I'm Importing Candle Jars for My Business — What Should I Test Before Approving Thousands of Containers for Production?

Stage 1: write the spec before you ask for samples

Importing changes one thing above all: distance. A domestic supplier can swap a carton next week; an overseas lot that fails on arrival may take months to replace, and the argument about whether it failed will be conducted by email against whatever was written down before the order. So write it down first. A one-page spec sheet, sent with your sample request, turns "the jars look fine" into numbers both sides can check.

Take your tolerances from your own wick and fill tests, not from the supplier's defaults. If your candle burns well across a narrow band of inside diameters, that band is the tolerance. How to measure an inside diameter properly, at pool height, is in external vs internal diameter; no CSI listing, incidentally, states an inside diameter (those that give a size give an outer, base or unspecified diameter or a width), which is why we tell our own customers to measure too.

Stage 2 and 3: pre-PO samples and the golden sample

Ask for samples from the production line, from the mould and glass that will make your order, and ask the supplier to say so in writing. Showroom samples are often hand-picked. Twenty-four is our judgement for one design: enough to measure a meaningful spread, burn a handful to the end and still keep some back. Budget a sample charge and international courier; we use ₹3,000.00 as a worked example, and your quote replaces it.

What to ask the supplier for in writing

Documents make disputes shorter and tests more targeted. None of them replaces a burn test on your candle. Ask for each in writing, keep them with the golden sample, and read them critically: what was tested, on which jars, by what method.

If a supplier cites a named test standard, ask for the report and check that the jars tested were your design, from the relevant period, and that the test resembles a candle burning down in a jar; a lab thermal-shock test is not a burn-down, and a burn-down on someone else's candle is not a test of yours. We deliberately do not list standards here: which, if any, applies to your product and market is a question for the report's authors and your own advisers. What a supplier's "candle-safe" claim covers, and the independent tests to set against it, is the subject of the supplier-claims guide.

Stages 5 to 7: inspecting each lot on arrival

A lot is one delivery of one design from one production run, identified by the code on its cartons. If a shipment mixes lot codes, inspect each lot separately; if it has none, treat the whole delivery as one lot and ask for codes next time.

Stage 5: arrival, before you open anything

Count the cartons against the packing list, photograph any crushed, wet or rattling carton before you open it, and open those first. Transit breakage is a carrier and packing question, and any claim is easier to make with photographs taken on the day. Then choose the cartons you will sample at random, by carton number, not from the top of the stack: the cartons nearest the door were packed last or loaded last, and neither tells you about the middle of the run.

Stage 6: the three samples

From the randomly chosen cartons, take jars from different positions. Inspect about 80 by eye, measure about 20 of those, and burn about 5 of the measured ones to the stop-line. These numbers are our planning judgement for lots of about 1,000 to 3,000 jars of one design. Formal acceptance-sampling tables exist, and inspection agencies use them; they share one idea worth borrowing, that the sample grows far more slowly than the lot. For a lot of about 5,000 to 10,000 we would take roughly half as many again at each level, not three times as many.

The burn tests are the slow part: a burn-down to the stop-line takes several sessions per jar, and each lot waits for them. Start them the day the lot arrives, and meanwhile release nothing for production. If you cannot wait, pour only a small first run from the lot and hold its candles until the burn tests pass. A chip at the rim is a critical defect in our classes for a reason: it is a common crack origin, a cut risk and a lid-fit problem (the chipped-rim reject policy).

Stage 7: traceability and the next delivery

Write the jar lot code on every production batch record, and retain five unused jars per lot (worth ₹275.00 at the illustrative price). When a complaint arrives in six months, you can go from the candle to its batch, from the batch to its jar lot, and from the lot to its retained jars and test log. Then the next delivery starts again at stage 5. If a lot fails, the doubled sample and the classification table tell you whether to sort by band (sorting jars with different inside diameters) or reject; random breakage that starts later in production is a different diagnosis (separating jar quality from pour temperature).

What to buy for your approval bench

CSI does not import jars on your behalf and cannot vouch for anyone else's glass. What we sell is the bench: the constants that make one jar comparable with another. Five items, chosen on page wording, price and stock on 26 September 2026; none comes with burn or thermal-shock data, including our jar.

What the programme costs, against what it protects

Every import price here is an illustrative input, not a quote: a landed cost of ₹55.00 per jar, a 3,000-jar lot, and ₹3,000.00 for samples and courier. Replace them with your own. CSI prices are live on 26 September 2026. Costs exclude your oil, labour, freight on CSI items and GST.

Each later lot costs only its arrival burn tests, about ₹672.31 in this example (0.4% of the lot's value), plus inspection time; the jars inspected and measured go back into stock unless they fail. If you already own a thermometer and scale, the pre-PO materials are ₹1,168.20 rather than ₹1,876.20.

What it protects

Worked example: one 150 g candle in the imported jar costs about ₹197.11 in materials: the jar at ₹55.00, 138 g of CSI 464 at the 10 kg price (₹70.02), 12 g of Lavender Fragrance Oil at the 1 kg price (₹62.65) and a C2 wick (₹9.44), excluding labour, freight, packaging and GST. If a faulty lot is found only after 500 candles have been poured, those candles represent about ₹98,556.60 of materials, roughly 17.8 times the whole first-lot programme, before you count labour, customer refunds or the candles already sold. Lavender is only a costing example; its page gives a candle range of "Recommended fragrance load 6-10% for candles", every size in stock on 26 September 2026.

Frequently asked questions