Free trials · Randomized field experiment

Social Referral Programs for Freemium Platforms

Management Science · 2022

Rodrigo Belo, Ting Li

Summary

Referral programs increased referrals but could reduce revenue when referred users gained access to premium features without paying.

For your marketing

Possible founder application

Useful when a product has social invitations or a free tier: track invited users, paid conversion, and engagement together, and test which benefits are available to nonpayers.

Reported result

What the study found

Referral programs increased referrals but could reduce revenue when referred users gained access to premium features without paying. Higher referral thresholds increased referrals and total revenue, while reducing some user engagement. Reserving selected premium features for paying users helped avoid the revenue loss.

Research question

What the authors investigated

How do social referral programs affect growth, payment, and engagement on freemium platforms?

Design

Two-year randomized field experiment that introduced referral programs and varied referral requirements on an online dating platform.

Sample and setting

52,782 approved new users on a dating platform in a major European city, observed from March 2015 to February 2017.

Scope and transfer

Limitations to keep in view

One online dating platform with distinctive social-network and turnover dynamics. Customer lifetime value results also depend on modeling assumptions; the exact trade-offs may differ for SaaS products.

Research sources

Publisher abstract and source page inspected with Firecrawl; summary is abstract-level and limited to the reported experiment.

Study strength

4.5/5 Strong design for its specific question

How this was assessed

The 1–5 evidence-check score is a practical reading aid, not a formal peer review or a probability that a conclusion is true. It combines design strength (2 points), sample reporting (1), comparison/control clarity (1), and how much of the paper could be checked (1). Access affects how confidently this directory can assess a study, not the study's underlying quality. Blinding is reported separately because it is not relevant to every design.

Design · 2/2

Random assignment provides a strong comparison for effects in the tested setting.

Sample · 1/1

Large sample reported (52,782); size alone does not guarantee a representative sample.

Comparison · 1/1

Randomized comparison; confirm the control condition and analysis in the paper.

Can we check it? · 0.5/1

This profile could be checked against an abstract or metadata, not the full paper.

Double-blind? Blinding is not reported in the source reviewed; that does not establish whether blinding was or was not used.