My Fragrance Oil Has Become More Expensive — Should I Reformulate Immediately or Calculate How Much the Increase Actually Changes COGS?

The impact formula and your BOM before and after

Supplier price notices are written per kilo, so the number that lands in your inbox looks large: ₹500.00 a kilo. Your customer buys a candle, not a kilo. Convert the rise to the unit you sell:

Impact per candle = (new price per kg − old price per kg) ÷ 1,000 × grams of oil per candle.

Fragrance load is a percentage of total candle weight (CSI's convention), so a 200 g candle at 8% holds 16 g of oil and 184 g of wax. With your oil going from ₹4,000.00 to ₹4,500.00 a kilo (worked example — use your own), the rise is 0.50 rupees a gram and the candle carries ₹8.00 more. The rest of the BOM has not moved. Here it is, with CSI's Luxury Soy Wax Chunks and an Eco C2 wick as the wax and wick lines, and your own jar and box as labelled worked examples. Materials only: labour, freight, GST and marketing are excluded.

Fragrance goes from 23.2% to 25.4% of the candle. The oil price rose 12.5%; the candle's materials rose 2.9%. That gap is the whole reason to calculate before you act: the rest of the candle dilutes the rise, and the bigger your jar and packaging lines, the more it dilutes. If your fragrance is a bigger share of COGS than here (a smaller jar, a perfume-grade oil, a 10% load), the same rise bites harder; run the same table with your own lines. The method for building the full BOM is the cost-per-gram and cost-per-candle guide; how fragrance compares with the other lines when you are choosing what to attack is the fragrance-share cluster hub.

How big is it really? Three ways to size the rise

One number per candle is easy to dismiss and easy to panic over. Put it in three frames before you decide anything.

1. Share of COGS

₹8.00 on ₹275.82 is 2.9%. This tells you how far a price change has to move to stand still: a 2.9% cost rise is not a reason for a 12% price rise.

2. Margin points on the selling price

At a ₹699.00 selling price (worked example), materials margin goes from 60.5% to 59.4%: you lose 1.14 points. Contribution per candle falls from ₹423.18 to ₹415.18, which is 1.9% of what each candle earned you before. To keep the same rupees per candle you need ₹707.00; to keep the same percentage margin you would need ₹719.27. Most makers aim at the rupees, not the percentage, and round to their next price point at the next scheduled change rather than mid-season.

3. Rupees per month and per year

At 500 candles a month of this SKU (worked example — use your own), the rise costs ₹4,000.00 a month and ₹48,000.00 a year. This is the number to hold against the cost of doing something about it. A fix that costs more than a few months of the rise, or that carries a risk of losing customers, has to be clearly better than absorbing it.

Two things the table shows. First, load multiplies the rise: the same ₹500.00 costs a 10% candle ₹10.00 and a 6% candle ₹6.00. If you have never tested your current oil below its current load, the rise is a good reason to do it (lever two below). Second, it takes a large rise to cross 3 points on a candle at this price: about ₹1,310.62 a kilo at 16 g. A cheaper candle crosses sooner: at half the selling price, every threshold arrives at half the rise.

Thresholds for action: absorb, act, review, re-source

These thresholds are business judgement, not a standard. They are written in margin points (impact ÷ selling price × 100) because that scales with your price. Set your own once and apply them every time.

The worked rise sits at 1.14 points: just inside the middle zone. So the answer to the title, for this candle, is: calculate (done), use the pack ladder and a load test, and price at the next review. Reformulation with a new oil is not yet on the table. A rise twice as big would put it there, and the next sections show why even then it should be a controlled test, not a reaction.

Lever one: the same oil in a bigger pack

If you buy small bottles, the per-gram price of the same oil at a bigger pack may recover some or all of the rise. It is the same oil, so it is not a formulation change, though a new lot of any oil is worth a quick side-by-side sniff and burn (batch checks are the batch-consistency questions to ask a supplier). Here is what the pack ladder is worth per 8% candle on three CSI oils, at live prices on 4 October 2026:

On Jasmine, a maker buying 100 g bottles recovers ₹6.05 of an ₹8.00 rise by buying 500 g; at 500 candles a month that is ₹3,025.00. On Lavender the ladder is shallower. On Woody Oud there is nothing to recover above 100 g. This is why the check comes before anything else: it is free to do and the answer differs oil by oil. Of the 78 CSI oils sold in both 500 g and 1 kg, the kilo costs exactly two 500 g packs on 24 and more than two on 20, so read the per-gram figure for your oil rather than assuming the kilo wins.

If your current oil is not from CSI, the same check applies at your own supplier: ask for the per-gram price at the next pack up. Do not change supplier to chase a pack discount on the same scent name; a different supplier's "lavender" is a different oil.

Lever two: a load test on the same oil

If you set your load once, years ago, and never tested lower, the rise is the moment to find out whether you need it. One point lower in a 200 g candle takes 2 g of oil out and puts 2 g of wax in. At the new price, with Chunks wax, that is worth ₹7.80 a candle, 98% of the rise, and ₹3,900.00 a month at 500 candles. At 7% the candle holds 14 g of oil and 186 g of wax.

It is the same oil, the same wax and the same wick, so it is a smaller change than a new oil. It is still a change. Bank it only if the 7% candle cannot be reliably told apart from the 8% candle in a blind comparison after cure and through to the last third of the jar, and if it burns as well on the same wick. The test design (which loads, how many candles, how to sniff) is the lower-load test design post; the method for finding the lowest load that still throws is the lowest effective load guide; and the money from a confirmed cut is the 10% to 8% saving across 1,000 candles.

Why panic reformulation costs more than the rise

"Reformulate immediately" usually means "buy a cheaper oil with a similar name and pour it next week". That skips the one thing that made your current candle sellable: the tests it passed. A new oil can need a different load, can burn differently on the same wick, can discolour differently and can smell different in the jar from the bottle. There are two ways to do it, and both cost money.

The controlled way: a three-oil re-source test

Three candidate oils, each poured at 7% and 9% (3 candles each, 96 g of oil per candidate, 4 g left in a 100 g bottle), against 3 controls of your current oil at 8%, in your jar with an Eco C2 wick and Chunks wax (worked example — the candidate bottle is priced at CSI Jasmine's 100 g, ₹430.00):

That is 1.8 months of the rise at 500 candles, before anything has passed, plus weeks of cure and full-life burns. What it repays depends entirely on the result. Suppose (hypothetically) a candidate costs ₹3,500.00 a kilo. If it passes at 9%, the oil-plus-wax cost per candle is ₹172.18 against ₹182.38 for your current oil at the new price: ₹10.20 saved a candle, and the test repays in about 1.4 months. If it needs 10%, the cost is ₹177.98 and the saving shrinks to ₹4.40, repaid in about 3.3 months. If none of the three passes, you have spent ₹7,334.25 to learn that your current oil was right. The full switching-cost and payback method belongs to the 15%-cheaper supplier payback post; how to compare two oils at the loads each needs is the ₹5,000 vs ₹3,500 per kg comparison.

The panic way: swap and hope

Skip the test and the risk moves to your customers. Suppose (hypothetically) the swapped candle is noticeably weaker and 3% of 500 buyers a month ask for a refund at ₹699.00: that is ₹10,485.00 a month, 2.6 times the ₹4,000.00 the rise was costing you, before the reviews and the repeat orders you do not see. The number is illustrative; the direction is not. A rise is a known, measurable cost. A failed reformulation is an unknown one.

A seven-day response plan

A price notice should start a short, fixed routine, not a scramble. Here it is in the order that protects the candle you already sell.

Nothing on this list removes or changes the scent your customers buy until a test says it is safe to. That is the point: the rise is the supplier's change; you should not add a second, uncontrolled change of your own on top of it.

The oils and wax behind the numbers

Three oils that show three different pack ladders, and the wax in the worked BOM. Chosen on page statements, price and stock on 4 October 2026; CSI publishes no throw data, and none of these is offered as a drop-in replacement for your oil. The last column is the oil cost of a 200 g candle at 8% (16 g) at each pack's price.

Frequently asked questions