My Candle Looks Perfect and Burns Cleanly but Customers Say the Scent Is Weak — Which Formulation Variable Should I Change Without Disturbing the Burn?

Is it weak, or weak where they burn it?

Before changing anything, place the complaint. A structured intake for one customer's report (burn length, room size, fan and AC, wick trim, other scents in the home) is the customer-says-no-scent post; a spread of "too strong", "too weak" and "perfect" is the conflicting-feedback post. Here the reports agree, the candle burns cleanly, and you need to know whether the fix lives in the jar or in how it reaches and is used by the customer. The table ranks the usual patterns; the order is editorial judgement, and CSI publishes no complaint statistics.

Pattern 5 sends you out of this post: if the pool is short, the burn is not clean in the customer's hands, and the multi-problem hub and the customer-tunnelling post take over. Pattern 4 is an outlier scent, handled by the one-weak-scent post. Patterns 1 to 3 are this page, and patterns 1 and 2 cost nothing to fix.

The burn-neutral lever ranking

Five levers can raise the scent a customer notices. They differ in how much they disturb a burn that works, and in what they add to each candle. The ranking below is editorial judgement; the costs are worked examples at CSI live pricing on 26 September 2026, excluding labour, freight, packaging and GST. Levers that raise throw at the cost of a new wick, wax or bigger pool (maximum-throw optimisation) belong to the maximum-hot-throw post; this page only uses levers that keep the burn where it is.

Lever 1: cure age at sale

Soy commonly needs a cure before hot throw is judged; CSI oil pages that state one say 48–72 hours for soy (25 of 102 in the text we captured), and many makers judge at day 7 and day 14 (common practice, not a rule). A candle dispatched on day three reaches its customer a few days later, still well short of day 14 (CSI publishes no delivery times; use your own courier's), and cure is time since pour wherever the candle sits. How curing works is the cure post; whether a candle keeps curing in transit is the ship-immediately post. The lever here is simple: hold one batch to day 14 and see whether the complaints stop. It costs stock-holding time, which before a festival is a real cost; it costs no materials and cannot disturb the burn.

Lever 2: use instructions

A clean-burning candle can easily be judged on a 45-minute first burn in a large room with the fan on. A care sticker or card that says "burn for long enough that the melt pool reaches the edge; trim the wick before lighting; for the fullest scent, burn in a closed room" moves the customer, not the formula. It fixes nothing if the candle is genuinely faint, and it is honest only if your own test shows the candle performs when used that way. The first-level version of this problem, strong near the jar and faint across the room, is the doesn't-fill-the-room post.

Lever 3: a one-point load step

If the candle is honestly faint at 3 m in a customer-sized room at day 14, the cheapest formula change is one point more oil: Roasted Coffee from 12 g to 13.5 g per 150 g candle (138 g to 136.5 g of wax). Its page states 6–10% for candles; CSI 464's page states up to 10%; 9% of total is 9.89% of wax weight, inside both readings of the ceiling. It adds ₹5.85 of oil a candle at the 100 g price, and a kilogram of oil makes 74 candles instead of 83. It is the lever most likely to disturb the burn of the ones that keep the wick, because more oil in the same wick tends toward sweating and a struggling flame; that is what the guardrails catch. More oil does not raise throw in a straight line, and at the top of a range it can lower it.

Lever 4: a different oil at the same load

Oils differ in how they throw in a given wax, wick and jar, and a second oil at the same load keeps the formula's arithmetic unchanged. But it has two costs people underrate. First, cash: in the worked example the swap oil costs ₹60.96 more per candle at the test sizes. Compare oils by cost per finished candle, not per bottle (the cost-per-finished-candle post). Second, identity: a different oil is a different scent, and "stronger" is not worth much if customers bought the old one. And an oil that smells similar can still burn differently with the same wick (the new-supplier-oil post), which is why this arm is guardrailed too.

Lever 5: the jar opening

A narrow mouth over a wider body can hold scent in the jar (a tendency), and a vessel swap is the test for it (3209). But a new jar means a new internal diameter, and a new diameter means a new wick bracket; it is the one lever here that cannot keep the burn still. If your jar's mouth is as wide as its body, skip it.

The guardrail test

The test measures levers 1, 3 and 4 in one pour day; lever 2 is tested in the same session by how you burn, and lever 5 is left alone. All candles are poured the same day in the same wax, with your current wick, jar, pour temperature and cooling, weighed on total weight. CSI 464 in the 150 g jar pours at 80–85°C (the jar page's 85°C governs).

Costs are worked examples and exclude labour, freight, packaging and GST. If you make this candle already, the new cash is ₹449.00 for the swap oil; from nothing it is ₹2,349.40. Two 50 g bottles of Roasted Coffee (₹460.00) would cost ₹70.00 more than the 100 g for the same test. The 1 kg of wax covers the 963 g with a little spare, not a repeat round.

Reading the test

Notice the order in the first two rows. If the sale-age candle is weaker than the day-14 candle, you may not need a formula change at all; and if the day-14 candle is clear at 3 m in a customer-sized room, the formula is fine and the gap is use. Only when the day-14 control is faint in a realistic room do the formula levers come in, and then the cheaper one that passes wins. In our judgement, many "weak" clean-burning candles are fixed in the first two rows.

Spending that won't close the gap

In plain terms: the most likely fixes for a clean-burning, "weak" candle are holding stock longer and telling customers how to burn it, and we sell nothing for either beyond a roll of stickers you may not need. We would rather you found that out before an oil order than after.

Oils, stickers and the jar

Four items: the worked-example oil and the swap oil, the care stickers for the use-instructions lever, and the jar, included to explain why it is the lever to leave alone. Chosen on page statements, price per unit and stock on 26 September 2026; CSI has no measured throw data, so nothing is ranked on throw. Oil tables show every size, with candle counts at 8% (12 g) and 9% (13.5 g).

Frequently asked questions