Financial statements are summaries. They aggregate millions of transactions into line items: revenue, COGS, operating expenses, depreciation, interest, tax, profit. The aggregation is necessary for reporting. It is dangerous for diligence, because it conceals the composition, timing, counterparties and patterns that determine whether the summary is trustworthy.
The ERP contains what the financial statements aggregate away. Every sales invoice. Every purchase order. Every journal entry. Every inventory movement. Every receivable posting. Every payment. Every user who processed the transaction and every timestamp. The ERP is the evidentiary substrate from which the financial statements were constructed. And the substrate, when examined directly, frequently tells a different story.
What the Financial Statements Say vs What the ERP Shows
Revenue: the headline vs the composition
The financial statements report: Rs 500 crore of revenue. Growth of 22%.
The ERP reveals: Rs 180 crore from the top three customers (36% concentration undisclosed because each is below the 10% individual disclosure threshold). Rs 45 crore recognised in the last five business days of each quarter (9% of total, suggesting period-end loading). Rs 12 crore of credit notes issued in Q1 reversing Q4 revenue (suggesting Q4 pull-forward). Rs 8 crore from two customers whose first transaction occurred in the same quarter as the fiscal year-end (new counterparties appearing precisely when targets are due).
The revenue figure is correct. The ERP reveals what the summary conceals: concentration, timing manipulation indicators and counterparty quality.
The transaction impact: if the top three customers are lost (36% concentration) or the period-end revenue is subsequently reversed (9% risk), the sustainable revenue base may be Rs 275 crore to Rs 440 crore, not Rs 500 crore. At 10x EBITDA, the concentration and timing risks represent Rs 90 crore to Rs 340 crore of enterprise value that depends on assumptions the financial statements do not test.
Customer profitability: the blended margin vs the real economics
The P&L reports 15% EBITDA margin. The business appears healthy.
The ERP, when analysed at the customer level (allocating direct costs, cost-to-serve, returns, credit notes and working-capital consumption by customer), reveals margins ranging from +35% for the top 20 customers to -12% for the bottom 50. The company is cross-subsidising unprofitable customers with profitable ones.
The transaction impact: if the buyer’s integration plan changes the service model (which frequently happens post-close), the unprofitable customers may leave without affecting the margin, but the profitable customers may also leave if the service changes affect them. The customer-level economics determine the integration risk. The blended margin conceals it.
Procurement: the cost line vs the vendor reality
The P&L reports Rs 300 crore of purchases and 40% gross margin.
The ERP reveals: 60% of procurement from three vendors (supply-chain concentration). Purchase orders systematically split into amounts just below the Rs 5 lakh competitive-bidding threshold (200 transactions at Rs 4.7 to Rs 4.9 lakh from the same vendor, suggesting deliberate avoidance of the control). Vendors incorporated within months of their first transaction. Vendors sharing registered addresses with the promoter’s family residences.
The transaction impact: vendor concentration creates operational risk. Threshold-splitting suggests control override. Connected vendors suggest potential related-party transactions that are not disclosed. Each finding reduces the reliability of the reported cost base and warrants investigation before the transaction price is finalised.
Inventory: the balance vs the saleability
The balance sheet reports Rs 80 crore of inventory at cost. Turns of 4.5x.
The ERP shows: Rs 25 crore that has not moved in more than 180 days. Rs 8 crore with production dates more than 12 months old. Rs 3 crore of items with zero sales in six months. The provision for slow-moving inventory is Rs 2 crore. The actual exposure, based on historical recovery rates for aged stock, may be Rs 12 crore to Rs 15 crore.
The transaction impact: the Rs 10 crore to Rs 13 crore provision shortfall reduces the target’s net assets and, if flowing through COGS, reduces sustainable EBITDA. At 10x, that is Rs 100 crore to Rs 130 crore of enterprise value attributable to inventory the market may not absorb.
Journal entries: the result vs the mechanism
The financial statements report profit. The ERP shows the journal entries that produced it.
A ten-filter analytics run identifies: 150 manual entries posted on weekends. 80 entries affecting revenue accounts in the last three days of each period. 45 entries posted directly by the CFO. 30 entries with vague descriptions (“adjustment,” “correction”). 12 entries posted and reversed within the same week.
None of these is necessarily fraudulent. Each is statistically anomalous. The entries that trigger five or more filters simultaneously (manual + weekend + CFO + revenue account + period-end + vague description) are the starting population for forensic investigation.
The transaction impact: if the investigation reveals that any of these entries were designed to inflate reported revenue or reduce reported costs, the earnings quality adjustment directly reduces the EBITDA base and the transaction price.
The Methodology
In Northrop Management Private Limited’s due diligence practice, ERP analysis follows a structured eight-step sequence.
Step 1: Full data extraction. Complete GL, all sub-ledgers, user master, vendor master, customer master, chart of accounts.
Step 2: Reconciliation. Reconcile extracted data to the trial balance and financial statements. Discrepancies are themselves findings.
Step 3: Revenue deep-dive. Revenue by customer, product, geography, month and last-five-days-of-period. Concentration, timing, new counterparties, credit notes and returns.
Step 4: Procurement deep-dive. Purchases by vendor, amount (threshold-splitting), timing and vendor substance.
Step 5: Inventory analytics. Ageing by item and location. Provision adequacy. Items with no movement for six months.
Step 6: Journal entry analytics. Ten-filter triage on the full population.
Step 7: Receivable and payable analysis. Ageing by counterparty. Concentration. Related-party balances. Unusual movements.
Step 8: Data quality assessment. Completeness, consistency and integrity of the ERP data itself. Incomplete data is a finding about the quality of the target’s financial infrastructure.
Ashish Chaudhary, frames the analytical shift directly: “The question should move from ‘what do the financial statements say?’ to ‘what does the underlying transaction data show?’ The financial statements are the summary. The ERP is the evidence. And the evidence, examined at the transaction level, almost always reveals patterns, concentrations and anomalies that the summary was not designed to show and that the board should see before approving the price.”
Questions for the Boardroom
- Has the diligence team analysed the target’s transaction-level data (full GL, sub-ledgers, vendor and customer masters), or has it relied on the financial statements and management schedules?
- What is the customer concentration at the transaction level, and is any single customer or group of connected customers responsible for more than 15% of revenue?
- What percentage of revenue is recognised in the last five business days of each quarter, and how much of Q4 revenue is reversed through Q1 credit notes?
- How many vendors share attributes (address, director, contact details) with the company or its promoters, and what is the total procurement value flowing to those vendors?
- If a journal entry analytics run were applied to the full GL, how many entries would trigger five or more anomaly filters simultaneously?
Practical Checklist
Request: Full GL export, sub-ledger exports, vendor master, customer master, user master, chart of accounts.
Test within first week: Revenue by customer and month. Procurement by vendor and amount. Inventory by age. Journal entries by ten-filter triage. Entity resolution across vendor, customer and employee databases.
Red flags to escalate: Revenue concentrated in last five days. Credit notes reversing prior-quarter revenue. Vendors with substance concerns. Inventory older than 12 months without adequate provision. Manual journal entries by senior personnel affecting revenue or provision accounts.
Quantify every finding: What does it mean for EBITDA? For working capital? For net debt? For the purchase price?
Closing Implication
Financial statements tell the board what management wants reported. The ERP tells the investigator what actually happened. The gap between the two is where diligence findings reside, and the gap is identifiable, quantifiable and, critically, negotiable before the purchase price is committed. A board that insists on ERP-level diligence governs the transaction with evidence. A board that accepts financial-statement-level diligence governs it with summaries.