Analytics without the fog.

An observation or pattern that differs materially from an expected baseline.

A rule for assigning credit for an outcome across observed touchpoints.

The comparison state used to interpret a change.

A group sharing a starting event or characteristic.

A range produced by a procedure that quantifies estimation uncertainty.

The share of eligible entities completing a defined action.

The trace from a metric back through transformations to its original source.

The eligible population used beneath a rate or ratio.

A field used to segment a metric, such as source, product, or region.

An estimate of future outcomes conditional on data and assumptions.

A metric that must not deteriorate while optimizing another outcome.

The outcome caused by an intervention beyond what would otherwise occur.

A metric selected to track progress toward a specific objective.

A measure expected to move before the final business outcome.

The level at which each metric record is uniquely defined.

A durable measure of customer value creation for product direction.

An extreme observation that may be real, exceptional, or erroneous.

A probability calculated under a null model, not the probability a hypothesis is true.

The share of a cohort remaining active or valuable over time.

A conditional model that explores a chosen set of assumptions.

A meaningful subset used to compare behaviour or performance.

The probability that a test detects an effect of a specified size when it exists.

The average squared distance of observations from their mean.

Won opportunities divided by opportunities eligible for a decision.