⏰ Schedule Experiment Start

You no longer need to manually start your experiments. This is especially helpful when coordinating launches with other activities, like email campaigns, or when organizing experiment start times across your organization.

You can find this feature today in the "Start Experiment" dialogue.

💪 Power Analysis Enhancements

We’ve rolled out several updates to make Statsig’s Power Analysis Calculator more powerful and easier to use when planning experiments.

You can now:

• Review multiple metrics side-by-side to see how power and duration vary across key metrics in one view.

• Set a target MDE across your metrics to clearly see the optimal experiment duration across your metrics.

• Share results with saved configurations, so your team sees the same inputs and assumptions when reviewing analysis.

For Statsig Cloud users, Power Analysis now supports multiple qualifying events and event parameter filters, allowing more precise reflection of your experiment setup and traffic definitions.

With these updates, users can plan well-powered experiments and better align on timelines with your.

🔎 Warehouse Events

All events across your multi-event metric sources can now be found in one place. No more digging through different metric sources to find the event you want. Just search for the event and use it.

What You Can Do Now

• Search across all events from multi-event metric sources in a single view.

• Add events to your metrics without opening individual metric sources.

• Search by event name or by metric source name to keep related events together.

How It Works

• This unified view includes events only from metric sources that are explicitly configured as multi-event.

• A multi-event metric source contains multiple event types, for example page_view and checkout. It is not a single computed metric, and it is not a single-event source.

• To make a source multi-event, go to the metric source page and set Configure for Metrics Explorer to Multi-event.

• The unified view supports search by event name and by metric source name.

• This view is optional and does not replace existing create or edit flows. If you prefer working from Metric Sources, those flows are unchanged.

Impact on Your Analysis

• Faster event discovery when you do not remember which source an event belongs to.

• Less context switching while building metrics in Metrics Explorer.

• Better visibility across similar events from different sources in large warehouses.

Warehouse Native users can use the unified event view to quickly find, add, and work with events across your multi-event metric sources.

🚪Feature Gate Exposures in Metrics Explorer

You can now treat Feature Gate exposure events like any other event in Metric Drilldown and Funnel analyses. This capability is available on both Cloud and Warehouse-Native and currently includes first exposures for each user.

What You Can Do Now

• Use Feature Gate exposure events in Drilldown charts to track exposure trends over time

• Add a gate exposure as the first step of a funnel to measure post-exposure conversion paths

• Group or filter by gate-related properties, such as gate name, pass/fail result, or environment

• Compare user behavior across gate conditions to understand the impact of gated rollouts

How It Works

Exposure logging When a user is evaluated against a gate, a first-exposure event is recorded with relevant gate properties and user context.

Event selection In both Drilldown and Funnel charts, gate exposures appear in the same event picker you already use for other events.

Property handling Exposure events include gate metadata and user properties, enabling the same group-by and filter controls available for other event types.

Impact on Your Analysis

Drilldown

• Validate rollout health by visualizing exposure volume and distribution over time

• Debug exposure logging by spotting spikes, drops, or unexpected gaps

• Align exposure activity with key metrics to confirm rollout timing and behavior

Funnels

• Measure user journeys starting from a user’s first gate evaluation

• Identify conversion differences between users who passed or failed a gate

• Attribute downstream changes to specific gated experiences

This brings Feature Gate exposure analysis into Metrics Explorer, helping you debug, validate, and measure the real-world effects of gated rollouts across both Cloud and Warehouse-Native environments.

👨‍⚕️ Docs v2

Following 4.5 years in our previous codebase and infrastructure, we're excited to announce the second iteration of the Statsig Documentation!

New UI/UX

Docs v2 comes with a vastly updated UI, including revamped codeblocks, tabs and dropdowns, and tidied-up navigation and page structure.

New Features

Docs v2 features some new features that make it easy to pull Statsig's docs into your LLMs' context, get answers with AI, and more. We also have a brand new API playground for our Console API and HTTP API, making it easy to grab code snippets in your language of choice or get a sample cURL.

As always, our docs remain open source and free for the community to contribute to. We make our docs better every day, if you find anything that needs some love, let us know in Slack!

📈 Autocapture Updates

Statsig Autocapture allows you to track events on your website such as page views, clicks, scroll depth, etc. with just one line of code. Today, we are excited to announce 4 major updates that will help you measure and provide more context around user behavior and site performance.

🔗 To start using Autocapture, refer to the setup guide in our docs:

1. Expanded Autocapture Events

Autocapture now tracks additional user actions, including:

• Form changes

• Clicks across more element types

• Text input

• Rage clicks (repeated clicks in frustration)

• Dead clicks (clicks that lead to no action or broken links)

• Copy/paste actions

2. UTM Parameters

UTM tags are now automatically captured, allowing teams to understand where traffic is coming from. Captured parameters include source, campaign, medium, content, and term.

3. Web Vitals

You can now track key performance metrics directly through Autocapture:

• Cumulative Layout Shift (CLS)

• First Contentful Paint (FCP)

• Last Contentful Paint (LCP)

• Time to First Byte (TTFB)

4. Web Analytics Dashboard

A refreshed Web Analytics Dashboard is now generated when you set up autocapture. It provides visibility into traffic trends, channel breakdowns, visitor demographics, and performance metrics. Each chart is customizable for filtering and segmentation.

📝 Statsig Visual Editor

Following two previous iterations under the "Sidecar" title, Statsig's third-generation visual editor is here, simply called the Statsig Visual Editor. The Visual Editor, like Sidecar before it, lets you start experiments without writing code but still use Statsig's powerful Stats Engine to get results faster. The Visual Editor experience is centered around the Statsig console (instead of in a Chrome Extension), meaning your Visual Editor experiments sit alongside your product experiments in the console. Alongside the in-console experience, the Statsig Visual Editor is designed to be vastly simpler to use vs. previous iterations, and have fewer hiccups going from idea-to-experiment.

The Editor is in an open beta, you can get started by reading the docs or choosing the "Visual Editor" experiment type when creating an experiment in the console!

💰 Rich Text Widgets on Dashboards

Overview

You can now add rich text widgets to any dashboard. This is a new option in addition to existing header text widgets, which remain for simple dividers.

What You Can Do Now

• Write context directly on dashboards with formatted text

• Use headings, bold, italics, lists, and links

• Format with Markdown or the widget’s built-in controls

How It Works

1. On a dashboard, add a widget and choose Rich Text.

2. Enter your content.

3. Format using Markdown or the toolbar, then save.

Impact on Your Analysis

• Make dashboards self-explanatory with metric definitions, scope, and caveats next to the charts

• Reduce back-and-forth by capturing conclusions, decisions, and next steps inline after reviews

• Speed up onboarding by explaining how to read the dashboard and why certain cuts or filters are used

• Link out to specs, tickets, or experiments so readers can get more context without leaving the page

Available now on all dashboards. Try adding a Rich Text widget to provide context where it’s most useful.

📈 Change Alerts

You can now set Change Alerts to track relative shifts in your metrics. Instead of relying on fixed thresholds, these alerts notify you when a metric moves up or down by the percentage or amount you choose.

What You Can Do Now

• Create alerts that trigger on % increases or decreases

• Catch major swings like a 20% drop in signups or a 50% jump in errors

• Use Change Alerts with Threshold Alerts to cover both relative and topline changes

Getting Started

1. In the left product menu, open Topline Alerts.

2. Create a new alert and choose your desired Condition Type.

When to Use Each Alert Type

• Threshold - use to monitor against a fixed limit. ("Alert me when total daily signups drops below 1000")

• Change - use to monitor absolute shifts. ("Alert me when daily signups drop by 200 compared to yesterday")

• Change (%) - use to monitor percentage shifts ("Alert me when daily signups drop 20% compared to yesterday")

🧪 Exposure Imbalance / Sample Ratio Mismatch (SRM) Debugging in Metrics Explorer

What is SRM?

Sample Ratio Mismatch (SRM) happens when the share of users in experiment groups is different from what you expected. For example, if you set up a 50/50 split between control and treatment but the actual traffic is 60/40, that’s an SRM.

What is the SRM p-value?

The SRM p-value is a statistical measure that tells you whether the observed imbalance could have happened by chance.

• A p-value above 0.01 generally means the imbalance is within expected random variation.

• A p-value below 0.01 suggests the imbalance is unlikely due to chance and may warrant investigation.

What You Can Do Now

• View SRM results and p-values across experiment groups in Metrics Explorer

• Group results by different properties to identify potential causes of imbalance

• Start from experiment exposure diagnostics and click on suggested properties to pre-apply them as group-bys in Metrics Explorer

How It Works

Metrics Explorer applies the SRM formula across experiment groups and shows the resulting p-value. From there, you can add group-bys (such as country, platform, or custom properties) to spot where imbalance is happening.

Experiment diagnostics also highlight properties that may be driving the imbalance. Clicking the icon next to one of these properties takes you into Metrics Explorer with that property already grouped, so you can continue the investigation seamlessly.

Impact on Your Analysis

This workflow makes it faster to detect and understand exposure imbalances. By moving directly from diagnostics to group-by analysis, you save time and get clearer visibility into which properties are linked to the imbalance.

Sample Ratio Mismatch debugging is available now across Cloud and Warehouse Native.