Skip to content
Commerce Fieldbook
Fieldbook/Measurement and growth

Commerce measurement reconciliation

Match orders, analytics events, and attributed channel results at transaction level before comparing totals.

stableVersion 0.1Reviewed 2026-07-25
ClaudeAnalytics engineersFinance and growth teams

Run this reconciliation when commerce, analytics, and advertising totals disagree and the team needs to know which transactions account for the gap.

Inputs

  • Order-ledger export
  • Analytics purchase events
  • Channel reports
  • Date range, timezone, and currency rules
  • Tax, refund, cancellation, and order-state policy
  • Consent and tracking setup

Output

  • Transaction-level match table
  • Missing, duplicate, late, and transformed events
  • Notes on attribution differences
  • Reconciliation rules
  • Fixes grouped by owner

Steps

  1. Normalize the date range, timezone, currency, and included order states.
  2. Use the commerce ledger for booked orders.
  3. Match transaction IDs across systems.
  4. Distinguish collection loss from normal attribution differences.
  5. Check consent, browser, server, and checkout event coverage.
  6. Record attribution windows and modeled conversions.
  7. Assign fixes and rerun the same reconciliation.

Checks

  • Transaction IDs
  • Currency, tax, shipping, refunds, and cancellations
  • Event order and deduplication
  • Consent and browser coverage
  • Attribution windows and identity rules

Limits

Related pages

Keep working on this issue