Monitor an Experiment

Track health checks, exposures, and diagnostics for active experiments in Statsig Cloud.

Statsig runs health checks on every live experiment: SDK checks arriving, unit IDs matching, exposures balanced with no sample ratio mismatch (SRM), and no crossover units. Review them in the Experiment Health Checks banner at the top of the scorecard after launch and again before you read results. Use this page to confirm that assignment and logging are correct. To interpret metric lifts, go to How to read experiment results.

Experiment health checks

Open Experiments from the navigation.
Select the experiment you want to inspect.
Review the Experiment Health Checks banner at the top of the scorecard.

Hover over a status icon to read the summary, then click it for full context. Common health checks include:

Checks started: Verifies that the SDK is reporting config checks shortly after launch.
Checks have valid unit type: Confirms that checks include the configured unit identifier (userID by default).
Event metrics have data: Ensures that events carry the same unit ID as exposures so Pulse can compute metrics. This check often surfaces problems when downstream tooling (for example, Segment) omits stableID or custom IDs.
Pulse metrics available: Indicates that Pulse results are ready (typically the day after launch).
Exposures are balanced: Runs a chi-squared test for sample ratio mismatch (SRM). Occasional warnings happen because of randomness, but persistent red alerts point to assignment or logging issues. Refer to SRM thresholds.
Crossover units detected: Flags users exposed to multiple variants. Statsig Cloud keeps these users in both groups, because the SDK rarely produces crossovers, but highlights them so you can address root causes. Contact Statsig in Slack if you see rates above 1%.
Default value type mismatch: Warns if an experiment's fallback default value type disagrees with the parameter definition.
Group assignment healthy: Surfaces unexpected assignment reasons (for example, Uninitialized or InvalidBootstrap). Click View Assignment Reasons to see the hourly breakdown.

SRM thresholds

The Exposures are balanced check sets its status from the chi-squared p-value and the absolute deviation from the expected split.

p-value	Absolute deviation	Status
Between 0.001 and 0.01	Any	Warning (yellow)
Less than 0.01	Less than 0.1%	Warning (yellow) with low expected impact
Less than 0.001	0.1% or more	Alert (red) that requires investigation

To diagnose and resolve SRM, go to Managing SRM.

Crossover troubleshooting

Crossover warnings usually mean one of the following:

The request bootstrapped with a different stableID (BootstrapStableIDMismatch).
Both client and server SDKs are checking the same gate or experiment without synchronized updates.

If you can't pinpoint the cause, contact Statsig in Slack for help.

Exposure streams

Scroll below the health checks to view exposure streams. These tables show every recent check, including the rule that matched and any secondary exposures, such as holdouts and targeting gates. Use them to validate targeting and confirm ramp progress.

Cumulative exposures

To track growth per variant:

Open the Results tab.
Locate the Cumulative Exposures chart.

The chart shows how many users have entered each group over time, so you can identify ramp issues early.

Was this helpful?