The Money That Leaves After the Sale

The money that leaves after a jewellery sale falls into four piles: fees the platform charges you, taxes it collects out of the sale for the government, and recoveries it takes back when a delivered order goes wrong. The fourth pile never appears as a line anywhere. Each one answers to a different lever.

This section is about the distance between the order value on your dashboard and the credit in your bank. Three guides sit under it: one reads the settlement report, one packs the parcel so a claim can be won, one builds the claim after a return arrives wrong. This page holds the part above them — which pile you are actually looking at, what chasing it costs you, and which of the four is worth your Sunday.

Start with one delivered order, because everything below reads more easily once you know what a single piece leaves you after fees and tax.

None of this is tax advice. Where a rule is named, it is named with its section and handed to the accountant who signs your return.

What are the four kinds of money that leave after a sale?

Sort them by who decided the amount and they separate themselves. A fee was set by a rate card you agreed to months ago. A tax was set by law, and the platform is holding it, on its way to a department. A recovery was caused by something that happened after dispatch. The fourth pile has no author at all: nobody bills you, nobody records it, and it is the one most sellers never count.

The order matters as much as the names. Three of the four are settled before the parcel leaves your table, and the fourth is settled by what you do in the weeks after it comes back.

Telling one line on a report from another is a job in itself, and the guide to a short settlement takes a report apart line by line. Read it once with a real report open, and the sorting below becomes mechanical.

Order ke paise kitne din mein aate hain?

Longer than the delivery date, and the exact number is published in your own seller panel, where it can change. What does not change is the order of events, and every stage in it holds your money a little longer.

A piece is bought from your supplier or made at the bench. It sits as stock. It sells, it travels, it is delivered. The buyer’s return window then runs, the order becomes eligible for settlement, the cycle closes, and a payout is released. The tax lines surface later still, on a portal, once the operator has filed its monthly statement.

Now count the days for yourself, on one real piece. From the day you paid the karigar or the supplier to the day the credit for that piece landed, that stretch is your float. The working capital you need is the value of every order standing between those two dates at once, counted on one day. Where that stretch runs past a month, you are financing more than a month of sales at any one time.

Two things stretch it. A COD order puts no money into the system until it is delivered, so a COD-heavy catalogue finances its own stock for longer. And a return restarts the clock instead of reversing it: the piece comes back, is checked, is listed again, and waits for a second buyer.

What can you price into the listing before anything goes wrong?

Two of the four piles can be worked out before a single order arrives. Fees come out to the rupee from your own rate card. Recoveries come out only as a likelihood: your return rate times what one return costs you, spread across the orders that stay delivered.

Work it backwards from the money you want in hand, one piece at a time, on the Reverse Price Calculator. Two lines behave differently as the price moves. Commission and GST on fees scale with the price; the fixed or closing fee, the courier and the packing do not. That is why a light imitation jhumka carries the flat lines worst, and why the honest answer for some pieces is a higher price band or a different piece.

Remember whose money the tax is. A finished piece sold ready-made is one supply, taxed at 3% on the whole value, and on a marketplace the listed price a buyer pays already includes it. Part of that order value was never yours to keep. Where the making is billed on its own, as work done on metal that is not yours, 5% is the figure to look at instead, and which rate a jewellery bill actually carries is a question with a guide of its own.

There is a ceiling on this, and it is the thing sellers discover late. Every rupee you add to cover returns sits on the price a buyer compares against forty other jhumkas. Add enough and the piece stops selling, which spreads the same return cost over fewer delivered orders and asks for another rupee. Past that ceiling, adding rupees stops working, and the lever moves to the parcel, the listing photograph and the size chart.

Which losses never show up as a line on any report?

Four of them, and they share one feature: no document anywhere names them, so nothing catches them for you.

The first is stock standing still. A piece out on a journey and back again is out of your stock for weeks, and it cannot be sold to anyone else while it travels. In a festival window those weeks are the whole season for that piece.

The second is the piece that comes back unsellable. A tarnished chain, a scratched plating, an opened jump ring, a stone that moved. Some of that goes back to the bench as a small karigar bill; some of it becomes a piece you quietly discount to move.

The third is your own hours: the file you assemble, the calls, the evening spent reading a report. The fourth is what those weeks do to a rate-linked piece, where the metal it was priced on has moved by the time it is listed again.

That audit prices the blocked weeks and the damaged piece for you. Two lines stay yours to judge: your hours, and the discount you eventually take on a piece that has been out and back. That is an honest limit on our own bench. A calculator prices what you can name, and in this pile the largest figures are usually the ones nobody has written down.

When is a small recovery worth filing a claim for?

Price the claim against your own evening first. Say a freight recovery comes to ₹120 and arguing it takes an hour: that is a poor trade even on the days you win. The same line appearing on eleven parcels of one piece is a different thing altogether. That is a listing telling you its declared weight or its box size is wrong, and the fix there is worth ten claims.

So read the pattern before the instance. Sort a month of recoveries by piece and by reason, and most of them collapse into two or three causes you can close for good. What is left after that is the short list genuinely worth filing.

Recoveries are the pile where most errors live, because a recovery is a record of an event, and events get recorded wrongly. Commission at your agreed rate, the GST on those fees and both tax lines are the arrangement itself. The settlement guide sets out one exception: a fee charged at a band your rate card does not carry is an error like any other.

When a return arrives with the wrong piece inside, or empty, a desk opens the file and reads three records and very little else. The honest summary of that guide is that the claim was decided a week before it was filed.

Keep your own record of what you filed and what came back. After a dozen claims you hold the only success rate that describes your catalogue, which is worth more than any figure quoted in a seller group. Some of what you are owed will not come back, and knowing which kind is worth as much as filing well.

What has to exist before the parcel leaves for any of this to work?

One order number, five records, one name on all of them. The order and its invoice, the stock code on the piece and on the outer, the weight the courier recorded at pickup, the recording made at dispatch, and the settlement line that order finally lands on. Name every file with the order number and the day, and the file assembles itself.

A recording nobody can find loses a claim as completely as a recording nobody made. A phone holding four thousand unnamed clips and a folder named by order number are two different businesses.

The records are then asked for by three different people on three different clocks. The platform wants order-level evidence inside the window its panel publishes. The carrier wants the booking and the manifest, on its own terms. Your accountant wants invoices and settlement reports at the year end, and section 36 of the CGST Act sets how long a registered person keeps books and records, with the income-tax side running its own period. Your accountant reads both against your registration and signs the return; this page does not.

What goes into the box, where the seal sits and which weight you write down settle most of this before the parcel moves, and the layers are worked through in the guide to packing a piece so the claim is winnable. Everything on that list is cheap, and all of it costs less than one return.

Does any of this change on a sale you make yourself?

The piles change size, and one of them nearly disappears. On a sale through Instagram, WhatsApp or your own site there is no marketplace commission, and nobody is collecting tax out of the payment on its way to you. The tax on that sale is due on your own return in the normal way, and your accountant confirms that against your registration. What is left of the fee pile is what your payment gateway charges, and a prepaid order reaches your bank on that gateway’s own settlement cycle, published in its dashboard. A COD parcel you booked yourself is remitted on the carrier’s cycle.

Everything the platform was doing is now yours. You book the courier, so its terms are the only cover the parcel has, and what each carrier will actually cover on a jewellery parcel is worth reading before the first one goes. You write the return policy, because no window exists until you publish one. You are the dispute desk. Where the money arrived through a payment gateway, a buyer can still raise a chargeback there, and that runs on the gateway’s rules and the gateway’s clock.

Then there is the money that leaves months later, which no marketplace guide covers at all. A customer comes back with a broken clasp, a loose stone, or a piece she wants exchanged. No window applies and no desk decides it; you do. The repair is a karigar bill, and on an exchange the metal comes back at today’s bhav against the rate printed on her bill, with the difference sitting on your side of the counter.

Write that policy down before it is needed — what you take back, in what condition, within how long, and what an exchange is valued at. Print it on the bill or the tag. A policy invented at the counter with a customer waiting is always the expensive one.

What is the one thing to do on the same day every month?

Write four numbers on one sheet, in four columns, one row a month. The delivered order value. Everything deducted, as a share of it. Your return rate, split by reason — courier return, buyer return, wrong or empty. And what one return costs you.

Each number answers a different question. The share tells you whether your price band is carrying the fees. The split tells you which lever to pull: courier returns are a pin-code and payment-mode problem, buyer returns are a photograph, a size chart or a description problem, and wrong or empty returns are a packing and evidence problem. The cost per return sets what your packing table is worth spending on.

One month of these numbers is noise. Three months is a decision, and a year of them describes your catalogue better than any single month of it can.

Put your last three settlement reports on the table tonight and fill in those three rows. The month with the worst share is the one to explain first, and the explanation is usually sitting in a run of identical small recoveries you have been paying without reading.