Coverage Report — daily_revenue

Generated from Scope x Basis matrix on 2026-06-29. Platform: dbt.

Tests generated

#	Scope	Basis	Check	Severity	Output
1	row-level	absolute	amount must be >= 0	error	schema.yml (dbt_expectations.expect_column_values_to_be_between)
2	aggregate-level	relative	sum(amount) within 2 std-dev of trailing-30-day mean	error	tests/daily_revenue_amount_sum_within_2_stddev.sql
3	aggregate-level	temporal	daily row count within 20% of prior day	warn	tests/daily_revenue_row_count_within_20pct_of_prior_day.sql
—	row-level	human	spot-check 5 rows against the billing UI	skip (manual)	noted in schema.yml, not automated

Row-level:            ✓ (1 test)        [+1 human check, manual]
Aggregate-level:      ✓ (2 tests)
Transformation-level: ✗ (0 tests — GAP)

Absolute:  ✓ (1 test)
Relative:  ✓ (1 test)
Temporal:  ✓ (1 test)
Human:     — (1 noted, not automatable)

Transformation-level scope: 0 automated tests. Every transformation-level cell in the matrix is empty. Per the framework definition-of-done (at least one check per scope level), this scope is uncovered. Recommend adding a source-fidelity check (e.g. sum(amount) downstream reconciles to the upstream billing source).

Tests 2 and 3 assume a date column revenue_date on the model. Adjust if the date grain column is named differently.
Test 3 (temporal) is set to warn severity per the basis mapping (temporal -> warn). It uses a window function over the model's own history; no snapshot table required because daily_revenue retains historical days. If the model is truncated/replaced each run, point the query at a daily_revenue_snapshot instead.
The human sanity check is a manual reminder only — no test code emitted.