From Costed Sample To First Production Run

The sample has arrived and it is good. The supplier has quoted a per unit price. The instinct now is to raise the purchase order and get the launch date locked. Slow down. The distance between a costed sample and a first production run is where most launch calendars break, and where most cost sheets quietly stop being true.

What a costed sample actually proves

A sample proves the supplier can make the thing once. It was probably made with hand picked material, by an experienced operator, off the line, with no time pressure and with someone watching. That is genuinely useful. It confirms the form, the fit, the finish and the function are achievable in that factory. It confirms the quote is a real number at some volume, on some day, under some assumptions.

That is the full list. Everything else you believe about the sample is an assumption you have not tested yet.

What the sample does not prove

It does not prove consistency. Unit to unit variation only appears at volume. It does not prove material availability, because the fabric, resin, glass or fragrance in the sample may have come from a partial lot the supplier already had. It does not prove the price holds at your quantity, your payment terms, or at input prices three months from now. It does not prove the packaging survives a Delhi to Chennai transit in June. And it does not prove the regulatory labelling fits on the pack, which is a surprisingly common reason for a redesign after everything else is signed off.

Ask three questions before you treat a sample as settled. Was this made on the production line or on a bench. What changes when you run it at my quantity. How long is this quote valid, and what input price is it built on.

Tooling and moulds change who has leverage

The moment your product needs a custom mould, die, cutting tool, printing cylinder or dedicated bottle, the commercial relationship changes shape. Before tooling, you can walk. After tooling, walking means abandoning an asset and restarting a development cycle of several weeks or months.

Three things belong in writing before you pay a tooling deposit. Ownership, stated explicitly, with the tool marked and photographed with an identification number. Location, with the right to inspect. And transfer, meaning what happens to the tool if you move production elsewhere, on what notice and at what cost. Tooling amounts vary widely by category, cavity count and complexity, so the number matters less than the clause. A tool you paid for but cannot move is a switching cost you did not put in the cost sheet.

Where the design allows it, prefer standard tooling for the first run. A stock bottle with a custom label costs less, moves faster and keeps you free. Earn your custom mould with proven volume.

MOQ negotiation and the real cost of a high minimum

A minimum order quantity is not a price term. It is a cash and risk position, so evaluate it as one. Convert the MOQ into months of cover at your honest forecast, not your optimistic one. If the answer is more than a couple of quarters for a new product with no sales history, the low per unit price you negotiated is not a saving. It is capital locked in a warehouse, plus storage, plus obsolescence risk, plus the near certain discounting you will do later to clear it.

There are levers other than begging for a smaller number. Place one purchase order with phased call offs, so the supplier gets the volume commitment and you get staggered cash outflow. Accept a higher per unit price for a genuinely short first run and treat the premium as the cost of information. Standardise components across your range so the minimum applies at component level rather than finished goods level, which is the single most effective structural fix. And ask what the minimum is driven by, because a minimum set by a fabric roll, a resin batch or a printing plate has a different negotiating surface than one set by the sales team.

Agree the quality standard in writing before the PO

Quality arguments after production are unwinnable, because both sides are arguing from memory about a standard nobody wrote down. Fix this before the order.

Sign a golden sample, two copies, one held by each party, sealed and dated. Write a defect classification splitting critical, major and minor defects with examples for your category. Name the inspection standard and sampling level you will use, the inspection point, who conducts it and who pays. State what happens on failure, with the options ranked: rework, price adjustment, or rejection, and who bears freight in each case. Add the packaging and labelling specification, since a large share of real world rejections are labelling failures rather than product failures. Keep retention samples from every run.

Lead time is what ruins launch calendars

Founders count lead time as production time. The real clock runs from purchase order to sellable, and it has far more segments: component procurement, production, inspection, rework if any, packing, dispatch, transit, inbound at the fulfilment centre, goods receipt and put away, listing activation, and the ad ramp that follows. Any one of those can add a week.

Then add the Indian calendar. Pre festive congestion at plants and in transport. Monsoon transit delays on certain lanes. Plant shutdowns around Diwali. Quick commerce and marketplace inbound appointment slots that tighten exactly when everyone else is also shipping.

Work backwards from the launch date, not forwards from the purchase order. Put the required PO date on the calendar with the buffer already inside it, and treat that date as the commitment. A launch date is a supply chain decision wearing a marketing costume.

The sequence that works

Costed sample approved. Regulatory and labelling confirmed against the actual artwork. Quality standard and golden sample signed. Tooling ownership documented if tooling is involved. MOQ converted into months of cover and negotiated on structure rather than sentiment. Pilot run produced on the real line and inspected. Pilot sold through one channel. Only then the scale purchase order, with a lead time counted to sellable and a buffer nobody is allowed to spend early.