Nobody plans for claims, so nobody staffs for them. Then a quarter of your working capital is sitting in cartons waiting for someone to approve a credit note.
"Returns" gets used for three situations that share almost nothing operationally.
Damages are goods made unsaleable by handling — crushed in transit, leaked, punctured. The question is who was holding it when it happened, which decides whether the claim goes to the carrier, the company, or nowhere.
Expiry and near-expiry is stock that ran out of shelf life while unsold. Nothing went wrong physically. The claim is against whatever near-expiry policy your principal operates, and those policies vary widely.
Saleable returns are goods coming back in good condition — over-ordering, a scheme that did not move, a retailer closing down. Often the most negotiable and the least documented.
Running all three through one register is a reliable way to make all three settle slowly, because each needs different evidence and goes to a different approver. Separating them is usually the single cheapest improvement available.
A claim moves through roughly six stages: goods identified at the outlet, collected and brought back, logged and inspected, claim raised with evidence, approved by the principal, and credit note issued.
It stalls at stage four almost every time, and for the same reason: the evidence was assembled after the fact.
At the outlet, the salesman knows exactly what happened — which batch, what condition, when. A week later, back at the office, someone is trying to raise a claim from a line in a notebook that says "2 cases damaged". The claim goes up thin, comes back queried, and each round trip is days during which the stock sits in your godown.
The fix is not a better claims department. It is capturing the evidence at stage one, when the person who knows is standing in front of the goods. Batch number, quantity, condition, photograph, and an acknowledgement from whoever handed it over. Everything after that is administration.
Numbers to re-run with your own, not a benchmark.
Take 60,000 rupees of stock coming back in a month, settling in 75 days instead of 30.
The 45 extra days are 45 days of your money held in cartons that earn nothing. If your working capital costs 14% a year, that is roughly 60,000 × 0.14 × 45 ÷ 365, or about 1,035 rupees on that month alone. Run it every month and it is over 12,000 a year — for no reason except that claims are raised slowly. The Working Capital Calculator puts your own numbers through the same arithmetic.
But the carrying cost is the smaller half. The larger half is the claims that fail entirely because the evidence never existed, and get written off. A written-off claim is not a delay, it is the full value gone, and it comes straight out of the margin the Profit Margin Calculator says you made on the original sale.
There is a third cost that never appears in any register: the retailer whose claim sat unresolved for two months and who now orders defensively, or from someone else.
Near-expiry has something damages do not — a window in which the outcome is still under your control.
Stock at six months of remaining life can usually be sold through with a push. At two months, it needs a scheme and a retailer willing to move it fast. At two weeks, the only question is whether the claim gets filed cleanly.
Which means the decision that matters happens well before the stock becomes a claim, and it depends on knowing what is ageing. That is a stock-ageing report by batch, run often enough to act on — monthly at minimum, and it is not an exotic report. Any billing system tracking batch and expiry can produce it; most distributors simply never ask it to.
The decision at each stage:
That third band is where money is won and lost. A discount that clears the stock is almost always cheaper than a claim that might not settle, and considerably cheaper than a write-off. But it only works if you knew a month earlier, which brings it back to the ageing report.
The claims that settle fast share a pattern. Captured at the point of collection, in a form nobody has to reconstruct:
If the ordering conversation with the retailer already happens on WhatsApp, the collection acknowledgement can happen in the same thread — a photograph and a batch number sent from the outlet takes seconds and lands in a place with a timestamp on it.
A return against a tax invoice is normally handled with a credit note under section 34 of the CGST Act, which adjusts both the taxable value and the tax originally charged.
The timing constraint is the part that catches people. The adjustment against a credit note has an outer limit — broadly the 30th of November following the end of the financial year in which the original supply was made, or the date of filing the relevant annual return, whichever is earlier. A claim from March that drifts unresolved for eight months can cross that line, at which point the commercial settlement may still happen but the tax adjustment does not.
That is another reason slow claims cost more than the carrying cost suggests, and it is worth reading alongside GST e-invoicing in practice, since credit notes flow through the same reporting machinery. Treatment depends on the specific facts of each case, so confirm your own handling with your CA rather than working from a general description.
If claims are currently a monthly source of unpleasant surprises, the sequence that pays off fastest is not a new system.
Split the register into three. Damages, expiry, saleable. Different evidence, different approver, different timeline. Doing this alone tends to expose which category is actually the problem, and it is often not the one people assume.
Move evidence capture to the point of collection. Photograph, batch, quantity, acknowledgement — at the outlet, by the person who is already there.
Run a stock ageing report monthly and act on the three-to-six month band while action is still cheap.
Track days-to-settle as a number somebody owns. A claim cycle nobody measures drifts, and it drifts in one direction. Once it is a number on a report, it tends to fix itself, in much the same way that outstanding recovery does once someone is accountable for it — the argument made at length in building an udhaar recovery system.
The questions that come up when a distributor tries to get claims under control.
A damage claim covers stock made unsaleable by handling — crushed cartons, leaked packs, transit breakage — and is usually raised close to when it happened, with the carrier or the company depending on where it occurred. An expiry claim covers stock that passed or approached its shelf-life date while unsold, and it is settled against whatever near-expiry policy the principal company operates. They follow different approval paths and different evidence requirements, and mixing them in one register is a common reason claims stall.
A return against a tax invoice is normally handled by issuing a credit note under section 34 of the CGST Act, which adjusts the taxable value and the tax already charged. There is a time limit on claiming the tax adjustment against a credit note — broadly, the 30th of November following the end of the financial year of the original supply, or the date of filing the relevant annual return, whichever comes first. Because the treatment depends on the specific facts, confirm the handling of your own cases with your CA.
Almost always because the evidence was assembled after the fact rather than at the moment the goods came back. A claim raised weeks later with no photograph, no batch number and no signed acknowledgement from the person who collected it will be queried, and each query round trip adds days while the stock sits in your godown occupying working capital.
It depends on whether your principal will settle it, and that is a question to answer before the situation arises rather than during it. Accepting returns your company will not reimburse converts a retailer relationship problem into a direct write-off on your books, so the workable position is usually a clear, published cut-off that matches the policy you can actually claim against.
The FlowKartAI team builds WhatsApp-native ordering for Indian B2B distributors and the kirana stores they serve. We write about distribution economics, GST compliance, and the practical side of putting AI in front of retailers who have never opened an app.
FlowKartAI parses natural language WhatsApp messages into ERP-ready orders in seconds.
Try Free Demo