I'm Making 50 × 200 g Candles at 8% Fragrance Load — How Many Grams of Fragrance Oil Should I Weigh for the Entire Batch?

The 800 g figure, line by line

Every figure on this page starts from the same convention CSI uses across every guide: fragrance load is a share of the finished candle's weight. If a candle weighs 200 g when it is poured and set, and the load is 8%, then 16 g of that 200 g is fragrance oil and 184 g is wax. The formula is plain multiplication: oil = finished weight × load, and wax = finished weight minus the oil. Nothing else in the calculation depends on which wax, which oil or which jar you use.

The check row matters more than it looks. If your oil and wax do not add back up to the finished weight, one of the two figures is wrong, and the most common cause is the convention mix-up explained in the next section. Put the check on your batch sheet and let it catch the error before the wax is melted, not after the candles are cured and weak.

Notice what the 800 g does not include yet. It contains no allowance for oil that stays in the pouring jug, no test candles and no spare. It is the net quantity that ends up inside fifty saleable candles, so treat it as the floor of your purchase, not the purchase. Fifty candles is also a small enough run that the allowances are large relative to the batch, and that is why they get their own section below. For the wax side of the same fifty candles, including how many kilograms to order, see the 50 × 200 g wax calculation.

The load convention trap

Many recipes and calculators online say "add 8% fragrance to your wax". Read literally, that means oil = wax × 0.08. It is a genuine convention, and some suppliers use it, but it is not the same load as 8% of the finished candle. Adding 8% of the wax weight always produces a true load below 8%, because the oil is then being measured against a smaller base. On a 50-candle batch the gap is large enough to matter.

The third row is the most expensive mistake. A maker who treats "50 × 200 g" as 10 kg of wax buys 800 g too much wax, then adds 800 g of oil on top and ends up with 10,800 g of candle mix at 7.41%. The jars are filled with a weaker candle, and there is enough left over for four more candles nobody ordered. The fourth row is the fix when you already have the wax weighed out. To hit a true 8% on a fixed amount of wax, divide by 0.92 rather than multiplying by 0.08: oil = wax × 0.08 ÷ 0.92. That gives exactly 16 g for 184 g of wax, and exactly 800 g for 9,200 g.

If your wax supplier or a recipe you trust states its figure as a percentage of wax, convert it before you weigh. A "10% of wax" recipe is 9.09% of the finished candle, and "6% of wax" is 5.66%. Once every figure is in one convention, your test results, your production batches and your reorders all line up. Ordering from the finished weight rather than the wax weight takes this further for a 10% run, where the gap is wider.

The allowance stack: 800 g to 872 g

Between the 800 g in the candles and the oil you should own sit two allowances. Both are working assumptions, not rules. The figures below are sensible starting points for a first 50-candle run, and you should replace them with your own once you have made a few batches and weighed what is left in the jug.

Why 3% for pour loss. Oil does not evaporate in any measurable way in a short melt, but it is mixed into wax, and some of that mix stays behind: a film on the pot walls, a skin in the pouring jug, a drip on the stirrer, a candle topped up after sinking. Makers commonly plan 3–5% extra for pour loss and test burns. Three percent is the lower end, appropriate for a small, careful batch. If you pour from a large pot through a funnel, or top up every candle after the first set, use 5%. That would make step 2 40 g and the bench total 888 g, which still sits inside the same 900 g of bottles.

Why three test candles. One candle tells you almost nothing, because you cannot tell a wick problem from a load problem from a pouring problem. Three candles poured from the same melt let you burn one at the curing checkpoint you normally use, keep one for a longer burn and keep one as a retained reference for the batch. At 16 g each, they cost 48 g of oil. If you have already approved this exact oil, wax, wick and jar combination, you can drop them. If you have not, the testing allowance for a 100-candle run shows how to size a proper test round.

The order of steps matters. The pour allowance is a percentage of what you are melting for production, so it applies to the 800 g, not to the test candles, which you usually pour from a small separate melt. The test candles are a fixed number of units, not a percentage. Rounding to packs always comes last, because the pack sizes are what you can actually buy, and rounding earlier compounds the error. Adding a 5% safety stock shows where a third allowance slots into the same stack.

Weighing it pour by pour

Most makers melt in rounds: a pot or a 1 litre electric melter takes a few candles' worth at a time. The right way to weigh is to decide how many candles each round makes, then weigh the oil and wax for exactly that number. The table gives the net figures and the figures with the 3% melt allowance built in, rounded to one decimal so you can see the arithmetic. On a scale that reads to 1 g, round the oil to the nearest gram.

Pick a pour size and stay with it for the whole run, so every round uses the same two weights and the batch sheet stays simple. Ten candles per round is a comfortable size for many home setups: 1,840 g of wax and 160 g of oil net, or about 1,895 g and 165 g with the allowance. Five rounds complete the order. Whatever melt capacity you have, check it by weighing a full melt of your own wax the first time. Capacity in litres is not the same as grams, and 200 ml does not mean 200 g explains why.

Weighing per pour also protects accuracy. A kitchen or candle scale has a resolution, often 1 g. The rounding error on any single reading can be up to half of that. On one candle's 16 g, that is a large share of the dose. On a 10-candle pour it is negligible. The table below shows the worst-case rounding error as a share of the oil being weighed, assuming a scale that reads to 1 g. Check your own scale's resolution on its label or manual.

The cheapest real bottles

CSI lists Roasted Coffee in five in-stock sizes. Rounding 872 g up to real bottles is a small puzzle with several answers, and the cheapest is not the obvious one. Every combination below covers at least 872 g, and every price is the listed in-stock price pulled on 24 September 2026.

The cheapest cover is one 500 g and four 100 g, which is 900 g for ₹3,450.00. That is only ₹288.00 less than the kilo, and the kilo leaves you 128 g spare instead of 28 g. The extra 100 g is enough for six more candles of the same size. So the real choice is not about price per gram, which is close to flat for this oil (₹3.90 at 100 g, ₹3.78 at 500 g, ₹3.74 at 1 kg). It is whether you expect to make more Roasted Coffee candles soon. If you do, the kilo is the better buy. If this is a one-off order, the 900 g mix keeps ₹288.00 in your pocket.

The same 872 g lands differently on other oils, which is why the calculation belongs to the oil, not to the batch. Here are three oils side by side:

British Rose is the interesting case. It is one of the oils with a real per-gram break at 1 kg (10.3% cheaper per gram than its 100 g), so the kilo costs only ₹106.20 more than the 900 g mix and brings 100 g more oil. For British Rose the kilo is the sensible buy even for a one-off batch. Orchard Pear behaves like Roasted Coffee: the kilo is a convenience purchase, not a saving. At what quantity the kilo becomes more economical works through the break-even for any oil.

The full 50-candle order

Here is a complete shopping list for the fifty candles plus three tests, with the same 3% melt allowance on the wax. Wax is Luxury Soy Wax CSI 464, which is rated up to 10% fragrance load by the wax maker, so 8% sits comfortably inside it. The wax need is 9,200 g net, 9,476 g with 3%, and 10,028 g with the three test candles, so a 10 kg bag alone is 28 g short. Wicks are Eco Candle Wicks Thin C1 and jars White Matte Glass Jars. Wick size for your jar diameter is not published, so confirm it in a burn test before the full pour.

That is ₹260.64 per saleable candle in raw materials, with the test candles and spare stock absorbed into the fifty. It excludes labour, packaging, freight and GST, which you add from your own figures. Buying seven more wicks than you need costs ₹41.30 and is worth it; a bent or badly centred wick is a remade candle, and how many spare wicks to keep covers that allowance properly.

Once this batch is done, keep the numbers. Weigh the oil left in the bottles and the wax left in the bag, and note how many candles needed top-ups. Those three observations turn next month's allowance into a measured figure. For a regular monthly run, how much stock to keep for 100 candles a month and when a 16 g-a-candle bestseller should trigger a reorder take the same maths into inventory.

Frequently asked questions