The impact of an AI platform advisor on informal firms in Indonesia

Last registered on May 18, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
The impact of an AI platform advisor on informal firms in Indonesia
RCT ID
AEARCTR-0018503
Initial registration date
May 14, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 18, 2026, 7:20 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Xavier Cirera
Affiliation
The World Bank
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Xavier Cirera
PI Affiliation
World Bank
Contact Investigator
PI Name
Diego Comin
PI Affiliation
Darthmouth College
Contact Investigator
PI Name
Marcio Cruz
PI Affiliation
International Finance Corporation
Contact Investigator
PI Name
Kyungmin Lee
PI Affiliation
World Bank
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-05-17
End date
2027-07-31
Keywords
Firms & Productivity
Additional Keywords
Informal sector, technology adoption, artificial intelligence, ERP software, small and medium enterprises, digitalization, Indonesia, formalization
JEL code(s)
O12, O14, O17, D22, L26, C93
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Informality remains a persistent challenge in developing economies, constraining firm growth, aggregate productivity, and job quality. In Indonesia, where an estimated 93 percent of firms operate informally, reducing informality requires more than lowering registration costs. It requires addressing information gaps, behavioral constraints, and limited access to business services. The emergence of AI-powered technologies offers a cost-effective channel to deliver a range of services that will encourage firms to formalize, without the need for physical infrastructure. Moreover, recent evidence suggests that entrepreneurs often overestimate their technological and managerial capabilities, leading to systematically low adoption of technologies despite their potential productivity gains.

This study uses a randomized controlled trial (RCT) to evaluate whether access to AI tools can improve firm performance and encourage formalization among informal firms in Indonesia. Six-hundred (600) informal firms drawn from the 2023 World Bank Enterprise Survey sampling frame in Jakarta and one additional large Indonesian city are randomly assigned to treatment and control groups. Treatment firms receive access to digital business tools along with mobile-delivered information nudges on business registration and industry benchmarking designed to address overconfidence. Primary outcomes include intensity of digital tool adoption, business registration, sales, profits, employment, and management quality. The study includes a midline survey in August–September 2026 and an endline survey in 2027.
External Link(s)

Registration Citation

Citation
Cirera, Xavier et al. 2026. "The impact of an AI platform advisor on informal firms in Indonesia." AEA RCT Registry. May 18. https://doi.org/10.1257/rct.18503-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
This study evaluates the impact of providing informal firms in Indonesia with access to AI tools to support business operations and decision-making. Firms in the treatment group receive access to AI tools along with information nudges via mobile messaging on business formalization and industry benchmarking. The study examines whether access to these tools improves firm performance, technology adoption, and business formalization among informal sector enterprises.
Intervention Start Date
2026-05-18
Intervention End Date
2027-07-31

Primary Outcomes

Primary Outcomes (end points)
Adoption and intensity of technology adoption; business registration
Primary Outcomes (explanation)
Digital tool adoption will be measured using platform usage logs (AI assistant and ERP app traffic/activity data) as well as self-reported use from follow-up surveys. Business registration will be measured as a binary indicator of whether the firm has formally registered or reported intent to do so, captured at midline and endline surveys.

Secondary Outcomes

Secondary Outcomes (end points)
Sales; profits; employment; cost per unit of output; management quality; adoption of additional digital technologies
Secondary Outcomes (explanation)
All secondary outcomes are self-reported through midline (August–September 2026) and endline (2027) surveys. Sales and profits are measured in Indonesian Rupiah over the reference period. Employment is measured as the total number of workers including the owner. Management quality is assessed using a structured index of planning, monitoring, and operational practices. Technology adoption is measured as an index of digital tools used within the firm for general business tasks (business administration, procurement, sales, payment, production planning, fabrication, and quality control).We measure the breadth of technologies used (MAX) and the depth of use of the most intensively adopted technology (MOST), following the methodology in Cirera, Comin, and Cruz (2026).

Experimental Design

Experimental Design
This study uses a randomized controlled trial (RCT) to evaluate the impact of access to AI-powered digital tools on the performance and formalization of informal firms in Indonesia. Eligible firms are drawn from a sampling frame of informal enterprises in large Indonesian cities. Following a baseline survey, firms are randomly assigned to either a treatment or control group. Treatment firms receive access to digital business tools along with information nudges via mobile messaging. The study includes a midline survey (August–September 2026) and an endline survey (2027) to measure impacts on firm performance, technology adoption, and formalization.
Experimental Design Details
Not available
Randomization Method
Computer-generated random assignment conducted after the baseline survey. Firms are assigned to treatment and control groups using a random number generator, with assignment stratified as needed to ensure balance across arms.
Randomization Unit
Establishment
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
Not applicable — treatment is not clustered.
Sample size: planned number of observations
600 firms
Sample size (or number of clusters) by treatment arms
150 firms: control group;
150 firms: AI-assistant only;
150 firms: enterprise resource planning (ERP) software only;
150 firms: both AI-assistant and enterprise resource planning (ERP) software

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations are based on a total sample size of 600 firms, with 150 firms per arm. Assuming a significance level of 0.05 and 80% statistical power, the study is powered to detect a minimum detectable effect (MDE) of [INSERT MDE] standard deviations on the primary outcomes of digital tool adoption and business formalization. Calculations assume a standard deviation of [INSERT SD] for the primary outcome and an anticipated attrition rate of [INSERT %, e.g., 10–15%] between baseline and endline.
Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan