Skip to main content

Set Up the SDK

1

Install the SDK

To install the Statsig Web SDK, add the package via your preferred package manager. Include optional packages if you plan to enable Session Replay or Auto Capture.
If you don’t need Session Replay or Auto Capture, omit the @statsig/session-replay and @statsig/web-analytics packages.
After installation, configure the SDK in your app entry point before rendering your UI.
2

Initialize the SDK

Next, initialize the SDK with a client SDK key from the “API Keys” tab on the Statsig console. These keys are safe to embed in a client application.Along with the key, pass in a User Object with the attributes you’d like to target later on in a gate or experiment.
Use initializeAsync when you need to await the latest values. For a non-blocking approach, you can call initializeAsync() without awaiting and rely on cached values until the promise resolves.

Use the SDK

Checking a Feature Flag/Gate

Now that your SDK is initialized, let’s check a Feature Gate. Feature Gates can be used to create logic branches in code that can be rolled out to different users from the Statsig Console. Gates are always CLOSED or OFF (think return false;) by default.

Reading a Dynamic Config

Feature Gates can be very useful for simple on/off switches, with optional but advanced user targeting. However, if you want to be able send a different set of values (strings, numbers, and etc.) to your clients based on specific user attributes, e.g. country, Dynamic Configs can help you with that. The API is very similar to Feature Gates, but you get an entire json object you can configure on the server and you can fetch typed parameters from it. For example:

Getting a Layer/Experiment

Then we have Layers/Experiments, which you can use to run A/B/n experiments. We offer two APIs, but we recommend the use of layers to enable quicker iterations with parameter reuse.

Logging an Event

Now that you have a Feature Gate or an Experiment set up, you may want to track some custom events and see how your new features or different experiment groups affect these events. This is super easy with Statsig - simply call the Log Event API for the event, and you can additionally provide some value and/or an object of metadata to be logged together with the event:

Typed Getters

Layer, Experiment, and DynamicConfig objects support a typed get method. Using a fallback that matches the expected type helps avoid returning unintended values.
Passing a fallback of the wrong type returns that fallback. When type safety is not needed, omit the fallback to receive the raw value.

Evaluation Details

Each gate, config, experiment, and layer exposes details describing how the value was resolved.
  • reason explains the source (e.g., Network:Recognized, Cache:Unrecognized).
  • lcut is the last time any configuration changed in your project.
  • receivedAt marks when this response was received, useful for judging cache staleness.
See /sdk/debugging for the full list of reason values.

Sample Projects

Explore end-to-end examples in the js-client-monorepo samples folder for React, Next.js, precomputed clients, and more.

Parameter Stores

Parameter Stores hold a set of parameters for your mobile app. These parameters can be remapped on-the-fly from a static value to a Statsig entity (Feature Gates, Experiments, and Layers), so you can decouple your code from the configuration in Statsig. Read more about Param Stores here.

Statsig User

You need to provide a StatsigUser object to check/get your configurations. You should pass as much information as possible in order to take advantage of advanced gate and config conditions. Most of the time, the userID field is needed in order to provide a consistent experience for a given user (see logged-out experiments to understand how to correctly run experiments for logged-out users). Besides userID, we also have email, ip, userAgent, country, locale and appVersion as top-level fields on StatsigUser. In addition, you can pass any key-value pairs in an object/dictionary to the custom field and be able to create targeting based on them. Once the user logs in or has an update/changed, make sure to call updateUser with the updated userID and/or any other updated user attributes:

Updating Users

Call updateUserAsync when the signed-in user changes to fetch fresh values for that identity.
For advanced flows—such as bootstrapping or prefetching users—see Using EvaluationsDataAdapter.

Prefetching Users

Use prefetchData to prepare values for another user so you can switch synchronously later.

Statsig Options

LoggingEnabledOption
default:"browser-only"
Controls logging behavior.
  • browser-only (default): log events from browser environments.
  • disabled: never send events.
  • always: log in every environment, including non-browser contexts.
boolean
deprecated
deprecated
Use loggingEnabled: 'disabled' instead.
boolean
default:"false"
Skip generating a device-level Stable ID.
boolean
default:"false"
Recompute every evaluation instead of using the memoized result.
string
Override the generated session ID.
boolean
default:"false"
Persist Stable ID in cookies for cross-domain tracking.
boolean
Prevent any local storage writes (disables caching).
NetworkConfig
Override network endpoints per request type.
StatsigEnvironment
Set environment-wide defaults (for example { tier: 'staging' }).
LogLevel
default:"Warn"
Console verbosity.
number
default:"50"
Max events per log batch.
number
default:"10_000"
Interval between automatic flushes.
OverrideAdapter
Modify evaluations before returning them.
boolean
default:"true"
Attach the current page URL to logged events.
boolean
default:"false"
Send requests without Statsig-specific encoding.
LogEventCompressionMode
default:"Enabled"
Control compression for batched events.
boolean
deprecated
deprecated
Use logEventCompressionMode instead.
EvaluationsDataAdapter
Provide a custom data adapter to control caching/fetching.
CustomCacheKeyGenerator
Override cache key generation for stored evaluations.
string
default:"https://api.statsig.com"
Base URL for all requests (append /v1).
string
default:"https://prodregistryv2.org/v1/rgstr"
Endpoint for event uploads.
string[]
Fallback endpoints for event uploads.
number
default:"10000"
Request timeout in milliseconds.
boolean
Disable all outbound requests; combine with loggingEnabled: 'disabled' to silence log warnings.
function
Provide custom transport (e.g., Axios).
string
default:"https://featureassets.org/v1/initialize"
Endpoint for initialization requests.

Manual Exposures

Manual logging is error-prone and can often introduce issues like uneven exposures, which compromise experiment results.
You can query your gates/experiments without triggering an exposure, and manually log the exposures later:

Session Replay

Install @statsig/session-replay and register the plugin to record user sessions.

Web Analytics / Auto Capture

Install @statsig/web-analytics to automatically capture clicks, page views, and other standard web events.

Content Security Policy

Add Statsig endpoints to your CSP connect-src directive when running the web SDK.
Statsig occasionally updates its network domains. Verify the latest list in Statsig Domains.

Lifecycle & Advanced Usage

Shutting Statsig Down

In order to save users’ data and battery usage, as well as prevent logged events from being dropped, we keep event logs in client cache and flush periodically. Because of this, some events may not have been sent when your app shuts down. To make sure all logged events are properly flushed or saved locally, you should tell Statsig to shutdown when your app is closing:

StableID

Each client SDK has the notion of stableID, a devive-level identifier that is generated the first time the SDK is initialized and is stored locally for all future sessions. Unless storage is wiped (or app deleted), the stableID will not change. This allows us to run device level experiments and experiments when other user identifiable information is unavailable (Logged out users).

Stable ID

Stable ID provides a consistent device identifier. It lets you run logged-out experiments and target gates at the device level.

How Stable ID Works

  • On first initialization the SDK generates a Stable ID and stores it in localStorage under statsig.stable_id.<SDK_KEY_HASH>.
  • Subsequent sessions reuse the stored value. Each client SDK key has its own Stable ID entry.
  • Local storage is scoped per domain, so cross-domain usage requires sharing the value manually (see below).

Reading the Stable ID

Overriding the Stable ID

Provide a custom Stable ID through StatsigUser.customIDs.stableID if you already manage a durable device identifier.
When you override the Stable ID it is persisted to local storage, so subsequent sessions reuse your custom value.

Sharing Stable ID Across Subdomains

Add this helper script before initializing the SDK and then copy the stored value onto your user object.

Aligning Stable ID Between Client and Server

To share Stable ID with a backend Statsig SDK, send the value with requests and persist it server-side when missing. The server can bootstrap the client with the same Stable ID.

Using multiple instances of the SDK

Up to this point, we’ve used the SDK’s singleton. We also support creating multiples instances of the SDK - the Statsig singleton wraps a single instance of the SDK (typically called a StatsigClient) that you can instantiate.
You must use a different SDK key for each sdk instance you create for this to work. Various functionality of the Statsig client is keyed on the SDK key being used. Using the same key will lead to collisions.
All top level static methods from the singleton carry over as instance methods. To create an instance of the Statsig sdk:

Override Adapter

Use the LocalOverrideAdapter to define local overrides for gates, configs, experiments, or layers.

Persisting Overrides

Pass your client SDK key to the adapter to persist overrides between sessions when using multi-instance setups.

Using Persistent Evaluations

Persist experiment assignments so users keep the same variant even if targeting rules change.
See Client Persistent Assignment for additional patterns and storage options.

Common Targeting Use Cases

Capture cookies or URL parameters and pass them through StatsigUser.custom for targeting rules.
Targeting in Console

Async Timeouts

Limit how long initializeAsync and updateUserAsync wait for network responses before falling back to cached values.

Data Adapter

StatsigClient uses an EvaluationsDataAdapter to manage caching and network fetches. The default implementation (StatsigEvaluationsDataAdapter) reads from local storage synchronously and refreshes values from Statsig asynchronously. See Using EvaluationsDataAdapter for full examples, including bootstrapping, prefetching, and custom adapters.

Partial User Matching

Use customUserCacheKeyFunc with updateUserSync when you need to enrich a user locally without triggering a full network refresh.
Custom cache keys can produce stale or incorrect evaluations if multiple users map to the same key. Await updateUserAsync when you need guaranteed fresh values per user.

Client Event Emitter

Subscribe to Statsig client lifecycle events to respond to initialization, logging, or evaluation changes.

Quality & Troubleshooting

Testing

Mock Statsig APIs in Jest to isolate business logic.

Debugging

When results look unexpected, use these tools to inspect what the SDK is doing.

Enable Verbose Logging

Inspect the __STATSIG__ Global

Open your browser console and run __STATSIG__ to inspect the current client instance. Useful properties include _logger._queue for pending events.
Statsig Global

Review Network Traffic

Filter network requests by client- to see initialization and logging calls.
Network Logs

Check Evaluation Reasons

Common reasons:
  • Network | NetworkNotModified — latest values from the API.
  • Cache — loaded from local storage.
  • NoValues — no cached values and network failed.
  • Bootstrap — values provided via dataAdapter.setData.
  • Prefetch — values from dataAdapter.prefetchData.
See /sdk/debugging for full details.

FAQs

Does the SDK use local storage or cookies?

Statsig’s web SDK does not set cookies. It stores gate/config values and unsent events in localStorage so features keep working when offline.

Can I access the SDK instance globally?

With multiple instances, pass the SDK key: Statsig.instance('client-YOUR_KEY').
Start with logging disabled and storage blocked, then enable them after consent.
The SDK buffers up to 500 events in memory and flushes them once logging is re-enabled.

Additional Resources