Skip to main content

January 2026

Overview​

This release deepens our investment in metrics governance while also bringing key performance improvements to the experiment experience. With this update, managing, discovering, and using metrics becomes faster, more transparent, and better integrated into your workflow.

Metrics​

Metric Catalog in Main Menu​

The metric catalog is now accessible directly from the main menu, reflecting the central role metrics play across the platform.

We've upgraded the metric list to a full catalog with enhanced search and filtering, matching the experience already available in experiment creation.
This makes it easier for:

  • Metric owners to manage and maintain their metrics
  • Experimenters to find the right metrics for their experiments

Variance reduction with CUPED​

CUPED (Controlled-experiment Using Pre-Experiment Data) is a well-known variance reduction technique that makes metrics more sensitive by leveraging pre-experiment data about users. It allows you to detect smaller effects with the same sample size, or reach statistical significance faster with fewer users.

In ABsmartly you can now :

  • Easily enable CUPED for new metrics, or by creating a new version of an existing metric
  • Choose a lookback period between 1 and 4 weeks
  • Enjoy a shorter time to decision

Metric Duplication​

You can now duplicate metrics with a single action, no need to re-enter definitions by hand. Ideal when creating a variation of an existing metric.

Draft & Active Status​

Manage your metrics' lifecycle more clearly:

  • All new metrics are created in draft by default
  • Draft can be edited without restrictions
  • Draft metrics cannot be added to experiments
  • Once a metric is made active it becomes targetable and can be used in experiments
  • All existing metrics will be made active by default so you can keep using all your existing metrics

This is the first step toward the upcoming metric approval workflow, which will allow greater control over metric governance in the next release.

Experiments​

Faster Experiment Overview​

We've added caching to improve the performance of the experiment overview page, especially for experiments with large datasets.

Data Freshness Indicators​

Each experiment now includes:

  • A data freshness indicator to help you understand how recent the data is
  • A force refresh button so you can manually update results when needed

Graph Improvements​

We've improved overall graph responsiveness and rendering speed, giving you a smoother experience when navigating and interpreting results. We've also improved the histogram graph so now the buckets match making comparison between variants much easier.

Questions or Feedback?​

We're always happy to help, so reach out if you have any questions or want to explore how to make the most of these new capabilities.