# Measuring performance, attribution and growth

Availability: Planned API.
Last reviewed: 2026-08-28.

Measure what actually happened and understand how the measurement was produced. The planned output is a scoped performance view with defined events, attribution and experiment evidence.

## Define the metric

Choose the business/Store/campaign scope, period, event definition, denominator and attribution model. Distinguish page views, QR scans, media engagement, leads, orders and settled revenue.

A modeled opportunity range belongs in planning—not in a realized performance total.

## Retain evidence and freshness

Keep the source events or permitted aggregates, measurement time, coverage and any exclusions. A missing feed is unavailable, not zero conversions. Visitor fingerprints, precise locations and detailed participant histories require narrower privacy treatment than aggregate metrics.

For example, a campaign can have observed scans without a verified sale attribution.

## Compare experiments responsibly

An experiment definition or generated variant does not prove traffic allocation or a live test. Report the actual exposure, sample and metric before interpreting differences. Maintain the revision and conditions used for each comparison.

## Availability

Business APIs are planned. Store analytics, media engagement, placement statistics and ABN scaffolds need consistent retrieval and metric contracts. They do not establish ad-buying availability, universal attribution accuracy or a guaranteed growth result. No hidden tracking or new customer-data collection is authorized by reading this guide.