/Docs

Metrics

Step 3 of the experiment builder is Metrics. This is where you select which goals you are measuring for this experiment. The metrics you choose determine how A vs B decides which variation is winning.

What metrics do

A metric is a tracked event that represents a goal. Examples: a visitor clicking a button, visiting a specific page, completing a purchase, or triggering a custom event in your code. Every time a visitor in your experiment completes a metric event, it counts as a conversion for their variation group.

A vs B compares the conversion rate of each variation group and uses statistical analysis to determine whether any difference is real or just random noise.

Selecting metrics

The Metrics step shows you all metrics that have been created for this project. Click a metric to add it to the experiment. Selected metrics appear in an ordered list below the metric picker.

You can select as many metrics as you want, but every experiment must have at least one.

The primary metric

The first metric in your list is the primary metric. This is the one that drives statistical significance for your experiment. When A vs B shows you a result like “Variant 1 is winning with 95% confidence”, it is referring to the primary metric.

The primary metric should be the main goal of your experiment — the single most important thing you are trying to improve. For a button color test, this would be “Button Click”. For a pricing page test, it might be “Plan Selected”.

To change which metric is primary, drag the metrics in the list to reorder them. The metric at the top is always the primary metric.

Secondary metrics

All metrics after the first are secondary metrics. They are tracked and reported, but they do not affect the primary statistical significance calculation. Use secondary metrics to monitor for unintended side effects.

For example, if your primary metric is “CTA Button Click”, you might add secondary metrics like “Bounce Rate” (as a guardrail metric) and “Page Scroll Depth” (to see if engagement patterns change).

Keep it focused
Resist the temptation to add every metric you have. A focused experiment with one or two clear metrics is easier to interpret and act on than an experiment tracking fifteen things at once.

Creating a new metric

If the metric you need does not exist yet, click Create a new metric at the bottom of the metric picker — this jumps you to the project Metrics page, where you can define a metric by type:

  • Click — fires when a visitor clicks an element matching a CSS selector.
  • Pageview — fires when a visitor views a specific URL.
  • Custom event — fires when you call avsb.track.event(shortId) from your code.
  • Revenue — fires when you call avsb.track.event(shortId, { revenue: 49.99 }).

For more detail on each metric type, see the Metrics section of the documentation.

Reordering metrics

The metrics list is sortable by drag-and-drop. Grab the drag handle (the six-dot icon) on the left of any metric row and drag it to a new position. The metric at the top becomes the primary metric.

Saving and continuing

Once you have at least one metric selected, click Next to proceed to Step 4 (Review & Publish). You can click Back to return to the Variations step if needed.