Holdout Group

Understanding holdout groups in experimentation

Holdout groups: Think of them as a set of users you intentionally exclude from your experiments. Their main purpose is to serve as a baseline. This way, you can measure the effect of your changes against a group untouched by any experimental modifications.

Why are they important? Holdout groups give you a clear picture of how effective your experiments are. They help you see whether the changes you implement actually make a difference. Without them, you'd struggle to determine if improvements are due to your changes or just natural variations.

Control groups: Holdout groups essentially act as your control group. By comparing the behavior of the holdout group with the experimental group, you can identify any significant differences. This comparison is crucial for understanding the true impact of your interventions.

Creating effective holdout groups

Why are holdout groups crucial?

Holdout groups provide a baseline for comparison. They help you see the true effect of experiments. Without them, you'd struggle to identify real changes. For a detailed explanation, check out What is a holdout?.

How to implement a holdout group

Use prerequisite flags to separate groups. Assign a small percentage, like 5%, to the holdout group. This ensures you have a reliable control group for comparison. For more details, see How to use Holdouts and Best Practices.

For practical implementation, refer to Creating your First Holdout.

Analyzing holdout group data

What metrics to compare

Compare revenue per user and sign-up rates. These metrics show clear differences. They help identify experiment impacts. Learn more about revenue per user and sign-up rates.

How to interpret results

Look for significant differences between groups. Adjust your experiment design based on these findings. This ensures accurate and actionable insights. Learn more about how to create and analyze holdout groups, and explore best practices for holdouts.

For additional resources, visit the documentation and guides.

Real-world applications of holdout groups

E-commerce platforms

Use holdout groups to measure impact on sales and user engagement. Compare how different promotions work. Understand what drives customer behavior.

Subscription services

Evaluate sign-up and retention rates. Identify which features keep users subscribed. Adjust strategies based on data-driven insights.

Mobile apps

Assess feature usage and user behavior. Determine which updates improve user experience. Optimize app performance by learning from holdout data.

Join the #1 experimentation community

Connect with like-minded product leaders, data scientists, and engineers to share the latest in product experimentation.

Try Statsig Today

Get started for free. Add your whole team!

What builders love about us

OpenAI OpenAI
Brex Brex
Notion Notion
SoundCloud SoundCloud
Ancestry Ancestry
At OpenAI, we want to iterate as fast as possible. Statsig enables us to grow, scale, and learn efficiently. Integrating experimentation with product analytics and feature flagging has been crucial for quickly understanding and addressing our users' top priorities.
OpenAI
Dave Cummings
Engineering Manager, ChatGPT
Brex's mission is to help businesses move fast. Statsig is now helping our engineers move fast. It has been a game changer to automate the manual lift typical to running experiments and has helped product teams ship the right features to their users quickly.
Brex
Karandeep Anand
President
At Notion, we're continuously learning what our users value and want every team to run experiments to learn more. It’s also critical to maintain speed as a habit. Statsig's experimentation platform enables both this speed and learning for us.
Notion
Mengying Li
Data Science Manager
We evaluated Optimizely, LaunchDarkly, Split, and Eppo, but ultimately selected Statsig due to its comprehensive end-to-end integration. We wanted a complete solution rather than a partial one, including everything from the stats engine to data ingestion.
SoundCloud
Don Browning
SVP, Data & Platform Engineering
We only had so many analysts. Statsig provided the necessary tools to remove the bottleneck. I know that we are able to impact our key business metrics in a positive way with Statsig. We are definitely heading in the right direction with Statsig.
Ancestry
Partha Sarathi
Director of Engineering
We use cookies to ensure you get the best experience on our website.
Privacy Policy