Content authored by practitioners with experience at Amazon India, Intuit QuickBooks, and the Tata Group. Meet the team →
Indian finance teams treat manual reconciliation as one technique — a preparer walks the register, ticks each row, and flags variances. That framing hides the fact that seven distinct manual techniques exist, each covers a different failure class, and each has a different volume ceiling. Teams that run only one or two techniques miss the failure modes the other techniques were designed to catch — a team running two-way tick-and-tie on the TDS ledger without a ratio-analysis overlay misses systemic rate-band misapplication on Section 194J code 1005 (technical services at two percent booked as professional services at ten percent), a team running conservation checks on GST without a three-way match against GSTR-2B misses Rule 36(4) unsupported ITC availment, and a team running peer review without an aging queue lets Section 16(4) 30 November permanent-loss exposures drift past the cutoff.
Publish the seven techniques as a detection portfolio: ratio analysis for aggregate drift on rates and ratios (TDS-payable-to-purchase, GST-output-to-turnover, ITC-to-purchase, bank-charges-to-turnover); two-way tick-and-tie for missing entries and content errors between two independent records (invoice-to-bank, challan-to-ledger); three-way tick-and-tie for the ITC stream where purchase register, GSTR-2B, and IMS action must reconcile under Rule 36(4); exception aging queue with 0-60, 61-120, 121-180, 180-plus day buckets and named escalation triggers for slow-clearing exceptions; independent peer review with statutory audit-style SA 315 checklist for process-consistency gaps; conservation checks for structural errors (SGST + CGST = IGST for intra-state supply reclassification; debit + credit = 0; Rule 42/43 input + output + net = ITC balance); reasonableness testing anchored to the enterprise's own materiality floor for outliers. Match every High Action Priority row on the register to two or three of the seven techniques based on the failure class it belongs to, and re-rate Detection on the anchored SOD scale accordingly.
One technique portfolio published in the reconciliation policy document; every reconciliation stream carries a documented mapping of failure class to technique to owner to cadence; the seven techniques' volume ceilings are recorded against the current transaction base and re-tested every quarter; when any technique reaches its ceiling the corresponding High Action Priority rows are re-evaluated for detection uplift; the aging queue's 180-plus day bucket escalation to the audit committee before the Section 16(4) 30 November cutoff is a standing agenda item every October audit committee meeting; the independent peer review's SA 315 checklist is refreshed annually against the ICAI's Standards on Auditing pronouncements; every failure caught by a manual technique that would not have been caught by the other six is documented as evidence for the technique's design under Section 143(3)(i) Internal Financial Controls testing.
A reconciliation detection layer whose coverage is defensible under statutory audit — every High Action Priority row on the register carries at least two of the seven techniques against its failure class; every Section 16(4) 30 November exposure is aged and escalated through the 180-plus day bucket before the deadline; every Section 200A short-deduction exposure on the TDS ledger is caught by a challan-to-Form 168 two-way match before quarterly filing; every Rule 36(4) unsupported ITC availment is caught by the three-way match against GSTR-2B before it reaches GSTR-3B. When any of the seven techniques crosses its documented volume ceiling — 10,000 transactions per month for ratio analysis, 2,000 for two-way tick-and-tie, 1,500 line items for three-way tick-and-tie, 500 open items for the aging queue — the register itself names the exposure the manual layer cannot close, and the finance team's board case for a continuous detection layer writes itself.
Manual reconciliation is not one technique. It is a portfolio of seven — ratio analysis, two-way tick-and-tie, three-way tick-and-tie, exception aging with escalation, independent peer review with a written checklist, conservation checks, and reasonableness testing. Each covers a different class of failure modes, holds a different volume ceiling, and produces a different evidence artefact for a statutory auditor to test under Section 143(3)(i) of the Companies Act 2013. A finance team that runs only two of the seven has three or four failure classes that no detection control on its register catches — and the register itself, if it has been walked through Terra Insight’s anchored SOD scale, names those uncovered rows as High Action Priority every quarter until the technique portfolio is completed.
This method article publishes the working discipline for each of the seven techniques — what it catches, how it runs on an Indian reconciliation stream, its volume ceiling, and where it hands off to the next layer.
The detection portfolio — one technique per failure class
The 12-class failure-mode taxonomy that Terra Insight publishes on the reconciliation process design pillar names data extraction, classification, completeness, matching, timing, partner, precision, policy, aging, cutoff, evidence, and portal as the twelve failure classes any Indian reconciliation stream can produce. No single detection technique covers all twelve. Ratio analysis catches aggregate drift but misses individual missing entries. Two-way tick-and-tie catches missing entries but misses classification errors where two matching amounts sit under the wrong Section code. Three-way tick-and-tie catches classification but only inside the ITC stream where a GSTR-2B row exists to match against. Aging queues catch slow-clearing exceptions but do nothing on the day-one exception itself. Peer review catches process-consistency gaps but consumes an entire reviewer day. Conservation checks catch structural errors like a SGST plus CGST total that does not equal the IGST total, but flag nothing on rate misapplication. Reasonableness testing catches outliers but produces false positives without an anchored materiality reference.
The right posture is a portfolio. Every High Action Priority row on the reconciliation control plan is mapped to the two or three techniques that specifically address its failure class, and the Detection rating on the anchored SOD scale is re-rated against the composite catch-rate rather than the strongest single technique.
Technique 1 — Ratio analysis for aggregate drift
Ratio analysis compares a computed ratio at aggregate level against its prior-period value, its budgeted value, or an external benchmark, and flags a step change outside a defined band. The failure classes it catches are classification (rate-band misapplication), completeness (missing entries at scale), and portal (a portal-side aggregate that has drifted).
Four ratios anchor the Indian reconciliation cycle:
- TDS-payable-to-purchase ratio. Aggregate TDS deducted in the period divided by aggregate purchases subject to withholding. A step change outside a plus-or-minus one-percentage-point band signals a rate-band misapplication — for example, a batch of Section 194J professional-services invoices booked at the two percent technical-services rate under payment code 1005 instead of the ten percent professional-services rate under the same code 1005, or a batch of Section 194Q purchases missing the 0.1 percent deduction against payment code 1031. The Section 194Q trigger walkthrough covers the payment-code mechanics in detail.
- GST-output-to-turnover ratio. Aggregate GST charged on outward supplies divided by aggregate turnover. A step change outside a plus-or-minus half-a-percentage-point band signals a place-of-supply misclassification (intra-state supply booked as inter-state or vice versa), a rate reclassification (18 percent output tax booked at 12 percent), or an exempt-supply completeness gap.
- ITC-to-purchase ratio. Aggregate ITC availed in GSTR-3B divided by aggregate GST-eligible purchases. A step change outside a plus-or-minus one-percentage-point band signals either unsupported ITC availment (Rule 36(4) violation, catchable by the three-way match) or a Rule 42 or Rule 43 common-credit apportionment error where the exempt-supply reversal has been miscalculated.
- Bank-charges-to-turnover ratio. Aggregate bank charges booked to the P&L divided by aggregate turnover. A step change flags missing pass-through of merchant discount rates, a missing GST reverse-charge on bank charges, or a bank tariff change that has not been surfaced to the finance team.
Volume ceiling. Ratio analysis holds up to roughly 10,000 transactions per month because the ratio is computed at aggregate level regardless of row count — the technique is scale-free on the arithmetic and only bounded by the reviewer’s ability to investigate the flagged variance. Above 10,000 transactions per month, the aggregate ratio still computes but the underlying investigation queue (a two-percentage-point drift on a ten-thousand-row base carries two hundred candidate rows to sample from) exceeds the reviewer’s capacity to walk manually.
Technique 2 — Two-way tick-and-tie for missing entries and content errors
The two-way tick-and-tie matches two independent records of the same economic event. The failure classes it catches are completeness (a record exists on one side and not the other), matching (both records exist but do not agree on amount, date, or counterparty), and partner (the counterparty identifier does not resolve to a valid master record).
Two Indian reconciliation streams anchor the technique:
- Invoice-to-bank two-way match. Every issued sales invoice must reconcile to a bank credit within the collection window, and every issued purchase invoice must reconcile to a bank debit within the payment window. The invoice-to-bank reconciliation failure modes analysis documents the twelve failure modes this technique catches, including the HDFC-Bank narration parsing patterns that surface merchant discount rate deductions and platform-settlement variance.
- TDS challan-to-ledger two-way match. Every TDS challan booked to the ledger must reconcile to a valid Challan Identification Number in the OLTAS system, with the correct payment code (1031 for Section 194Q, 1005 for Section 194J, and the ninety other Section 393 codes catalogued in the TDS payment codes 1001-1092 guide), and every deductee row must reconcile to a matching Form 26AS or Form 168 credit at the deductor’s CPC-TDS processing. A challan booked at ten percent under code 1005 against a technical-services vendor that should have carried the two percent code 1005 rate creates a Section 200A short-deduction exposure that only the two-way match catches before quarterly filing.
Volume ceiling. Two-way tick-and-tie tops out around 2,000 transactions per month per stream because the per-row walk consumes reviewer time linearly at roughly one to two minutes per row. Above 2,000 transactions per month on a single stream, the sampling posture shifts from full-population to statistical-sample, and the Detection rating on the anchored SOD scale drops from D5 (full population) to D7 (sample-based) unless a second detection layer is added.
Technique 3 — Three-way tick-and-tie for the ITC stream
The three-way tick-and-tie adds a third independent record to the two-way match. For the Indian ITC stream, the third record is the GSTR-2B download from the GSTN portal, which the enterprise did not create and cannot edit. The three-way match reconciles purchase register versus GSTR-2B versus the IMS action taken on the same document, and it is the only manual technique that satisfies Rule 36(4) at the record-keeping level.
The technique catches four failure classes the two-way match cannot: ITC misclassification (invoice booked to a GSTIN that does not match the vendor’s registered GSTIN), IMS action inconsistency (the invoice was accepted in IMS but the corresponding purchase register entry was reversed), portal-side omission (the invoice appears in the purchase register but not in GSTR-2B because the supplier missed the filing window), and Rule 37A cascading reversal risk (the invoice appears in GSTR-2B this month but the supplier’s Rule 37A default risk requires proactive reversal — see the Rule 37 and Rule 37A ITC reversal guide).
The three-way ITC reconciliation in Excel guide covers the working paper layout — three sheets (Purchase Register, GSTR-2B, IMS Actions) joined on invoice number and GSTIN, with a variance column that flags any row not present on all three.
Volume ceiling. Three-way tick-and-tie tops out around 1,500 line items per month per GSTIN because the per-row walk consumes roughly two to three minutes (the additional portal-download and IMS-action step). A multi-GSTIN enterprise with four GSTINs each processing 1,000 line items per month is already at 4,000 line items across the group and past the single-reviewer ceiling. The IMS versus GSTR-2B reconciliation walkthrough documents the same ceiling from the IMS-action side.
Technique 4 — Exception aging queue with escalation
An open-items list carries every unresolved reconciliation exception in one bucket. An exception aging queue with escalation puts every open item into a time-bounded bucket and attaches a named escalation trigger to each bucket. Four buckets anchor the discipline:
- 0 to 60 days. Analyst-owned. Daily follow-up. Standard cadence.
- 61 to 120 days. Escalated to the controller. Weekly review. Ledger-level entry in the exception log.
- 121 to 180 days. Escalated to the CFO. Personal follow-up with the supplier or the deductor. Rs 5 lakh threshold triggers a written acceptance rationale from the controller.
- 180 days and beyond. Escalated to the audit committee. Every October audit committee agenda carries the 180-plus day bucket as a standing item, and every GSTR-2B ITC row in the bucket is walked to a Section 16(4) accept-or-reverse decision before the 30 November cutoff.
The failure classes the aging queue catches are aging (the exception has been open longer than the process design permits) and cutoff (an aging exception is approaching a statutory deadline like Section 16(4) 30 November, Section 39(9) 30 November GST amendment window, or the FY-end 31 March for Section 43B(h) MSME payments and Section 194Q trigger recognition).
Volume ceiling. An aging queue with the four buckets and the named escalation triggers manages roughly 500 open items before the daily and weekly review cadence exceeds a single reviewer’s capacity. Above 500 open items, the aging queue itself is fine but the escalation cadence collapses — the controller cannot maintain a weekly review on 300-plus rows in the 61 to 120 day bucket, and the queue devolves into a snapshot rather than an escalation.
Technique 5 — Independent peer review with statutory audit-style checklist
Independent peer review with a written checklist is the enterprise’s own analogue of the statutory auditor’s SA 315 walk. ICAI’s Standard on Auditing SA 315 requires the statutory auditor to perform risk assessment procedures that walk the reconciliation function’s design and operating effectiveness at interim. The peer review technique compresses the same walk into a monthly one-close-cycle pass by a reviewer who did not run the reconciliation.
The reviewer works from a checklist tied to the reconciliation stream’s failure-mode inventory:
- Design walk. Was the reconciliation function performed against the documented Standard Operating Procedure? Are all data sources named on the working paper? Are the failure-mode-to-technique mappings from the reconciliation control plan visible on the working paper?
- Operating walk. Are the tie-out totals reproducible by the reviewer? Do the sampled two-way and three-way matches hold? Is every aging-queue escalation documented with the escalation trigger?
- Evidence walk. Is the working paper retained under the seven-year Section 128(5) rule? Are all supporting artefacts (bank statements, GSTR-2B downloads, Form 168 extracts, IMS action logs) named and retrievable?
The failure class the peer review catches is policy (the reconciliation did not follow the documented process and produced a wrong result silently). The peer review also serves as the second detection layer that the reconciliation control plan template requires on every High Action Priority row.
Volume ceiling. Independent peer review with a full SA 315-style checklist runs at approximately one full close cycle per month because the walk is compressed and the reviewer’s capacity to hold the entire stream in working memory is bounded by the close cycle’s length. Two peer reviews per month on the same stream produce diminishing catch-rate returns — the reviewer’s fresh-eyes effect wears off. The statutory audit reconciliation checklist publishes the SA 315-style checklist the peer review runs from.
Technique 6 — Conservation checks for structural errors
A conservation check is a mathematical identity that a correctly-recorded reconciliation must satisfy. Three identities anchor the Indian reconciliation cycle:
- SGST + CGST = IGST for intra-state supply reclassification. The total State GST plus the total Central GST charged on an intra-state supply must equal the Integrated GST that would have been charged on the same supply had it been treated as inter-state at the same rate. A break in the identity flags either a place-of-supply misclassification, a rate misapplication, or an entry error in the SGST-CGST split. The GSTR-1 versus GSTR-3B failure modes analysis walks the identity’s application on the outward-supply side.
- Debit + Credit = 0. The ledger’s debit and credit totals across the reconciliation working paper must reconcile to zero. A non-zero balance flags an entry error, a missing counterparty side, or a rounding accumulation that has crossed a materiality threshold. The identity is trivial arithmetically but non-trivial to enforce on a working paper that is edited across a close cycle.
- Rule 42/43 Input + Output + Net = ITC balance. The sum of Input tax credit availed plus Output tax paid plus Net ITC utilised across a Rule 42 or Rule 43 common-credit walk must reconcile to the ITC available balance at month-end. A break in the identity flags a common-credit apportionment error that would otherwise reach the annual reconciliation in the September-following-FY-end return as an aggregate variance.
The failure classes conservation checks catch are structural: place-of-supply misclassification, entry errors that break the double-entry identity, and common-credit apportionment errors. The failure classes conservation checks miss are content: a rate misapplication where the identity holds (SGST 9 percent plus CGST 9 percent equals IGST 18 percent even when the correct rate was 12 percent), a missing entry that is symmetrically missing on both sides, or a classification error where two amounts match under the wrong Section code.
Volume ceiling. Conservation checks are scale-free because the check is a formula, not a row-by-row walk. Coverage is bounded by the number of identities the working paper carries — three identities on a typical monthly close, up to seven on a full-scope statutory audit walk — but every identity computes in constant time regardless of the underlying transaction volume.
Technique 7 — Reasonableness testing anchored to materiality
Reasonableness testing compares a computed value against an expected value derived from an independent estimation and flags any value outside a defined tolerance band. The failure classes it catches are precision (the value is inside the identity but outside the reasonable range) and evidence (the underlying assumption cannot be defended).
Two variants anchor the discipline:
- Materiality-anchored reasonableness. Any variance below the enterprise’s own materiality floor of Rs 5,000 per invoice or Rs 1 lakh per close-cycle aggregate is auto-cleared without investigation. Any variance at or above the threshold is escalated to the analyst for content investigation. The materiality anchor prevents the reviewer from spending time on paise-level rounding differences while ensuring that a Section 16(4) exposure starting at Rs 12,000 per month (which sits above the threshold on the aggregate) is escalated to the aging queue.
- Variance-analysis reasonableness. A ratio, a rate, or an aggregate that has drifted more than a plus-or-minus five-percentage-point band from the prior-period value is escalated as a content anomaly. The variance-analysis pass complements the ratio analysis in Technique 1 — ratio analysis flags the drift, variance-analysis reasonableness investigates the content.
Volume ceiling. Reasonableness testing is scale-free on the arithmetic but produces false positives at high volume without an anchored materiality reference. A team running variance-analysis reasonableness on a 10,000-transaction base without a Rs 5,000-per-invoice materiality anchor generates roughly 500 false-positive escalations per month — the technique’s coverage collapses under its own escalation load.
Worked case 1 — the systemic Section 194J rate-band misapplication caught only by ratio analysis
A Rs 240 crore Indian IT services enterprise runs a Section 194J TDS ledger with roughly 480 professional-services vendor payments per month. The two-way tick-and-tie technique reconciles every challan to the ledger row and every ledger row to Form 168, and the reviewer signs off on the reconciliation at Detection 5 on the anchored SOD scale. In Q2 FY 2026-27, a batch of 120 technical-services vendor invoices totalling Rs 3.4 crore is booked at the ten percent professional-services rate under Section 194J payment code 1005 instead of the correct two percent technical-services rate under the same code 1005. The two-way tick-and-tie technique confirms every challan matches its ledger row (same amount, same code, same vendor) and produces a clean tie-out. No exception surfaces.
The TDS-payable-to-purchase ratio for the quarter walks from a prior-quarter value of 6.7 percent to a current-quarter value of 8.9 percent — a 2.2-percentage-point drift on a stream where the historical band is plus-or-minus one percentage point. The ratio-analysis technique surfaces the drift as a High-priority variance investigation, and the investigation traces the 2.2-percentage-point gap to the 120 technical-services invoices carrying the wrong rate band. The Section 200A short-deduction exposure of roughly Rs 27 lakh (the 8 percent gap on Rs 3.4 crore) plus Section 201(1A) interest at 1 percent per month for short-deduction is caught before quarterly filing under Rule 31A, and the correction is filed in the same quarter with no demand notice.
The two-way tick-and-tie technique cannot catch this failure class alone because the challan and ledger row agree on all four dimensions. Only the ratio analysis overlay surfaces the aggregate drift. The TDS reconciliation failure modes analysis catalogues rate-band misapplication as a Severity-9 row that the two-way match alone under-detects.
Worked case 2 — the Section 16(4) exposure that only the aging queue’s 180-plus day escalation catches
A pharmaceutical distributor with a Rs 300 crore annual GST-eligible purchase base runs a monthly GSTR-2B reconciliation using the three-way tick-and-tie against 220 suppliers and roughly 3,800 inbound invoices per month. The three-way match runs at Detection 5 on the anchored SOD scale on the population and catches most Rule 36(4) unsupported-ITC exposures at the month of invoice. But three suppliers — carrying an aggregate ITC of Rs 4.7 lakh per month — have filed no GSTR-1 for four consecutive months and their invoices sit in the purchase register with no GSTR-2B match. The three-way match flags the invoices as exceptions and rolls them into the open-items list.
The exception aging queue with escalation puts the 12 exception rows into the 0-60 day bucket at first surfacing. In month three, they cross into the 61 to 120 day bucket and the controller opens a written follow-up log with the three suppliers. In month five, they cross into the 121 to 180 day bucket, the CFO escalates the follow-up personally, and one of the three suppliers files the missing GSTR-1s in month six clearing Rs 1.6 lakh of the exposure. In October, the remaining 8 rows carrying Rs 3.1 lakh of ITC cross into the 180-plus day bucket. The October audit committee walks the rows to a Section 16(4) decision, decides to reverse Rs 2.4 lakh of the ITC in the November GSTR-3B ahead of the 30 November cutoff (avoiding a Section 74 fraud-recovery exposure on ITC availed without supplier compliance), and pursues the residual Rs 700,000 through a formal supplier demand. The Section 16(4) time bar guide covers the cut-off mechanics and the DRC-01B reconciliation reply guide covers the notice-response cadence that a reversed row can still attract.
Without the aging queue’s 180-plus day escalation the rows would have sat on the open-items list until the 30 November cutoff, at which point Rs 3.1 lakh of ITC would have crossed into permanent Section 16(4) loss and the Section 74 fraud-recovery exposure on the availed-and-unreversed Rs 2.4 lakh would have been alive.
Worked case 3 — the SGST plus CGST equals IGST conservation check that catches a place-of-supply misclassification
A mid-market Indian manufacturer with a Rs 180 crore annual outward-supply base runs a monthly GSTR-1 reconciliation against the sales register. In August 2026, the sales register shows Rs 42 crore of intra-state supply carrying SGST at 9 percent (Rs 3.78 crore) plus CGST at 9 percent (Rs 3.78 crore) totalling Rs 7.56 crore in State-and-Central tax. The IGST on the same value at 18 percent would have been Rs 7.56 crore — the conservation identity holds.
But one batch of Rs 4.2 crore of supply to a project site in an adjacent state has been mis-classified as intra-state and the SGST-CGST split has been recorded at 9-percent-each even though the destination-state address on the invoice is a different state code. The conservation identity holds on the arithmetic (Rs 37.8 lakh SGST plus Rs 37.8 lakh CGST equals Rs 75.6 lakh IGST at 18 percent) but the ledger classification is wrong — the invoice should have carried IGST at 18 percent under a place-of-supply rule for inter-state supply, not SGST-CGST.
The conservation check as stated at the total-invoice level misses the misclassification because the arithmetic holds. But a per-invoice conservation check — SGST plus CGST equals the IGST amount that would have been charged at the same rate if the supply were inter-state, tested against the destination-state code on the invoice — flags the batch: the destination-state code does not match the source-state code, so the intra-state classification is invalid, and the reviewer investigates the batch and reclassifies. Without the per-invoice conservation check keyed to the destination-state code, the misclassification would have flowed into GSTR-1, and a subsequent GSTR-1 versus GSTR-3B mismatch under Rule 88C and DRC-01B would have surfaced the misclassification only after the seven-day-reply-window notice was served.
The conservation-check technique needs the per-invoice application to catch the failure — the aggregate check would have passed silently.
Where this fits in the wider methodology
The seven manual detection techniques are the detection layer of the reconciliation process design method Terra Insight publishes on the methodology pillar. They pair with the 6P cause taxonomy that names the underlying cause of every failure mode, with the Action Priority table that ranks every row on Severity first, with the anchored SOD scale that re-rates Detection against the composite catch-rate of the techniques applied, and with the prevention controls working list that reduces the likelihood of the cause.
On the operational side, the seven techniques are the day-to-day discipline that the monthly close reconciliation playbook sequences across the close cycle. The bank reconciliation runbook for days 1 to 5 runs the two-way match and the ratio analysis. The GSTR-2B ITC runbook for days 11 to 15 runs the three-way match and the aging queue. The GSTR-1 versus GSTR-3B runbook for days 16 to 20 runs the conservation checks and the reasonableness tests. The TDS reconciliation runbook runs the two-way tick-and-tie and the ratio-analysis overlay. The TDS receivable aging workbook instantiates the aging queue for the TDS receivable side. And the bank narration parsing Excel formulas surface the merchant discount rate and platform-settlement variance patterns the two-way match walks.
Every High Action Priority row on the reconciliation control plan is mapped to the two or three techniques from this portfolio that specifically address its failure class. The reconciliation control plan template publishes the mapping structure.
When the manual detection layer stops being economically viable
Each of the seven techniques has a documented volume ceiling — 10,000 transactions per month for ratio analysis, 2,000 for two-way tick-and-tie, 1,500 line items for three-way tick-and-tie, 500 open items for the aging queue, one full close cycle per month for peer review, scale-free but structural-only coverage for conservation checks, scale-free but false-positive-prone for reasonableness testing. Above roughly 200 vendors, 3,000 monthly invoices, or a multi-GSTIN structure with more than three GSTINs, the seven manual techniques cannot together sustain the composite Detection rating that a High Action Priority row on the register demands. The register itself begins to name the exposure — a Section 16(4) 30 November queue that cannot be walked at population, a Section 194Q payment-code 1031 aggregate that cannot be ratio-analysed at the required frequency, a Rule 36(4) three-way match that cannot be run at 4,000-plus line items per GSTIN — and the finance team’s board case for a continuous detection layer writes itself.
Terra Insight’s reconciliation software surface carries the continuously-refreshed detection layer that pairs with the manual technique portfolio on the streams where the manual ceiling has been crossed. The GST reconciliation software runs the three-way tick-and-tie at full population every day rather than once a month at reviewer capacity. The TDS reconciliation software runs the two-way tick-and-tie against Form 26AS and Form 168 continuously with a payment-code-aware validation gate that surfaces a Section 194J code 1005 rate-band drift before quarter-end. The monthly close reconciliation playbook sequences the manual technique portfolio and the continuous detection layer against a common cadence, so the finance team’s own reviewers remain the design authority on every High Action Priority row while the continuous layer carries the population-scale walk the manual ceiling could not sustain.
Where this fits
- Reconciliation process design — the methodology pillar
- The anchored SOD rating scale
- Action Priority vs materiality
- The 6P cause taxonomy
- Prevention controls for manual reconciliation
- Reconciliation control plan template
- Three-way ITC reconciliation in Excel
- Statutory audit reconciliation checklist
- Reconciliation process design worksheet
Frequently Asked Questions
Why publish seven manual detection techniques when a single strong technique should be enough?
No single manual technique catches every failure class. Ratio analysis catches aggregate drift but misses individual missing entries. Two-way tick-and-tie catches missing entries but misses classification errors between two matching amounts under the wrong Section code. Three-way tick-and-tie catches classification but only inside the ITC stream where a GSTR-2B row exists to match against. Aging queues catch slow-clearing exceptions but do nothing on the day-one exception itself. Peer review catches process-consistency gaps but consumes an entire reviewer day. Conservation checks catch structural errors like a SGST plus CGST total that does not equal the IGST total but flag nothing on rate misapplication. Reasonableness testing catches outliers but produces false positives without an anchored materiality reference. The seven techniques compose a detection layer — each covers a distinct failure class, and every High Action Priority row on the reconciliation register is walked through the two or three techniques that specifically address its class.
What is the difference between a two-way tick-and-tie and a three-way tick-and-tie in the Indian reconciliation context?
A two-way tick-and-tie matches two independent records of the same event — a bank credit against an invoice line, a TDS challan against a deductee row in the TDS receivable ledger, a supplier invoice against the purchase register entry. It confirms the two records agree on amount, date, counterparty, and identifier. A three-way tick-and-tie adds a third independent record — for the ITC stream in India, the third record is the GSTR-2B download from the GSTN portal, which the enterprise did not create and cannot edit. The three-way match reconciles purchase register versus GSTR-2B versus the IMS action taken on the same document, and it is the only manual technique that satisfies Rule 36(4) at the record-keeping level. The three-way match is more expensive per row but is the required design for any High Action Priority ITC row on the register.
How does exception aging with escalation differ from a simple open-items list?
An open-items list carries every unresolved reconciliation exception in one bucket. An exception aging queue with escalation puts every open item into a time-bounded bucket — 0 to 60 days, 61 to 120 days, 121 to 180 days, and 180-plus days — and attaches a named escalation trigger to each bucket. A GSTR-2B ITC mismatch in the 0 to 60 day bucket sits with the analyst for daily follow-up. In the 61 to 120 day bucket it escalates to the controller for weekly review. In the 121 to 180 day bucket it escalates to the CFO for personal follow-up with the supplier. In the 180-plus day bucket the row is escalated to the audit committee for a decision on Section 16(4) accept-or-reverse before the 30 November cutoff. The aging queue is the detection layer that stops a slow-moving exception from silently crossing the Section 16(4) 30 November cutoff, and it is the second detection layer that Terra Insight’s reconciliation control plan template requires on every High Action Priority row.
What is a conservation check and why does the SGST plus CGST equals IGST identity matter?
A conservation check is a mathematical identity that a correctly-recorded reconciliation must satisfy. The SGST plus CGST equals IGST identity is one of them — the total State GST plus the total Central GST charged on an intra-state supply must equal the Integrated GST that would have been charged on the same supply had it been treated as inter-state. If the ledger’s SGST-CGST split for an intra-state supply totals to more or less than the IGST amount at the same rate, one of three structural errors has occurred: the supply has been mis-classified as intra-state when it was inter-state, the tax rate has been misapplied, or the SGST-CGST split itself carries an entry error. A second conservation identity is that debit totals must equal credit totals across the reconciliation working paper. A third is that the sum of Input plus Output plus Net across a Rule 42 or Rule 43 common-credit walk must reconcile to the ITC available balance at month-end. Conservation checks scale to any transaction volume because the check is a formula, not a row-by-row walk, but they only surface structural errors — they miss content errors where the identity holds but the underlying classification is wrong.
At what point does the manual detection layer stop being economically viable?
Each of the seven techniques has a documented volume ceiling. Ratio analysis holds up to roughly 10,000 transactions per month because the ratio is computed at aggregate level regardless of row count. Two-way tick-and-tie tops out around 2,000 transactions per month per stream because the per-row walk consumes reviewer time linearly. Three-way tick-and-tie tops out around 1,500 line items per month for the same reason plus the added portal-download and IMS-action step. An aging queue with escalation manages roughly 500 open items before the daily and weekly review cadence exceeds a single reviewer’s capacity. Independent peer review with checklist runs at approximately one full close cycle per month because the walk is compressed. Conservation checks are scale-free but low-severity coverage. Reasonableness testing is scale-free but produces false positives at high volume without an anchored materiality reference. Above roughly 200 vendors, 3,000 monthly invoices, or a multi-GSTIN structure with more than three GSTINs, the seven manual techniques cannot together sustain the High Action Priority detection layer that the anchored SOD scale demands, and the finance team’s own risk register begins to name the exposure the manual layer cannot close.
Content authored by practitioners with experience at Amazon India, Intuit QuickBooks, and the Tata Group. Meet the team →
- ▸ Rule 36(4), Central Goods and Services Tax Rules 2017 — Input Tax Credit to be availed by a registered person in respect of invoices or debit notes shall be availed only if the details of such invoice or debit note have been furnished by the supplier in Form GSTR-1 or through the Invoice Furnishing Facility and such details have been communicated to the recipient in Form GSTR-2B. Rule 36(4) is the statutory anchor of the three-way tick-and-tie — purchase register versus GSTR-2B versus IMS action — and any ITC availed in Form GSTR-3B without a matching GSTR-2B entry is unsupported at the record-keeping level and reversible under Rule 88D through Form DRC-01C.
- ▸ Section 16(4), Central Goods and Services Tax Act 2017 — A registered person shall not be entitled to take Input Tax Credit in respect of any invoice or debit note for supply of goods or services after the 30th day of November following the end of the financial year to which such invoice or debit note pertains. The Section 16(4) 30 November cutoff is the aging-queue's outer wall — every open GSTR-2B mismatch item must be escalated and resolved before the deadline because there is no rectification, no condonation of delay, and no recovery mechanism after the cutoff.
- ▸ Rule 42 and Rule 43, Central Goods and Services Tax Rules 2017 — The Input Tax Credit attributable to exempt supplies and non-business purposes shall be reversed on a monthly basis under Rule 42 for inputs and input services, and under Rule 43 for capital goods, with an annual reconciliation in the return for September of the following financial year. Rule 42 and Rule 43 produce a monthly reversal identity that the ratio-analysis technique tests as ITC availed to purchase base ratio — a step change in the ratio outside a five-percentage-point band flags a common-credit apportionment error that would otherwise reach the annual reconciliation as an aggregate variance.
- ▸ Section 200A, Income-tax Act 1961 (retained in Income-tax Act 2025) — The Central Processing Centre shall process every TDS statement filed under Section 200, adjust any arithmetical error or incorrect claim, and issue a demand notice for the amount payable with interest under Section 201(1A) at 1 percent per month for short-deduction and 1.5 percent per month for short-payment, and Section 234E fee at Rs 200 per day capped at the tax deductible. The two-way tick-and-tie between challan and deductee row against Form 26AS or Form 168 is the manual detection technique that pre-empts a Section 200A demand — every challan booked to the ledger must reconcile to a valid CIN with the correct payment code before quarterly filing.
- ▸ Companies (Auditor's Report) Order 2020, Clause 3(xii) — The auditor is required to report whether the Nidhi Company has complied with the Net Owned Funds to deposits ratio of 1:20 and maintained ten percent unencumbered term deposits. Clause (xii) is one of several CARO 2020 clauses whose evidence base rests on a ratio-analysis reconciliation — the auditor's testing under this clause is a compressed form of the ratio-analysis technique this article covers, and the enterprise's own monthly ratio walk is the design-side evidence that pre-empts an audit exception.
- ▸ ICAI Standard on Auditing SA 315, Identifying and Assessing the Risks of Material Misstatement — The auditor shall perform risk assessment procedures to obtain an understanding of the entity and its environment, including the entity's internal control, sufficient to identify and assess the risks of material misstatement, whether due to fraud or error, at the financial statement and assertion levels. SA 315 requires the statutory auditor to walk the reconciliation function's design and operating effectiveness at interim, and the independent peer review technique covered here is the enterprise's own analogue of that walk — a monthly compressed SA 315 test on the reconciliation stream before the auditor's year-end pass.
- ▸ Section 194Q read with CBDT Notification specifying payment code 1031, Income-tax Act 1961 (retained in Income-tax Act 2025) — A buyer whose aggregate turnover exceeds ten crore rupees in the immediately preceding financial year shall deduct tax at 0.1 percent on the value of purchase of goods from a resident seller exceeding fifty lakh rupees in a financial year. From 1 April 2026, Section 194Q deductions are reported under the Section 393 four-digit payment code 1031 in place of the legacy Section 194x identifier. The two-way challan-to-ledger tick-and-tie on any FY 2026-27 quarter must map the challan to code 1031 and reconcile the aggregate against the buyer's Section 194Q register before Form 168 filing under Rule 31A.
- ▸ Section 194J read with CBDT Notification specifying payment code 1005, Income-tax Act 1961 (retained in Income-tax Act 2025) — Tax shall be deducted at ten percent on fees for professional services and at two percent on fees for technical services paid to a resident. From 1 April 2026, Section 194J deductions are reported under the Section 393 four-digit payment code 1005. The two-way tick-and-tie between the professional-services ledger and Form 168 must reconcile every code 1005 challan to the underlying vendor row and rate-band — a code 1005 challan booked at ten percent against a technical-services vendor that should have carried the two percent rate creates a Section 200A short-deduction exposure caught only by the manual detection layer.