiments and large-scale data, I study the human and technological systems that determine which ideas are discovered, recognized, and funded.

Problems Worth Solving

Which problems attract entrepreneurial effort and capital—and how is artificial intelligence changing that landscape?

Beyond a Technology Lens: Characterizing the Problems Entrepreneurs Solve
With Shai Bernstein, Mimi Chen, and Zilin Ma · HBS Working Paper 27-016, 2026

We map roughly 90,000 venture-backed startups into 265 problem areas, revealing how technologies, founder expertise, and capital allocation shape which problems receive entrepreneurial attention.

Judging the Problem: A Problem-Centric Approach to Early-Stage Venture Evaluations
With Miaomiao Zhang · HBS Working Paper 26-043, 2026

A field experiment with 178 judges evaluating 109 ventures shows that directing attention toward the customer problem changes how experts assess other dimensions of a venture—and that these effects depend on judges’ professional backgrounds.

Recognizing and Funding Ideas

How do evaluator expertise and process design determine which ideas receive support?

Beyond Feasibility Filters: How Domain-Spanning Expertise Enables Recognition of Innovation Potential
With Zoe Szajnfarber, Jason Crusan, and Michael Menietti · Strategic Management Journal, 2026 · 2023 TIM Best Paper Award

In a field experiment with NASA, we find that experts who span multiple domains are better able to recognize how novel solutions can improve an entire system.

Greenlighting Innovative Projects: How Evaluation Format Shapes the Perceived Feasibility of Early-Stage Ideas
With Simon Friis, Tianxi Cai, Michael Menietti, Griffin Weber, and Eva Guinan · Working paper

Focusing evaluators exclusively on feasibility leads them to uncover more implementation barriers, whereas evaluating multiple criteria together better reveals the relationships among feasibility, novelty, and impact.

Related work

• Designing Evaluation Panels: When Does Professional Panel Diversity Predict Startup Success?

• Forecasting the Impact of Ideas: The Role of Concrete Language in Idea Evaluation

• Conservatism Gets Funded? The Role of Negative Information in Expert Evaluations for Novel Projects, Management Science, 2022

• Engineering Serendipity: When Does Knowledge Sharing Lead to Knowledge Production?, Strategic Management Journal, 2021

Humans and AI

When does AI improve human judgment—and when does it make us confidently wrong?

The Narrative AI Advantage? A Field Experiment on AI-Augmented Evaluations of Early-Stage Innovations
With Léonard Boussioux and collaborators · Forthcoming, Management Science · First Place, 2025 Wharton People Analytics White Paper Competition

Black-box AI recommendations improve evaluation quality, but adding persuasive explanations increases compliance without improving decisions—and causes evaluators to reject more promising ideas incorrectly.

The Crowdless Future? Generative AI and Creative Problem-Solving
With Léonard Boussioux, Miaomiao Zhang, Vladimir Jaćimović, and Karim Lakhani · Organization Science, 2024

In a real-world innovation challenge, human-generated solutions are more novel, while AI-assisted approaches produce solutions that are more feasible and higher in overall quality.

The Mean–Variance Innovation Tradeoff in AI-Augmented Evaluations
With Christoph Grumbach and Georg von Krogh · Working paper · 2026 AIM Best Student Paper

AI can raise the average quality of selected ideas while narrowing their range—potentially disadvantaging unconventional ideas with greater upside.

View Full CV.