Stop Assignments

Stopping new user assignments in an experiment

After running an experiment for some time with users split across control and treatment groups, you may want to stop enrolling new users and analyze only the users who have been exposed so far. The Stop Assignment option lets you do this.

What Stop Assignment does

• Sets experiment allocation to 0%: no new users are enrolled in the experiment.
• Preserves existing user assignments: users already exposed continue to receive their control or treatment experience (only if Persistent Assignment is configured).
• New users aren’t checked against the experiment and instead get the project’s default experience. To ship a specific variant to all new users, make a decision on the experiment or set this in code.

Enabling Stop Assignment option

You must first enable the Stop Assignment option in Project Settings for it to appear in the Make Decision modal. To enable this option, go to Settings > Project Info and toggle on the Enable stop new user assignments for experiments setting.

How Stop Assignment works

To stop assignment for an experiment, select the Make Decision dropdown as shown below.

When you stop assignments for an experiment:

• After you stop assignment, the experiment stops enrolling new users. Statsig stops performing checks on new users.
• Previously exposed users continue receiving their consistent control or treatment experience. Configure Persistent Assignment to preserve the user variants.
• Analysis continues as the experiment records new data points for already-exposed users.

Stop Assignment is an irreversible decision. After you make this decision, you cannot resume assigning users into your experiment.