The Selective Tailwind Effect of A.I. on Startups: Predictions and Anomalies

How to cite

Conti, Annamaria; Messinese, Danilo. 2026. "The Selective Tailwind Effect of A.I. on Startups: Predictions and Anomalies." AEA RCT Registry. AEARCTR-0018524. https://doi.org/10.1257/rct.18524-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Annamaria Conti* , Danilo Messinese

* Primary investigator

Registry ID
AEARCTR-0018524
DOI
https://doi.org/10.1257/rct.18524-1.0
First registered
Last updated
Status
In development
Trial dates
2026-05-04 – 2026-05-31
Countries
Private
JEL codes
J24, L86, O31, O32, O33, D81, D83, M13
Registry record
http://www.socialscienceregistry.org/trials/18524

Abstract

What role does AI play in entrepreneurial decision making? We have partially explored this question by analyzing the impact of AI predictive tools on the performance of a large sample of US startups. We have exploited the release of Google Analytics 4 (GA4) which introduced AI predictive tools especially useful for mobile app developers. We found some evidence that post-GA4-release there is a boost in customer acquisition. The positive premium is largest for skilled human capital. These preliminary findings suggest that AI predictive tools are useful for complementing skilled human capital in formulating new testable business hypotheses, especially relevant for the detection of breakthroughs.

To further explore our results obtained from large-scale data, we plan to conduct a lab experiment organizing an in-person Hackathon involving a large spectrum of students enrolled at schools in Madrid area. The goal is to better understand the role of predictive AI tools for anomaly detection for the production of breakthrough innovations, as well as the interaction of these tools with human capital.

Intervention

The goal of this lab experiment is to test whether the use of AI for anomaly detection can improve the ability to identify innovation opportunities and whether these tools complement or not human capital. The treatment in this experiment is a nudge about the importance of anomalies detection to uncover novel ideas and business opportunities, delivered through a short video and an ad hoc AI platform which provides standard predictive tools plus anomaly detection. The control group is administered a video about the importance of predictive AI for innovating and the same platform as the treated group but without the anomaly detection tool.

More specifically, each group will access a platform developed by the research team to analyze data from a B2C software startup. This platform closely replicates Google Analytics 4 (GA4). Participants will be given a dataset simulating user behavior for a consumer wellness and sleep app as of 2026, including variables such as user sessions, revenue, engagement, and customer acquisition. The dataset, in addition to data trends, contains anomalies that, if detected, could inspire innovative solutions. Notably, the treated group will have access to a default option within the platform that highlights data anomalies as part of the treatment.

Experimental design

The goal of this study is to evaluate whether AI-based anomaly detection enhances the ability to identify innovation opportunities and to explore whether these tools complement human capital. Participants will participate in the IE VentureMind AI Hackathon and will be tasked with analyzing data from a B2C startup, with the objective of proposing innovations based on their analysis.

Participants will be provided with an ad hoc data-analysis platform developed by the research team containing time-series data on key performance metrics in 2026.

Experimental Design Details
We expect the group that receives the anomaly detection treatment to outperform the control group in terms of our primary outcomes, but with variance depending on the participants' human capital. To measure participants' human capital, we will administer a Qualtrics survey collecting information about participants' backgrounds and including questions aimed at assessing their ability to formulate business theories and hypotheses.

We will recruit at least 70 participants per group (depending on how many participants will show up at the hackathon). Participants are students (bachelor's, master's, MBA, and PhD) located in the Madrid area. We will reward the three most innovative projects with a monetary prize.

The Hackathon will last two days. During the first day, participants will be asked to act as the CEO of the Arora company and produce a business plan, which might revise the initial board's plan. To produce the business plan, the participants will have to analyze data displayed on an ad hoc AI platform.

Design details

Randomization method
Randomization into the treated and control group described above will be performed by Qualtrics
Randomization unit
Participants in the IE VentureMind AI Hackathon
Clusters
0
Observations
At least 70 units per group, but the final number dpends on how many people attend the Hackathon event.
Arms
At least 70 units per group.
Primary outcomes
Level of innovativeness of business plan submitted by participants
Secondary outcomes
Viability of the business plan

Cited by

No citation data available for this registration yet.

Keywords