Using Fraud Examination Methods to Investigate Allegations and Corroborate Evidence

How forensic practitioners plan allegation-driven fraud examinations, preserve evidence, perform targeted testing, conduct interviews, and report supported findings.

Fraud examination is different from a standard financial statement audit. A financial statement audit is designed to obtain reasonable assurance about the financial statements as a whole. A fraud examination is usually allegation-driven and is designed to determine what happened, how it happened, who was involved, what evidence supports the finding, and what loss or exposure resulted.

For AUD, the important point is that suspicion does not equal proof. Fraud examination methods identify leads, preserve evidence, corroborate facts, and support defensible conclusions.

    flowchart TD
	    A["Allegation, red flag, or irregularity"] --> B["Define objective and scope"]
	    B --> C["Preserve records and chain of custody"]
	    C --> D["Develop fraud hypothesis"]
	    D --> E["Perform targeted testing and analytics"]
	    E --> F["Interview witnesses and relevant parties"]
	    F --> G["Corroborate or refute findings"]
	    G --> H["Report methods, evidence, limitations, and conclusions"]

Fraud Examination Lifecycle

Phase Purpose Common evidence
Intake and triage Understand the allegation, source, urgency, and potential exposure. Hotline reports, complaints, management concerns, or audit findings.
Planning Define scope, roles, legal involvement, procedures, and evidence preservation. Engagement letter, legal instructions, investigation plan, and document hold notices.
Evidence preservation Protect records from alteration, deletion, or loss. Chain-of-custody logs, forensic images, document inventories, and access restrictions.
Targeted testing Test transactions or records most likely to show the scheme. Vendor files, journal entries, payroll records, invoices, bank records, and system logs.
Interviews Gather explanations, timelines, admissions, and corroborating facts. Interview notes, recordings when permitted, signed statements, and follow-up evidence.
Reporting Communicate supported findings, limitations, and quantified effects. Investigation report, exhibits, schedules, and recommendations.

The scope should be specific. A request to “look for fraud” is too vague. A stronger objective identifies the suspected scheme, affected period, relevant systems, and intended use of the work.

Targeted Testing

Fraud testing is designed around a hypothesis. The hypothesis may change as evidence develops.

Suspected scheme Targeted procedure
Fictitious vendor Compare vendor addresses, tax IDs, bank accounts, phone numbers, and ownership information against employee and public records.
Duplicate payment Search for repeated invoice numbers, amounts, dates, purchase orders, or bank accounts.
Payroll ghost employee Match payroll records to HR files, active employee listings, badges, direct-deposit accounts, and manager approvals.
Revenue cut-off manipulation Inspect shipping, delivery, invoice, and cash receipt evidence around period end.
Management override Analyze manual journal entries, late postings, round-dollar entries, and entries by privileged users.
Expense reimbursement fraud Compare expense claims to receipts, calendars, travel records, card statements, and policy limits.

Targeted tests produce leads. Each lead should be corroborated with source documents, external evidence, system logs, interviews, or other independent support.

Interviews

Fraud interviews are usually staged. The investigator starts with neutral witnesses and background information before interviewing people more directly connected to the allegation.

Interview practice Why it matters
Use open-ended questions Encourages narrative responses and reveals process details.
Avoid premature accusations Reduces defensiveness and protects the investigation.
Ask follow-up questions Tests consistency and fills gaps.
Corroborate statements Interview comments are stronger when supported by records.
Document carefully Notes should identify date, attendees, topics, and key statements.
Coordinate with counsel when needed Legal privilege, employment law, and regulatory issues may affect the process.

Nonverbal cues can be useful for planning follow-up, but they are not proof of fraud. The investigator should rely on corroborated evidence rather than body-language speculation.

Evidence Handling

Forensic work often supports discipline, insurance claims, civil litigation, criminal referral, or regulatory reporting. Evidence handling must be defensible.

Evidence issue Good practice
Chain of custody Track who collected, held, transferred, and analyzed evidence.
Digital evidence Preserve original media or forensic images and work from copies.
Emails and documents Maintain metadata when relevant and avoid uncontrolled forwarding.
System logs Capture logs before retention periods expire.
External confirmations Use independent sources when internal records may be compromised.
Workpapers Separate facts, assumptions, analysis, and conclusions.

The exam trap is treating ordinary audit documentation as enough for legal use. Forensic evidence may require more precise preservation and traceability.

Reporting Findings

A fraud examination report should state what procedures were performed, what evidence was obtained, what limitations existed, and what conclusions are supported. It should avoid unsupported legal conclusions.

Reporting item What to include
Objective and scope Allegation, period, entities, systems, and intended users.
Procedures Documents reviewed, data tests performed, interviews conducted, and external corroboration obtained.
Findings Facts supported by evidence, not speculation.
Quantification Loss estimate, unsupported amounts, or exposure range when supportable.
Limitations Missing records, scope limits, unavailable witnesses, or unresolved issues.
Recommendations Control improvements or remediation steps, when within scope.

The investigator should distinguish a supported finding from a suspicion, anomaly, or unresolved question.

Exam Traps

  • Fraud examination is allegation-driven; it is not the same as a normal audit opinion engagement.
  • Red flags and anomalies are leads, not proof.
  • Interview evidence should be corroborated with documents, logs, confirmations, or other records.
  • Nonverbal cues are not enough to conclude fraud occurred.
  • Chain of custody matters when evidence may be used in legal or disciplinary proceedings.
  • Legal counsel may be involved to preserve privilege and guide reporting obligations.
  • The report should avoid unsupported legal conclusions such as declaring criminal guilt.

Quick Review

Use this sequence for fraud examination questions:

  1. Define the allegation and engagement objective.
  2. Preserve evidence and establish chain of custody.
  3. Develop a fraud hypothesis.
  4. Perform targeted tests tied to the suspected scheme.
  5. Conduct interviews in a planned sequence.
  6. Corroborate anomalies before concluding.
  7. Report facts, methods, limitations, and supported conclusions.

Review Questions

### What best distinguishes a fraud examination from a standard financial statement audit? - [ ] A fraud examination always provides an audit opinion. - [x] A fraud examination is usually allegation-driven and focused on investigating specific suspected misconduct. - [ ] A fraud examination uses no documentation. - [ ] A fraud examination ignores evidence reliability. > **Explanation:** Fraud examinations are directed toward specific allegations, red flags, or suspected schemes rather than an opinion on the financial statements as a whole. ### What should occur early in a forensic investigation when records may be altered or destroyed? - [ ] Wait until the final report is drafted. - [x] Preserve records and establish chain of custody. - [ ] Delete duplicate records. - [ ] Interview the suspected person with no preparation. > **Explanation:** Evidence preservation protects the integrity and traceability of records. ### Why are targeted fraud tests designed around a hypothesis? - [ ] To avoid collecting evidence. - [x] To focus procedures on the suspected scheme and likely evidence trail. - [ ] To guarantee a fraud finding. - [ ] To replace all interviews. > **Explanation:** A hypothesis helps the investigator select procedures that can corroborate or refute the suspected scheme. ### Verifying vendor addresses and comparing them with employee records is primarily used to detect what? - [ ] Inventory obsolescence. - [x] Fictitious vendors, shell companies, or conflicts of interest. - [ ] Pension discount rates. - [ ] Bond amortization errors. > **Explanation:** Shared addresses, bank accounts, or ownership links can indicate vendor fraud or undisclosed relationships. ### What is the best use of nonverbal cues during an interview? - [ ] Treat them as conclusive proof of fraud. - [x] Use them as prompts for follow-up questions while relying on corroborated evidence. - [ ] Ignore all verbal answers. - [ ] Replace document inspection. > **Explanation:** Nonverbal cues may guide questioning, but conclusions require evidence. ### Which interview approach is generally preferred at the start of an investigative interview? - [ ] Immediate accusation. - [x] Open-ended, non-accusatory questions. - [ ] Refusal to take notes. - [ ] Asking only yes-or-no questions. > **Explanation:** Open-ended questions encourage fuller explanations and reduce premature defensiveness. ### What does chain of custody document? - [ ] The expected tax rate. - [x] Who collected, held, transferred, and analyzed evidence. - [ ] The audit fee schedule. - [ ] The client's marketing plan. > **Explanation:** Chain of custody supports the integrity and admissibility of evidence. ### What should a fraud examination report avoid? - [ ] Describing procedures performed. - [ ] Explaining limitations. - [x] Stating unsupported legal conclusions. - [ ] Quantifying supported losses. > **Explanation:** The report should distinguish supported findings from legal conclusions that may belong to courts or counsel. ### Why should anomalies from data tests be corroborated? - [ ] Analytics are never useful. - [x] An anomaly is a lead and may have a legitimate explanation. - [ ] Corroboration is prohibited. - [ ] Every anomaly proves fraud automatically. > **Explanation:** Data anomalies require follow-up before the investigator concludes fraud occurred. ### What is a common outcome of the reporting and recovery phase? - [ ] Ignoring control weaknesses. - [x] Communicating supported findings, loss estimates, limitations, and remediation recommendations. - [ ] Beginning the investigation for the first time. - [ ] Deleting the workpapers. > **Explanation:** The final phase communicates findings and may support recovery or corrective action.
Revised on Monday, June 15, 2026