We're still early in our journey with Statsig, but the impact is already clear. Historically, we’ve had limited visibility into how users actually interact with our app - what screens they visit, what they skip, and how engagement varies across the experience. Most of our feedback has come from a vocal few, which isn’t always representative. Statsig is starting to give us the additional data we need to make more informed and balanced decisions.

Even without a demo or diving into the documentation, the setup experience has been smooth. It was quick and intuitive to get our first dashboards and graphs running, and it’s already clear that it’ll integrate well into our existing development process.

Right now, the real-time dashboards are what we're most excited about. Looking ahead, we’re planning to make good use of the experimentation features as we continue to scale. Combining this data with qualitative user feedback and focus groups will give us a strong foundation for future product decisions. Over the next six months, we expect Statsig to significantly improve our confidence in roadmap planning and help us better understand user behaviour across platforms. Review collected by and hosted on G2.com.

At this early stage, I don’t have any dislikes to report. Our experience so far has been wholly positive, and we’ve found the platform intuitive and easy to get started with. Review collected by and hosted on G2.com.

Statsig has many automated processes that allow me to be more efficient at work. Previously, I used to download data directly from tables in Snowflake and run numerous queries to understand how metrics behaved during an experiment. I also had to perform additional processes to determine if those changes were statistically significant. Statsig has helped me save a lot of time because it does this automatically, allowing me to be more efficient in other areas of my work and dedicate more time to finding new opportunities. Love it! Review collected by and hosted on G2.com.

The reference channels are sometimes unclear to me. For example, if I want to learn about the functionality of a specific feature or something new, I sometimes don't know where to go. I know there's plenty of reference material out there, but that's not the problem; the issue is which of the portals I should go to. Review collected by and hosted on G2.com.

I've been using Statsig for almost two years, primarily for A/B testing and experimentation across multiple product teams. I appreciate that it offers a comprehensive all-in-one platform for running tests directly within the tool, avoiding the need for external analysis. The core A/B testing experience is robust with advanced statistical techniques, such as Bayesian testing and variance reduction, or Frequentist approach, allowing flexibility depending on our statistical approach. Another standout feature is the product analytics, which helps us visualize data and identify any major issues, complementing our experiments effectively. I find the ability to select statistical engines particularly valuable, along with their newer capabilities like feature flagging and potential for testing algorithms or large language models. Statsig’s setup is intuitive and supports a smooth workflow. Statsig allows me to keep the evaluation in the platform without export it to other custom analysis engine (like a notebook). Overall, I find Statsig to be a well-rounded product, worthy of a solid nine out of ten rating for recommendation. Review collected by and hosted on G2.com.

We had a challenge integrating in React Native environment, but to be fair - that is a shortcoming on our side that we had for our apps initially. As we switched to native apps the integration with SDK was very smooth. I would also always suggest a deeper technical chat with platform experts, as testing the platform integration with A/A test is very crucial to uncover hidden issues. Review collected by and hosted on G2.com.

Its UI is friendly and clear. I’ve been working with the tool for over a year now, and the experience has been very positive. I work directly in an experimentation and analytics team, and as a Statsig administrator, I’ve been involved in the entire experiment setup process. The interface for defining targeting rules is simple and intuitive, and the results section is truly powerful.

One feature I’ve recently started using and found especially useful is assignment sources. If you have an external process outside of Statsig and simply want to use the platform to analyze results, this option offers an incredible engine for standardizing metrics. It allows you to quickly understand the behavior of key metrics.

Today, I can confidently say that Statsig has become our main tool for daily analytics. The migration process was smooth in terms of implementation, and we integrated more than 28 teams across the company—always supported by the close guidance of Statsig’s technical team. Review collected by and hosted on G2.com.

There are sections I haven’t had the chance to explore yet, such as Power Analysis or Autotune. However, I wouldn’t say they aren’t useful—on the contrary, I recognize that it’s more a matter of my own unfamiliarity. I’m sure they are highly valuable and useful tools.

Regarding my experience, perhaps the only drawback I’ve identified so far is the flexibility when integrating certain fields from databases. We’re currently working on adding segments to experiments—such as user type, city, or boolean fields indicating, for example, whether a user has an active subscription. While working directly within Statsig, adding these segments is straightforward. However, if you want to bring that information into a large-scale database report, it can become more complex to integrate these fields efficiently. Review collected by and hosted on G2.com.

It's very flexible on how to define metrics - we use our own warehouse, so we basically just feed it whatever internal business logic we already had, and Statsig helps us track metrics, check for imbalances, and cut by segments as we need. I use it as a product analyst every time we experiment, and if there is some arcane custom logic we have to run on an experiment one single time, it even outputs the tables to our warehouse so we can easily poke around in SQL.

Documentation is good, and the service staff is responsive; they have helped quickly when there are bugs to fix or stuff we didn't understand. Review collected by and hosted on G2.com.

Not much, it can be a little hard to understand the underlying logic sometimes, but that's not the fault of the platform; rather that the users should be good at documenting definitions, because Statsig does allow you to write notes/descriptions where it matters.

The few times I've personally had troubles with Statsig, they've fixed them. Review collected by and hosted on G2.com.

Statsig does everything correctly. I've evaluated a dozen experimentation platforms. Their billing model makes sense, the two offerings (Cloud vs WHN) makes sense, the flexibility with treating different units as first-class citizens, the ability to really perform nuanced experiments with multiple goal metrics. Their SDKs are easy to implement, and performant too. Integrating Statsig into our organization was seamless with the Snowflake import/export functionality. Support is always prompt with a response, and they seemingly have the only (?) support bot I've seen so far that is actually helpful most of the time. Prior to Statsig, we only ran a handful experiments a year. We now run 4x-5x as many experiments. Review collected by and hosted on G2.com.

There are truly no real downsides to Statsig. It's very clear that it's a platform built by people that care about UX, analysis, and engineering. Review collected by and hosted on G2.com.

One of the biggest things I appreciate about this platform is how easy it was to get started. Across our company, three different tech-stack teams (PHP, JS, mobile) have integrated it into their workflows—and each time, it’s been a smooth process. The documentation is clear, and we’ve never hit any frustrating blockers during setup.

The tool itself is intuitive for both technical and non-technical users. Our stakeholders feel confident navigating the UI, creating experiments and interpreting the results by themselves, which frees up a lot of my time, as I am an analyst in the team. The documentation, blog articles, recorded sessions, and the Statsig University are all a great help through the whole process. They not only explain how to use the platform, but also deepen understanding of experimentation best practices in general and can be interesting for absolutely everyone involved in the product development.

The support team deserves a special mention. Every time we’ve reached out with a question or issue, we’ve heard back the same day. Having a dedicated Slack channel has honestly been a game changer, as it makes the support feel more like a partnership than a ticket queue! Review collected by and hosted on G2.com.

The main area where we’ve had issues is the Google Analytics (GA) integration. Right now, when events are forwarded from Statsig to GA, only the event_name gets sent, but any additional event parameters are lost. So for something like a purchase event, we lose all item details, which limits our ability to analyze user behavior fully outside of Statsig.

Also, the GA integration broke twice this year, and we didn’t know until we manually spotted issues and reached out to support. To their credit, both issues were fixed quickly, but it would be great to have some kind of monitoring or alerting in place to catch this earlier. Review collected by and hosted on G2.com.

I primarily use Statsig for Feature rollouts and A/B experimentations on a daily basis. The tool has an intutive UI, is easy to integrate within the existing systems and analyze the results.

Automatic stat analysis makes it seamless to understand and analyze the experiments and results even for non data scientists. I especially like it for the ease of it, ability to dive deeper into each experiment, user results and have the ability to be able to debug user connections, analyze user journeys and impact of each feature/rollout on the metrics all through one tool without the need for constant engineering involvement.

I especially appreciate the "Pulse" and "Holdout" features offering visibility into how experiments impact core business metrics over time.

Also, great customer support team which helps in resolving issues in a prompt manner. Review collected by and hosted on G2.com.

The onboarding experience could be smoother for teams not familiar with experimentation platforms. The terminology and setup flows require upfront learning and hands on experience which if improved could be a huge upgrade. Review collected by and hosted on G2.com.

Statsig makes it easy to run experiments and manage feature flags with minimal setup. The SDKs are straightforward, the dynamic configs are flexible, and the event logging is well-structured. I especially appreciate how experimentation logic can stay on the backend and remain decoupled from UI, which fits perfectly with our architecture. The ability to evaluate flags and variants in real time using console rules is also incredibly powerful. Lastly, the team is responsive and open to feedback, which makes a real difference.

Very smooth. The SDK was easy to integrate into our backend service, and initial setup of feature gates and dynamic configs was quick. We were up and running with our first experiment in a matter of hours.

We use Statsig consistently for experimentation logic, feature rollout control, and real-time configuration updates. It’s now an essential part of our development and deployment workflow.

Straightforward. Backend integration (in our case, .NET) was well-supported. The SDK offers a clean API, and the event logging + variant evaluation flow was simple to embed into our existing services. Review collected by and hosted on G2.com.

The documentation could go a bit deeper in some areas — especially around advanced configuration and production-ready patterns. We also noticed that the console UI can feel a bit clunky at times when dealing with large numbers of configs or gates. Some limitations around segment targeting and rule flexibility required us to build custom logic on top. Review collected by and hosted on G2.com.

We've been using Statsig for about a year and a half on our experimentation team, primarily for A/B testing and progressive rollouts. One of the biggest advantages is how quickly and easily we can spin up new metrics. The interface is clean and intuitive, especially the way it displays confidence intervals, which gives us a lot of clarity when interpreting experiment results. The built-in SRM (Sample Ratio Mismatch) alerting has also been incredibly helpful for catching issues early. Review collected by and hosted on G2.com.

The interface, while generally solid, can feel a bit rigid at times. We'd love to have more flexibility in setting up experiments and customising views or configurations to better match our internal processes. It's not a dealbreaker, but more adaptability would really take the platform to the next level for us. Review collected by and hosted on G2.com.