My Fragrance Oil Is 35% of My Raw-Material Cost — Should I Switch to a Cheaper Oil or Test Whether a Stronger Fragrance Performs at a Lower Dosage?

What 35% means in rupees

Your title gives one number: fragrance is 35% of your raw-material cost. To turn that into decisions we need a candle, so here is a worked example built to match it: a 200 g candle at 8% fragrance (load as a share of total weight: 16 g of oil, 184 g of wax), every unit price labelled as yours to replace. None of these is a CSI price, and none is a claim about your supplier.

Fragrance is the single biggest line at ₹84.00, and wax, close behind, is the line that grows whenever fragrance shrinks, because the candle still weighs 200 g. Everything that is not fragrance costs ₹156.00 here, and nothing in this cluster touches it.

Two candles can both be 35% for opposite reasons

Reader A pays ₹5,250.00 a kilo and uses 8%: 16 g × ₹5.25 = ₹84.00. Reader B pays ₹4,200.00 a kilo and uses 10%: 20 g × ₹4.20 = the same ₹84.00. Same share, different cure. Reader B's first lever is almost certainly the load, because 10% is the top of most CSI oil pages' stated range and many oils do not need it. Reader A's load is already mid-range, so the per-gram price deserves a harder look. Write down your own grams and price per gram before you choose; the decision follows from which of the two is out of line. If you are at 10% "for strength", the 10% against 8% of a better oil post is your next read; if you are at 12%, the 12% load post.

The one comparison that matters

Every fragrance decision in this cluster reduces to one line: fragrance ₹ per candle = price per gram × candle grams × approved load. The approved load is the lowest load at which your blind, full-life burn test says the candle throws as well as it must. It is not the bottle's maximum and not the supplier's suggestion. That one number turns "cheaper oil or stronger oil" into arithmetic.

Read the table both ways. A "stronger" oil approved at 6% wins only if it costs less than ₹7,000.00 a kilo; CSI's Dark Vanilla at ₹7,646.00 would cost ₹91.75 at 6%, ₹7.75 more than you pay now, however well it throws. A "cheaper" oil that needs 10% wins only below ₹4,200.00; at ₹4,531.20, Woody Oud at 10% would cost ₹90.62. The general method, with seven CSI oils at three loads and two candle sizes, is the cost-per-finished-candle method; the worked ₹4,500.00 against ₹2,500.00 case is the ₹4,500.00 at 6% against ₹2,500.00 at 10% post; and comparing two suppliers, with documentation and consistency counted, is the supplier A against B post.

Two things the formula does not let you do. It does not let you approve a load above the wax page's or the oil page's stated maximum (the lower of the two governs; every oil on this page states 10% for candles), and it does not let you count a load you have not tested. "This oil is strong, so 6% will do" is a hope. Why bottle strength does not predict candle strength, and how to find the lowest load that still throws, belong to the lowest effective load post.

The decision tree: pack size, load, then oil

Three gates, in order of risk. Each one is priced on the worked candle, one change at a time, never stacked on a saving it overlaps.

Gate 1: pack size, the only saving with no test

The same oil, bought in a bigger pack, is the same candle. On CSI ladders the per-gram drop is steep from 15 g to 100 g and small after it. 16 g of Lavender costs ₹112.02 at the 15 g price, ₹86.40 at 100 g and ₹83.54 at 1 kg: ₹28.49 a candle between the smallest and largest bottle, but only ₹2.86 from 100 g up. Citrus Lemon drops ₹6.40 a candle from 50 g to 100 g and ₹2.69 more to 1 kg. Woody Oud saves nothing at all past 100 g (₹72.50 either way). If you already buy your oil by the kilo, gate 1 is closed: move on. Whether a supplier's 5 kg offer is worth the cash is the 5 kg fragrance pack post; one oil at 500 g a month has its own post in the map below.

Gate 2: a lower load of the oil you already use

Dropping from 8% to 7% removes 2 g of oil, worth ₹10.50 at ₹5.25 a gram. The candle still weighs 200 g, so 2 g of wax goes back in at ₹0.90: net ₹9.60 a candle, and fragrance falls to 31.9% of a ₹230.40 candle. A further point, to 6%, doubles that to ₹19.20. Both exist only if the blind burn test passes across the candle's whole life, and some oils will not pass: that is why it is a test and not a setting. The cluster splits the work: the percentages to compare and the test's own cost are the lower-load test design, the 1,000-candle money is the 10% to 8% savings post, and whether a lower load also lets you change the wick is the wick-behaviour post, a possibility to test, never a saving to count now.

Gate 3: another oil, last

Only after gates 1 and 2 does an oil swap earn its test. Inside your scent family it is a real option: for an oud candle, Woody Oud at 8% would cost ₹72.50, ₹11.50 under your ₹84.00. Outside your family it is not a saving at all; it is a new product with a new customer test. And a swap re-opens everything a load cut does not: wick behaviour, colour, cure, and how the throw holds to the last centimetre. That is why the change-control series treats a new oil as a formulation change (what to retest when changing fragrance oil), and why price comparisons across different loads need the ₹5,000 against ₹3,500 per kg method before anyone signs a purchase order.

CSI's price spread, and where switching saves nothing

The candle-grade oils this cluster prices run from Citrus Lemon at ₹2,832.00 a kilo, the cheapest 1 kg on CSI's list, to Lavender at ₹5,221.00: a ₹2,389.00 spread, or ₹38.23 a candle at 16 g. Above that sit dearer candle-grade oils such as Dark Vanilla at ₹7,646.00, and the perfume-grade "inspired" listings, which run as high as ₹24,056.00 a kilo and are often priced flat, or nearly flat, per gram from 50 g; their pages quote perfume usage rates, not candle loads.

Where CSI can help is the pack-size gate and the in-family candidate. If you buy a 50 g bottle at a time, our 100 g and 1 kg tiers may be a real per-gram drop on the oils you use; check each oil, because the ladders differ (Woody Oud is flat from 100 g, Lavender barely moves from 50 g to 100 g, Citrus Lemon falls at every step). And CSI lists no 5 kg fragrance packs: above 1 kg, pricing is a conversation on WhatsApp, and we will not print a discount here we have not agreed with you.

Test gate, annual saving and payback

No load cut and no new oil reaches a customer until it passes a controlled test: one variable, everything else frozen, at least three candles per arm, burned to the end, judged blind. This page owns the money; the method lives elsewhere: whether 7% performs like 10% on an expensive oil, the dosage-first order for cutting fragrance cost, and how to test cheaper alternatives without risking a bestseller.

Sensitivity: at 150 candles a month, 8% → 7% is worth ₹17,280.00 a year and the test takes about 1.9 months to pay back; at 2,000 a month it is worth ₹2,30,400.00. The load ladder costs ₹661.50 of oil, ₹753.30 of wax, ₹495.00 of jars, ₹63.00 of wicks and ₹750.00 of labour. The swap test costs ₹993.12 for two 100 g candidate bottles, ₹1,252.80 of wax, ₹252.00 of control oil, ₹825.00 of jars, ₹105.00 of wicks and ₹1,200.00 of labour. Every test candle is priced at full cost.

What to buy for the comparison

Four items chosen on page statements, price ladders and stock on 4 October 2026; CSI publishes no throw data, so nothing here is ranked on performance. Counts assume a 200 g candle at 8% (16 g of oil, 184 g of wax).

Skip for this decision: the perfume-grade inspired listings (flat per gram, perfume usage rates on the page, and far above your break-even) and any 1 kg bottle bought before a test approves the oil. A kilo of an oil that fails your panel is the most expensive fragrance you will ever buy.

Map of this cluster: 6021–6040

This hub owns the comparison rule, the gate order and the 35% answer; each sibling owns one decision. The four-lever summary that pulls them together is 6040, the last row.

Frequently asked questions