Evgeny Kagan

Evgeny Kagan

Associate Professor at Johns Hopkins University, Carey Business School

Current Research

Can AI Chat Replace Traditional Search? Evidence from a RAG-based Live Events Experiment (with Rakesh Allu)

Conversational AI is increasingly built into the search interfaces of online platforms. In this paper we run a pre-registered online experiment in which participants search a catalog of several thousand live events using one of three interfaces: an attribute-based filter interface, a GenAI chat interface, or a combination of the two. We find that the chat-only interface performs worst. Two mechanisms drive the result: users cannot see what the catalog contains, so they search for events that are not there, and they shift effort towards prompting and away from evaluating the output. Read more

What’s in a Queue? An Experimental Study of Job Ordering, Autonomy and Queue Visibility

How a queue of jobs is arranged and presented to workers is an important design problem in service operations. In this paper we use online experiments to examine how job ordering, worker autonomy and queue visibility affect the speed and quality of work. We find that workers who choose their own job order perform worse than workers with an imposed order, and that seeing new jobs arrive produces a short-term productivity burst. Read more

One at a Time? The Personal Productivity Bias in Emergency Department Patient Assignment (with Brett Hathaway and John Jones)

Emergency departments often use a shared-queue setup in which physicians self-assign cases from a pool of triaged patients. Using data from five emergency departments and experiments with healthcare workers and ED physicians, we show that batching, i.e., self-assigning multiple patients at once, is common, is associated with longer stays for batched patients, and persists even when it reduces physicians’ own payoffs, a behavior we term the personal productivity bias. Read more

Mentoring in Startup Ecosystems (with Jeffrey Sanchez-Burks, David Brophy, Thomas Jensen and Melanie Milovac)

In this report (commissioned by the Kauffman foundation) we look at how entrepreneurial accelerators and incubators organize their mentoring initiatives. We distill some of the best practices and discuss the differences between university and non-university accelerators, and the lessons learned for setting up such programs. Read more