I Ordered 5,000 Wicks for Production — How Many Should I Visually Inspect Before Accepting the Shipment?

There is no single correct number — but there is a sensible one

When 5,000 wicks arrive, the temptation is either to open one packet, glance at it and start pouring, or to feel that anything short of checking every piece is careless. Neither is good goods-in practice. A visual inspection of a sample exists to answer one question before production: is the defect level in this delivery low enough that you can use it, or high enough that it needs sorting, a supplier conversation or rejection?

Formal acceptance-sampling standards exist for exactly this kind of decision, and larger manufacturers use them to set sample sizes and acceptance numbers by lot size. You do not need to adopt one to run a sound check. What you need is a sample large enough to catch a problem that affects a meaningful share of the lot, drawn from enough places that one bad packet cannot hide, and a written rule for what you will do with the result. Everything below is a suggested starting point to validate against your own supplier history, not a standard and not CSI policy.

Two facts shape the number. First, visual inspection is quick, a few seconds per wick once you know what you are looking for, so a sample in the low hundreds is affordable for a small business. Second, the cost of a missed defect is not the wick. It is the jar, the wax, the fragrance, the labour and possibly a customer. That asymmetry argues for a larger visual sample than instinct suggests.

A suggested starting-point sample for 5,000 wicks

A practical plan for a first delivery from a supplier, or a first delivery of a new wick type, is to inspect around 125 wicks drawn evenly from every carton or pack in the shipment. For 5,000 pieces that is 2.5% of the lot. Once you have three or four clean deliveries on record, you might reduce to around 50 per delivery; if a delivery shows problems, you tighten the next one. Treat these numbers as a place to start and adjust them as your records build.

Spread matters as much as size. Wicks are made and packed in runs, and a machine that drifted for an hour can leave a few hundred poor pieces grouped together. Twenty-five wicks from each of five packs will find that cluster; 125 wicks from the first pack you open will not.

What to look for, and how serious each fault is

A sample is only useful if two people inspecting it would reach the same result. Write down what you are looking for and grade each fault before you start. The usual approach in manufacturing QC is three grades: critical, major and minor. Critical faults create a safety risk. Major faults are likely to change how the candle burns or looks to a customer. Minor faults are cosmetic or easily corrected at assembly.

Keep a small reference card: two or three wicks from the lot you last approved, taped down and labelled. Compare suspect pieces against it under the same light. Your eye adjusts quickly, and without a physical reference you will start to accept gradual drift as normal.

Worked example: 5,000 Eco Candle Wicks

Suppose the 5,000 wicks are CSI Eco Candle Wicks, Thin (C1), bought as five packs of 1,000 at ₹5,900.00 each. The delivery costs 5 × ₹5,900.00 = ₹29,500.00, and each wick costs ₹5,900.00 ÷ 1,000 = ₹5.90. The product page describes natural fibres of cotton and paper and suits them to soy; it does not publish dimensions, tab size or a burn chart, so your retained wicks from the previous lot are the only reference you have.

You draw 25 wicks from each pack, 125 in all. At an unhurried ten seconds per wick against the reference card, that is 125 × 10 = 1,250 seconds, or roughly 21 minutes including laying out and recording. Most of the sample returns to stock afterwards, because visual inspection does not consume the wick. Only pieces you pull for burning are lost.

Now the other side. If one pack in five were faulty and you had only inspected the first pack you opened, there would be a four-in-five chance that pack was one of the good ones. You would release all 5,000 wicks, and 1,000 of them, a fifth of the delivery, would go into jars before anyone noticed. The wicks are cheap to replace; 1,000 finished candles are not. Twenty-one minutes spread across all five packs removes that blind spot.

Turning the count into a goods-in decision

The inspection is finished only when the delivery carries a status label. Using the suggested rules above, the outcomes are straightforward. If the sample shows no critical faults and major faults within your limit, the lot is accepted for visual quality and moves to whatever burn check your system requires for a new lot. If major faults exceed the limit but are concentrated in one pack, that pack goes into quarantine and the rest can be released. If faults are spread across all packs, the whole delivery stays in quarantine, and you decide whether to sort it piece by piece or return it.

Visual inspection has limits. It tells you that the wicks look like the ones you approved. It does not prove they burn the same, and a braid that looks identical can still perform differently if its treatment changed. That is why a visual pass is the first gate, not the last. Pair it with a short comparison burn against a retained old-lot wick in your production jar before full release, and escalate to a complete burn test whenever the visual check turns up something unexpected. A delivery that fails your inspection stays in quarantine whoever supplied it, including us.

Where this page fits

CSI already has guides that cover parts of this question. This page covers the goods-in check for the question above; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions