OUTCOME CHECKWORKFLOW VERIFICATION
Evidence library / Sample 001

Inspect the finding. Follow the evidence.

These cases illustrate how Outcome Check can present discrepancies between supplied records and defined expectations. All examples are synthetic; they do not represent customer incidents, validated production findings or measured detection performance.

SYNTHETIC DATA · NO CUSTOMER RECORDS · NOT A FINANCIAL AUDIT
INCIDENT TYPES03
EXPANDED CASES01
DATA SOURCETEST
PRODUCTIONNONE
Case selector / 01—03

Choose an incident type.

Expanded case / OC-0041

One event. Two linked effects.

A synthetic event links to two successful downstream effect records when the illustrative rule expects one. The finding describes a count discrepancy in the supplied data; it does not establish why it happened or whether financial loss occurred.

Expected

One linked effect

Rule R-03 permits exactly one successful ledger.posted effect for the supplied event and object.

EXPECTED COUNT = 1
Observed

Two linked effects

The supplied synthetic input contains two matching effect records for the same event and object.

OBSERVED COUNT = 2
RECORDEVENTOBJECTEFFECTSTATUSSOURCE
eff_201evt_104pay_104ledger.postedsuccesseffects.csv · row 2
eff_202evt_104pay_104ledger.postedsuccesseffects.csv · row 3
Finding / R-03

DUPLICATE_EFFECT

Interpretation: the supplied records contain two matching successful effects for one expected posting. This violates the illustrative expected-count rule.

Review next: inspect retry and replay paths, idempotency handling, uniqueness constraints and any reversal or compensating entries. Confirm against the authoritative system before assessing impact.

This sample does not prove that money moved twice, that a customer was charged twice, or that a real system has a defect. The sample uses synthetic records and no production system was tested.
Report contract / Scope

What a finding should contain.

Evidence

Finding code, rule reference, expected and observed result, related record IDs and source-row references.

Interpretation

What the evidence supports, plausible alternative explanations, next review steps and known limitations.

A reproducible report requires the input set and rules to be fixed and the result to be checked against known clean and failure cases. This page is a sample report design, not a customer delivery.
Next / Scope

Have a workflow worth investigating?

Discuss whether a bounded evaluation fits the workflow, the available evidence and the verification question.

Discuss an evaluation ↗Read method & scope