What is the Beneish M-Score?

The Beneish M-Score is a mathematical model using 8 financial ratios to detect earnings manipulation. An M-Score above -1.78 suggests the company is likely a manipulator. The 8 components: Days Sales in Receivables Index (DSRI), Gross Margin Index (GMI), Asset Quality Index (AQI), Sales Growth Index (SGI), Depreciation Index (DEPI), Selling, General & Administrative Expenses Index (SGAI), Leverage Index (LVGI), and Total Accruals to Total Assets (TATA). Beneish identified Enron as a likely manipulator before its collapse.

What is Benford's Law in accounting?

Benford's Law states that in naturally occurring datasets (like financial transactions), the leading digit is 1 about 30% of the time, 2 about 17.6%, and so on, with 9 occurring as the first digit only about 4.6% of the time. Auditors use Benford's Law analysis to flag journal entries, expense reports, or revenue figures where the distribution of first digits significantly deviates from the expected pattern — a potential indicator of fabricated or manipulated numbers.

AI for Financial Statement Fraud Detection: Beneish M-Score, Altman Z-Score, and Red Flags with Claude (2026)

How forensic accountants and analysts use Claude AI to detect financial statement manipulation: Beneish M-Score computation, Benford's Law journal entry testing, Altman Z-Score distress analysis, and revenue recognition red flags.

Financial Statement Fraud Detection and AI

Financial statement fraud costs investors billions annually. The ACFE estimates the median fraud scheme runs for 12 months before detection. Claude with ClaudeFinLab enables forensic accountants, auditors, and equity analysts to systematically scan for manipulation indicators: quantitative models like the Beneish M-Score and Altman Z-Score, statistical anomalies via Benford's Law, and qualitative red flags in accounting policy changes and revenue recognition practices.

Beneish M-Score Computation

"Compute the Beneish M-Score for this company using 2-year financial data: Year 1: Revenue $185M, AR $32M, COGS $112M, PP&E net $45M, Depreciation $8M, SGA $28M, Total Assets $180M, Total Debt $85M, Net Income $18M. Year 2: Revenue $210M, AR $48M, COGS $128M, PP&E net $52M, Depreciation $9M, SGA $35M, Total Assets $200M, Total Debt $95M, Net Income $22M. Compute all 8 M-Score components (DSRI, GMI, AQI, SGI, DEPI, SGAI, LVGI, TATA) and the final M-Score. Interpret: is this company a likely manipulator (M-Score > -1.78)?"
"Explain the highest-risk M-Score component for this company: DSRI = 1.42 (AR growing much faster than revenue — suggests premature revenue recognition or channel stuffing). GMI = 0.89 (gross margin compression — competitive pressure or cost overruns masked by revenue manipulation). AQI = 1.18 (asset quality deteriorating — higher off-balance-sheet assets or capitalized expenses). Which component raises the most concern and what specific financial statement items should an auditor investigate?"

Revenue Recognition Red Flags

"Analyze revenue recognition red flags in this SaaS company's financials: (1) AR days increased from 52 to 78 days over 2 years while revenue grew 35% — suggests billing without collection; (2) deferred revenue grew only 12% while ARR claimed to grow 28% — inconsistency between cash ARR and deferred revenue; (3) Q4 revenue is consistently 42% of annual revenue (seasonality or channel stuffing?); (4) unbilled receivables line appeared in Year 3 ($8.2M) — what does this indicate? Draft a framework for investigating each red flag."
"Analyze the revenue recognition policy change: the company switched from percentage-of-completion to completed contract method for long-term contracts in Year 3. Under old method, Year 3 revenue was $185M. Under new method: $195M (increase of $10M). What is the impact on earnings quality? Is this change disclosed? What additional disclosure is required under ASC 606? Does the change reverse in future periods or is it a one-time acceleration?"

Benford's Law Journal Entry Testing

"Analyze this journal entry dataset (5,000 journal entries) for Benford's Law compliance: [paste digit distribution]. Expected first-digit distribution: 1=30.1%, 2=17.6%, 3=12.5%, 4=9.7%, 5=7.9%, 6=6.7%, 7=5.8%, 8=5.1%, 9=4.6%. Observed: digit 7 appears as first digit 18.3% (vs expected 5.8% — 3.2 standard deviations above expected). Flag the anomaly. Filtered analysis: pull all journal entries beginning with 7 — are they clustered around specific accounts, users, or time periods? This is the standard ACFE/PCAOB journal entry testing procedure."

Altman Z-Score and Distress Indicators

"Compute the Altman Z-Score (public company) for this manufacturer: Working Capital $18M, Retained Earnings $42M, EBIT $12M, Market Cap $85M, Total Liabilities $120M, Total Assets $180M, Revenue $210M. Z-Score formula: 1.2(WC/TA) + 1.4(RE/TA) + 3.3(EBIT/TA) + 0.6(MktCap/TL) + 1.0(Rev/TA). Compute Z-Score. Interpret: Z > 2.99 = safe zone; 1.81-2.99 = grey zone; Z < 1.81 = distress zone. If in the grey zone, which driver is most concerning? What would the Z-Score be if market cap fell 40% (stress scenario)?"

Where to Start

Run the Beneish M-Score first — it requires only 2 years of financial data and flags the companies most likely to be manipulating. For companies with M-Score > -1.78, follow up with AR aging analysis and revenue recognition policy review. For all companies in a screening context, Benford's Law testing on the journal entry population is the fastest way to flag anomalous entries for auditor follow-up.

Get your free API key