How to Automate COD Remittance Reconciliation for Indian Sellers

Why a short remittance is easy to miss

Cash on delivery settles in batches, and that is where errors hide. The courier collects cash at the door, keeps its own charges, and pays the seller a lump sum on its own schedule. The total often looks roughly right, so most owners accept it. A payout can cover fewer orders than were delivered, and the shortfall stays invisible unless someone compares each order with its payment. Across a few months, those small gaps add up to real money.

Shortfalls come from several places, not only from courier error. A delivered order can drop out of a batch. A weight correction can reduce the amount. An RTO on a related shipment can be netted against a later delivery. Each of these looks like a short payment when you read only the total.

What the problem costs

There are three cost areas, and they stack. The first is deductions. Shipmozo's 2026 guide notes that couriers often deduct forward freight, RTO fees and COD handling charges from the payout without a clear line for each item. Without an AWB-level breakdown, you cannot check whether each deduction is correct. Shipmozo's same guide puts cash lost to unexplained deductions and short remittances at 2 to 4 percent, but that figure comes from a vendor blog, so treat it as a rough signal and measure your own.

The second cost is returns. Estimates differ widely. Flexicommerce's 2026 COD calculator puts the average RTO rate for Indian COD orders at 15 to 25 percent. Clickpost's 2026 guide puts COD return-to-origin rates at 40 to 58 percent, with fashion and apparel at the high end. The gap is large enough that you should calculate your own RTO rate by pincode and category before deciding anything.

The third cost is the price of each failed delivery. Flexicommerce estimates a total loss of ₹100 to ₹150 per RTO order in its 2026 calculator. Toolbaz's 2026 calculator estimates that one RTO on a ₹499 order with ₹65 shipping can cost ₹130 to ₹180. Take those figures as ranges, not quotes from any courier.

A simple illustration: if you ship 1,000 COD orders a month at an average of ₹800, and 2 percent of the value goes unexplained, that is ₹16,000 a month, or about ₹1.9 lakh a year. Your own numbers will be different. The point is that a small leak repeated every month is worth a weekly check.

Cash timing matters too. Shiprocket's support page says COD amounts normally reach the bank after 8 to 9 days from delivery, and its processing standard is D+8. Early COD plans shorten that window for a fee, which means you pay to have your own money sooner. Reconciliation tells you whether the money you are owed has actually arrived.

Manual method versus automated method

How to automate it, step by step

1. Export three datasets each week. Pull your list of delivered COD orders from your order system or Shopify export. Pull the courier's remittance report with one line per AWB. On Shiprocket, that report sits under COD Reconciliation in the app, where you can apply a date filter. Pull the bank statement for the same period. Other couriers have similar reports under different menu names, so ask your account manager for the AWB-level version before you sign.

2. Match on AWB number first. AWB is the most reliable key for COD. Terra Insight's 2026 settlement guide recommends matching COD at AWB level and prepaid orders at order-ID level, because each flow has its own timing and deductions. Clean the AWB column first. Remove spaces, fix leading zeros lost in Excel, and make sure both files use the same format.

3. Match bank credits on the reference. Settlement files from Indian banks and payment gateways carry UTR numbers for NEFT, RTGS and IMPS transfers. ReconPe's 2026 review describes tools that match bank credits this way and surface COD shortfalls and aging in the same view. Use the UTR to link one courier transfer to the batch of AWB lines it covers.

4. Classify every exception. Each unmatched or short line gets one reason code: missing from batch, short amount, RTO netting, weight correction, or timing. Metaport's 2026 explainer describes the same kinds of gaps. Reason codes let you see which problem costs the most and which courier causes it.

5. Run the overdue tracker. Store the delivery date, the expected payout date and the escalation date for each exception. The expected date comes from your courier contract. Metaport's guide notes that disputes are often subject to a claim window of about 15 to 30 days, so the escalation date should be set from the contract, not from memory. A spreadsheet works for this stage, and so does a shared database view.

6. Draft the dispute, then let a person send it. A large language model can read the exception row and draft a short message that quotes the AWB, the expected amount, the amount received and the batch date. A person reviews and sends it from email to the courier's account team. WhatsApp is useful for internal alerts to the finance lead, not for legal notices to couriers.

7. Post the clean result to your books. For bank data, Zoho Books accepts CSV or XLS statement files with a date, a description, a reference number and either a signed amount or separate withdrawal and deposit columns. The reference number helps avoid duplicates on re-import. Tally users can post the matched batch as sales and receipt vouchers after the same checks. Keep the GST treatment of each sale consistent with the date of the invoice, not the date the cash arrives.

OCR fits in one place. Some couriers send remittance statements as PDF files rather than spreadsheets. An OCR step can convert those tables into rows, but you should check a sample each week, because a misread digit in an AWB number creates a false exception.

Keeping the process from breaking

Most failures come from data, not from the matching logic. Courier report formats change without notice. AWB numbers get truncated when they pass through spreadsheets. Bank PDF statements do not import directly into every accounting tool. Zoho's direct PDF import covers a set of mostly US banks, so Indian bank statements usually need to be exported as CSV first. Some co-operative, rural and small finance banks are not supported for automatic bank feeds at all.

Prevention is mostly routine:

• Lock the export format in writing with each courier and check it after any update.
• Run a control total each week: the number of delivered orders, the number of AWB lines, and the sum of amounts should agree before matching starts.
• Flag any AWB imported twice, and any RTO that also appears as a delivery.
• Keep the delivery proof for every disputed order, since the courier will ask for it.
• Review the reason codes monthly. A rising share of one code usually means a process problem, not a courier problem.

Two other checks protect you. Before you onboard a courier, get written terms on the remittance cycle, the real-time reconciliation reports, the escalation path, any penalty for late payment, and how RTO or partly paid orders appear in the report. Shipdaak's guide to its COD timeline lists these same items. Without them, the matching logic has nothing solid to match against.

When not to automate

Automation is not worth the set-up for every seller. If you ship a few dozen COD orders a month through one courier, a 30-minute weekly check of the payout against your sales sheet will usually catch the problems. Building a pipeline for that volume costs more attention than it saves.

Do not automate matching if the courier gives you only a lump-sum payout with no AWB breakdown. The tool will have nothing to match, and you will end up with a confident-looking report that proves nothing. Ask for the breakdown first, or change couriers.

There is also a limit on what automation can do. It shows you which orders are short and produces the evidence for a claim. It does not force a courier to pay. If a courier refuses, the outcome depends on your contract and your escalation path. A person still has to make that call, and some disputes will be lost.

What it takes in time

The main effort is not in building the matching step. It is in getting clean exports from every courier, agreeing on AWB formats, and deciding the escalation rules with your finance lead. Plan the first month as a test run. Process the same batches manually and automatically, then compare the exception lists. Where they differ, find out why before you trust the tool.

At scale, the saving is in staff hours. Metaport's 2026 guide estimates that automated reconciliation replaces 3 to 5 days a month that finance teams spend on this work when volumes are high. For a smaller business, the saving is less about hours and more about catching money that would otherwise go unnoticed. Once the process runs, the weekly task shrinks to reviewing the exception list, approving dispute messages and checking the control totals.