I Have a 500-Candle Wedding Order with a Custom Fragrance — How Much Extra Fragrance Oil Should I Buy for Testing and Production Wastage?

The oil budget, line by line

The order: 500 wedding favours for a November–February wedding, each a Heart Cutout Velvet Jar in red, with a scent the couple wants to feel like theirs. You propose a blend of Raat Ki Rani (Night-Blooming Jasmine) and Sandalwood, poured in Luxury Soy Wax Chunks at 8%. The jar is sold with a stated 50 g wax capacity; we treat 50 g as the finished candle, so each favour is 4 g of blend and 46 g of wax. Weigh one of yours to confirm.

Of the 278 g beyond production, only 104 g is wastage in the sense of oil lost to beakers, bottles and pour jugs. The rest is oil that goes into a candle on purpose: one you burn to check the wick, one a family member sniffs in September, one you keep in a drawer so that when the couple's cousin orders the same favours next year you can match them. Treat those as planned uses, not losses. The difference matters when the client asks why the oil line on your quote is higher than 500 × 4 g.

The wastage allowance here is 5%, at the top of the usual 3–5% working range, because a blend passes through more vessels than a single oil: two bottles into a blending jar, the blending jar into a pouring jug. Each transfer leaves a film. On a single-oil order you could use 3%. Neither figure is measured by CSI; they are planning assumptions to adjust once you have weighed your own losses on a run.

Development trials: how much oil a blend costs to find

A two-oil blend has two variables: the ratio and the load. Round one tests the ratio: 80:20, 70:30 and 60:40, two favours each at 8% — one to burn, one to smell cold. That's six candles and 24 g of blend. Round two tests the load on the winning ratio: two favours at 7% and two at 9% — 16 g more. Ten candles and 40 g of blend to arrive at a formula you can defend.

Weighing 0.8 g of oil into a 50 g candle needs a scale that reads to 0.1 g, or you mix a small trial blend first (say 20 g at each ratio) and weigh 4 g of that. The second method is easier and more accurate, and the leftovers of each trial blend are still usable — keep them labelled. Luxury Soy Wax Chunks are rated up to 13% fragrance load (the wax maker's stated formulation maximum, not a regulatory limit), so the 9% trial is well inside it.

If the couple prefers a single oil after all, trials shrink to a load round. That's cheaper to develop and easier to reproduce. For comparison, the post-approval quantity as straight Jasmine would be 2 × 1 kg (₹3,923.00) + 2 × 100 g (₹430.00) + 1 × 15 g (₹103.00) = ₹8,809.00, against ₹13,911.70 for the blend. The blend is dearer here because Raat Ki Rani is ₹6.77 a gram at 1 kg against Jasmine's ₹3.92. The per-favour oil cost of the blend is ₹25.00.

One master blend for the whole run

Once the ratio is approved, weigh the whole post-approval quantity into one clean, dry, closable container: 1,549.8 g of Raat Ki Rani and 664.2 g of Sandalwood, 2,214 g of blend. Mix it thoroughly. Pour the 30 g retained sample into a small sealed bottle, label it with the ratio and date, and put it away. Every production pour then weighs blend, not two oils. That removes a whole class of error: a slightly heavy hand on the Sandalwood in batch seven.

It also explains why the wastage allowance sits on the production blend: the blending container keeps a film, and so does every pouring jug that takes blend from it. If you blend in small lots per pour instead, you lose a film in the blending beaker every time, and the real wastage climbs well above 5%. Keeping batches identical covers the same principle for wax and colour.

What to buy before and after approval

Weddings change: the colour scheme moves from red to ivory, the guest list grows, a family elder dislikes jasmine. So the split is deliberate. Before approval, buy only what finds the formula and persuades the family. After approval — and in writing — buy everything for production in one go.

Rounding to packs adds a little more: you buy 2,295 g for a need of 2,278 g. The pre-approval packs leave about 5 g of Raat Ki Rani and 11 g of Sandalwood unused, which can go into the master blend or a test for another client. In total, the oil you buy is 14.8% above the production need. That is the honest answer to the title: plan on roughly 14–15% more oil than the favours contain on a blended custom order like this one, and know exactly what each part of it is for.

The rest of the favour

The favour needs 46 g of wax, a wick and the heart jar. Poured units are 536 (500 favours, 10 QC, 10 reserve, 6 approval, 10 trials), so wax is 25.89 kg with a 5% allowance: 2 × 10 kg (₹5,999.00) + 1 × 5 kg (₹2,995.00) + 1 × 1 kg (₹599.00) = ₹15,592.00. Jars: 546 with ten spare, bought as 54 × 10 (₹450.00) + 1 × 5 (₹225.00) + 1 × 1 (₹45.00) = ₹24,570.00 for 546. Wicks: 1 × 500 (₹2,950.00) + 1 × 50 (₹295.00) + 1 × 10 (₹59.00) = ₹3,304.00.

The per-favour column spreads every test, reserve and allowance across the 500 favours the couple pays for — the right basis for a quote. Ribbons, tags, boxes and printing are not CSI items; price them with your own suppliers and add them as your own lines. The heart jar is in stock in red only on the pull date (the white version is out of stock), so confirm the colour with the couple before you promise it. Custom-coloured candles covers colour decisions when the jar colour is fixed.

If the wedding favours are only part of your season, the neighbouring questions are how to cost the first quote (a 100-candle inquiry), how much inventory a 500-piece run needs before the first pour (500 custom candles), how much extra wax Diwali volumes call for (1,000 Diwali candles) and what buffer a fixed wedding date deserves (strict delivery dates).

Two related questions sit next to this one. If the wedding is booked months ahead, protecting your quoted price covers what happens if oil prices move before production. And if the order is large enough that the oil runs to several kilos of a single scent, calculating kilograms before ordering and testing the bulk batch against your sample take over.

Frequently asked questions