Platform
Resources
Product Updates
👩🏼‍💻 Statsig CLI (”Siggy”)
Statsig CLI is our new command-line interface that helps with the creation and management of Statsig Feature Gates, Experiments, and Dynamic Configs, all from the command line. Our CLI tool can also be used within scripts, as well as integrated into CI/CD pipelines to automate the management of Feature Gates and Experiments.
To get started with installing the Statsig CLI, see our documentation here.
We’re keen to keep building out functionality for this interface, so if you have any feedback or feature requests drop us a note in Slack!
Advanced Local Metrics on Warehouse Native
Local Metrics are metrics that are scoped to an individual experiment. They let you create a custom metric you want to measure in the context of your experiment or gate, without having to clutter up your Metrics Catalog with these metrics. Local Metrics can be created from the Setup tab sections of your experiment. They will be calculated for the duration of your experiment or rollout, and then will cease to exist when you make a decision on your experiment.
Our April release allowed you to create simple metrics (sums, count, unique user count). By popular demand, we've unlocked all the flexibility available in the metrics catalog - windowing, SQL filters and metrics that span multiple metric sources.
đź”˝ Pulse Metric Lifts View Selector
Today, we’re making it easier to view your Pulse metric lifts in the way that makes the most sense for you, whether this is viewing relative (%) deltas, absolute deltas, relative or absolute topline impact. To make this easier across all metrics in your Scorecard, we’ve introduced a selector inline in Pulse to toggle how you want to view your metric lifts.
We’ve also introduced the ability to view your daily and cumulative time-series either by Deltas (today’s default) or Totals (raw units). This new view exists within the Pulse detail hovercard.
Check out these new views and let us know what you think!
ℹ️ Metric Detail Cards
At Statsig, we believe that all builders on a team should be able to create the metrics that best measure what they and their teams care most about. This is why we built a powerful Custom Metrics product which enables you to build metrics with more complex business logic on top of raw events and input metrics.
As a team’s Metrics Catalog grows, it becomes increasingly important to provide visibility to non-metric creators on how a metric is defined to ensure they can correctly interpret metric lift results.
To this end, we’re excited to start rolling out a new Metric Detail Card. Now on hover in Pulse, you can get more information about a metric’s-
Definition
Input events
Applied statistical methods
Possible dimensions to break down by
…and more!
Check it out in your Pulse Scorecard and Explore queries today!
New Suite of Javascript SDKs
We’ve rewritten our Javascript SDKs (statsig-js, statsig-react, statsig-react-native, statsig-react-native-expo, statsig-js-lite, and statsig-js-on-device-eval-client) to reduce the package sizes, codify common initialization and callback patterns, and share core logic between each package. This new SDK also supports packages for Session Replay and Web Analytics.
Read more about the migration path here, or learn about the improvements in our blog post.
Meta-analysis/Experiment Timeline View
We're rolling out the first in a series of views that enable meta-analysis across your corpus of experiments. This view lets you to filter down to experiments a team has run. At a glance you can answer questions like
What experiments are running now?
When are they expected to end?
What % of experiments ship Control vs Test?
What is the typical duration?
Do experiments run for their planned duration - or much longer or shorter?
Do experiments impact key business metrics - or only shallow or team level metrics?
How much do they impact key business metrics?
It is rolling out now to experimenters that have expressed interest in or have offered feedback on meta-analysis needs. Reach out in Slack if this is an area of interest; we're adding more over the summer and would love to hear from you! It is homed under the Insights tab in the left navigation.
SDK events are now hourly in WHN
If you choose to use the Statsig SDKs to log custom events (user actions like e.g. clicks, searches or page loads), we now batch and write them to your warehouse hourly (instead of daily). This is helpful when you've just launched an experiment and want to preview data to see if it is working (or causing crashes).
If you use your own telemetry pipeline to capture events, this has no impact on you.
Exposures
If you use the Statsig SDKs for assignment, we batch, dedupe and write exposure information to your warehouse daily.
When you load Pulse (experiment scorecard), we perform a just-in-time update of exposures for the current day so you're looking at current results. This typically captures exposures as recent as ~15m old.
Saved Queries in Metrics Explorer
We’re excited to introduce a new feature in our Product Analytics product called Saved Queries. This update improves your insight gathering and exploration workflow by providing a lightweight way to save your analysis for later use, or share them with the whole team.
What’s New with Saved Queries
Previously, to leverage previous insights you've come to in Statsig product analytics, you had to bookmark the URL or save it to a dashboard. Now, with Saved Queries, you can preserve these insights more easily. This feature simplifies how you work with data, making it more straightforward to revisit and build upon your previous analyses.
How It Works
When you land in Metrics Explorer, our main product analytics surface, you start on an “Unsaved Exploration” and can immediately dive into user and product data. If you discover valuable insights, you can save the query directly, enabling you to access and refine your findings at any future point.
Personal and Published Queries
Saved Queries come in two types:
• Personal Queries: These are private and only accessible by you. You can save any number of insights that are particularly relevant to your ongoing projects.
• Published Queries: If your saved query could benefit the entire team, you can choose to publish it. Once published, these are available to everyone in the project through the Saved Query catalog in Metrics Explorer, enabling better collaboration and knowledge sharing across your team.
We believe that Saved Queries will streamline your analysis and exploration tasks, making it easier to manage and share important insights across your team.
We will start rolling out Saved Queries today. We look forward to seeing how saving queries makes building and sharing your insights easier!
đź’» Web Analytics
Web Analytics makes it dead simple to track and watch key measures for your website.
Debuting on our suite of Javascript SDKs, Web Analytics auto-captures key user-generated events like page views, errors, page performance stats, clicks, and form submits. With each of these auto-captured events, we include key user and website page metadata, making advanced exploration easy out-of-the-box.
Once you’ve gotten up and running with Web Analytics, you can easily:
Explore your website’s engagement via Metrics Explorer
Look at your user engagement trends, via Statsig’s auto-generated suite of User Accounting metrics including DAU/ WAU/ MAU, stickiness, retention, etc.
Curate and share your own custom set of dashboards, applying custom filters and aggregations to your suite of auto-captured events
Read more in our docs & let us know if you have any feedback or questions!
🎯 Inline Targeting Criteria
Often, you may want to experiment on a specific group of people as defined by targeting criteria. To-date to accomplish this you’ve had to define a Feature Gate with your targeting criteria and then reference this Feature Gate from the experiment.
While this is useful if your targeting criteria is relatively common (and you’ll want to reuse it again in a future experiment or rollout), this can introduce extra configs and overhead if this targeting criteria is only serving the experiment in question.
We say- extra overhead and unnecessary configs cluttering up your catalog BE GONE!
Today, we’re starting to roll out the ability to define targeting criteria inline within your experiment setup. This will be accessible via the same entry point you can select an existing Feature Gate to target your experiment against (don’t worry, that capability isn’t going anywhere).
Read more in our docs here and don’t hesitate to reach out if you have any questions!
Join the #1 experimentation community
Why the best build with us
