Statsig

Feature flags, A/B tests, and product analytics in one platform

Freemium Some setup needed Web · API

workflow research #feature-flags#experimentation#product-analytics

About

Add a feature flag, start an experiment, and read the impact in the same dashboard. Product, engineering, and data teams use it to manage rollouts, run A/B tests, and drill into product and web analytics with session replay. Warehouse-native deployment and 30+ SDKs set it apart from single-point tools.

Editor's Take

Worth trying if product and data teams need flags, experiments, and analytics in one place — especially when you want warehouse-native results and session replay. Expect some setup learning and watch usage-based pricing as experiments and event volume grow.

Key Features

Instrument your app with a supported client/server library → start sending events in minutes (30+ SDKs)
Choose from 30+ SDKs → roll out feature flags consistently across web, mobile, and backend services
Flip on an experiment from a flag → get statistical results and a declared winner in the analytics view
Open a spike in metrics → jump into session replay to see the exact user journey
Connect your data warehouse → analyze experiments and product metrics without moving data (warehouse-native)

Use Cases

A product manager gating a new checkout flow behind a flag and measuring conversion lift via an A/B test
A data scientist running holdout experiments across multiple features and reviewing results alongside product analytics
A mobile lead rolling out a risky feature to 1% of users across iOS and Android SDKs, then ramping based on metrics

Try It Like This

1
Gate a new checkout flow
PM creates a feature flag in the dashboard → instrument the web checkout with the client SDK and target 10% of users → start an A/B experiment from the flag and watch conversion metrics in the analytics view.
2
Run a cross-platform holdout test
Data scientist configures the same experiment across iOS, Android, and backend SDKs → define metric events and run the test for a defined cohort → inspect statistical results and declare a winner without moving data from the warehouse.
3
Safe gradual rollout for mobile
Mobile lead creates a flag and targets 1% of users on both iOS and Android → monitor crash, engagement, and conversion metrics in the experiment panel → ramp to larger percentages when preconfigured thresholds are met.
4
Investigate metric spike with session replay
Product analyst opens an experiment that shows a sudden drop in a metric → use the analytics view to open session replay for affected users → follow the user journeys to identify UI bugs or confusing flows.
5
Warehouse-backed metric validation
Engineer connects Statsig to the company data warehouse (warehouse-native) → run the same experiment analytics against warehouse tables to validate dashboard numbers → export or join results with other warehouse data for deeper analysis.

Pros & Cons

Pros

30+ official SDKs for consistent flagging across web, mobile, and backend services — reduces fragmentation when rolling out features.
Warehouse-native analytics lets teams analyze experiments directly against their data warehouse without ETL or separate exports.
Integrated flow from feature flag → experiment → analytics (with session replay) so teams can run a test and inspect user journeys in one dashboard.

Cons

Dashboard and feature set have a mild learning curve; initial configuration and deeper features can be unclear to new users.

Getting Started

1 Create an account at statsig.com
2 Install one of the 30+ SDKs and send your first event from your app
3 Create a feature flag and watch exposures and key metrics populate in the dashboard