Date of slack thread: 8/9/24
Anonymous: When running experiments does StatSig bucket users only if they are a first-time visitor or do we bucket on ANY user ever? The reason I ask is that the effect of tests can be different depending on if someone is a returning user vs a new user. Essentially how are users defined?
Statbot: Statsig buckets users deterministically based on a unique identifier, such as userId
or organizationId
. This means that given the same user object and the same state of the experiment or gate, the user will always be assigned to the same bucket, regardless of whether they are a first-time visitor or a returning user. This ensures consistent and repeatable results across different sessions and devices [source]. If you need to differentiate between new and returning users, you might consider using different identifiers or custom properties to segment your users accordingly.
Statbot: I’d suggest reading through this to learn how bucketing works: https://docs.statsig.com/faq#how-does-bucketing-within-the-statsig-sdks-work. It’s a deterministic hash based on the input, and does not account for state like first-time visitor vs repeat user. If you have a flag for that, you can set it as a custom field to target separately.