Happy Moments

How to cite

Serra-Garcia, Marta; Gneezy, Uri. 2024. "Happy Moments." AEA RCT Registry. AEARCTR-0014626. https://doi.org/10.1257/rct.14626-1.0

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

Marta Serra-Garcia* , Uri Gneezy

* Primary investigator

Registry ID
AEARCTR-0014626
DOI
https://doi.org/10.1257/rct.14626-1.0
First registered
Last updated
Status
In development
Trial dates
2024-06-01 – 2025-02-28
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/14626

Abstract

This experiment aims to test whether reminding customers of “happy moments” (HM) with a product (a positive experience) and varying the timing of the reminder and ask to refer (call to action or CTA) affects the likelihood that a customer refers the product. The customers are randomly assigned to one of 5 main conditions: (1) No communication, (2) Communication 1 day after HM + CTA to refer with reminder of HM; (3) Communication 1 day after HM + CTA to refer with NO reminder of HM; (4) Users have a HM - Communication 1 week after HM + CTA to refer with reminder of HM; and (5) Users have a HM - Communication 1 week after HM + CTA to refer with NO reminder of HM.
A credit card company implements the field experiment, and provides us the data thereafter.

Intervention

There are 4 treatment conditions and one control condition:
1. Users have a HM - No communication (control)
2. Users have a HM - Communication 1 day after HM + CTA to refer with reminder of HM
3. Users have a HM - Communication 1 day after HM + CTA to refer with NO reminder of HM
4. Users have a HM - Communication 1 week after HM + CTA to refer with reminder of HM
5. Users have a HM - Communication 1 week after HM + CTA to refer with NO reminder of HM
There are different types of messages/HM: NPS, Cashback redemption, MSI, CSAT, referral, and generic message.

Experimental design

There are 4 treatment conditions and one control condition:
1. Users have a HM - No communication (Control)
2. Users have a HM - Communication 1 day after HM + CTA to refer with reminder of HM
3. Users have a HM - Communication 1 day after HM + CTA to refer with NO reminder of HM
4. Users have a HM - Communication 1 week after HM + CTA to refer with reminder of HM
5. Users have a HM - Communication 1 week after HM + CTA to refer with NO reminder of HM

Design details

Randomization method
Randomization done by a computer.
Randomization unit
Individual
Clusters
500,000 active users. There are strata: Age, Months at the company, Number of logins in previous month, Dormant, Previous referrals, Happy moment in previous month, Gender, Spending level previous month, Credit limit.
Observations
500,000
Arms
Equally divided by treatments.
Primary outcomes
The fraction of participants who refer the product to others, the number of referrals, and whether the referrals are approved. We also plan to measure their engagement and repayment behavior with the card later on.

Cited by

No citation data available for this registration yet.

Keywords