I'm Making 500 Candles Across Five Fragrances — How Do I Divide My Fragrance Oil Purchase When Some Scents Are Expected to Sell More than Others?

The forecast split

The forecast share is your estimate of how the 500 candles will sell across the five scents. It can come from a previous season, a pre-order list, a stall where you watched which jars people picked up, or plain judgement. It does not need to be right, because the wave method corrects it. It does need to be written down, as a percentage per scent that adds to 100. The example uses Lavender 30%, Vanilla Bean 25%, Sandalwood 20%, Mango Coconut 15% and Fresh Strawberries 10%. These shares are illustrative, not CSI sales data. Lavender is first in the collection's best-selling sort on the pull date, which is a reasonable reason to lead with it, not a forecast.

Each scent's grams are candles × 16 g × 1.05. That is 16.8 g of oil for every candle poured, counting the jug film and the odd scrapped candle as a working 5%. The cost column is the cheapest combination of in-stock 100 g, 500 g and 1 kg bottles that covers the grams. Testers are left out, because production from 15 g bottles pays a median 23.1% per-gram premium. The up-front total is ₹46,973.96. If you also need trial oil or QC pulls, add them per scent as in the single-scent build-up.

Compare the equal split. At 100 candles each, Fresh Strawberries gets twice the oil its forecast suggests and Lavender a third less. The equal split is common because it feels fair to the scents. Your customers are not fair to them, and neither should the order be. If you have no data at all, even a rough 40/60 split between the two scents you are most and least sure of is better than equal. I'm Launching 10 Candle Fragrances — How Do I Decide Which Scents Deserve 1 Kg Fragrance Oil Purchases and Which Should Remain in Smaller Trial Sizes? covers how to get that first data cheaply.

Rounding each scent to packs

Five scents means five separate roundings, and the small scents suffer most. A scent needing 840 g cannot be bought as 840 g. It has to be some mix of 100 g, 500 g and 1 kg bottles. Whether that mix costs more per gram depends on the oil's ladder, and the five example oils have different ones:

Mango Coconut and Fresh Strawberries have their whole kilo saving already at 500 g, so a small allocation in 500 g bottles pays no penalty. Lavender and Vanilla Bean are within a percent between 500 g and 1 kg. Sandalwood is ₹5.03 a gram at every size from 100 g up. None of the five punishes a scent for being small. That is the fact that makes the next section work. Some oils do have a real step between sizes, such as Cinnamon's 10.0% at 1 kg with no step at 500 g, and for those, where you split matters more. Check the ladder for every scent in the range. I'm Launching 10 Candle Fragrances — How Do I Decide Which Scents Deserve 1 Kg Fragrance Oil Purchases and Which Should Remain in Smaller Trial Sizes? shows how to read it.

Two waves, not one guess

The wave method splits the 500 candles into two pours. Wave 1 is 60% of each scent's forecast, 300 candles in total, bought and poured now. Wave 2 is the remaining 200 candles, allocated only after wave 1 has been on sale long enough to show real proportions. You choose the window, for example three or four weeks. Wave 2's oil for each scent is its new candle count × 16.8 g, minus whatever wave 1's bottles left over, so no rounding is paid for twice. The 60/40 split is a working choice. Lean toward 70/30 if your forecast rests on last season's sales, or toward 50/50 for a brand-new range.

In the example, wave 1 sales came in as 34% Lavender, 22% Vanilla Bean, 12% Sandalwood, 22% Mango Coconut and 10% Fresh Strawberries. Wave 2's 200 candles follow those shares. Sandalwood gets 24 more candles, not the 40 its forecast implied, and Mango Coconut gets 44, not 30. Each scent's wave 2 purchase is reduced by the oil wave 1's bottles left over. For Sandalwood that is 92 g, which is why its wave 2 needs only 4 × 100 g. Wave 1 cost ₹29,556.08 and wave 2 ₹17,306.52, ₹46,862.60 in total.

What the waves save

Compare the two plans against the same observed demand. The up-front order bought each scent's candles by forecast. The waves poured each scent's candles by forecast, then by sales:

The up-front order leaves 370 g of oil in the wrong scents, worth ₹2,121.92 at their 1 kg rates. That oil will be used eventually, but it is cash on the shelf. It also leaves Lavender and Mango Coconut 22 candles short during the season, which means either lost sales or a rushed reorder. The two-wave plan costs ₹46,862.60 against ₹46,973.96, a saving of ₹111.36, and ends with every scent matched to its sales. The oil cost hardly moves, because the waves cost almost nothing on a flat ladder. What changes is where the oil ends up.

The tail scent problem

Every range has a tail, the scent at 10% that you keep because it rounds out the collection. The tail is where allocation errors are proportionally largest. A 10-candle miss on a 150-candle scent is 7%, and on a 50-candle scent it is 20%. It is also where pack rounding bites hardest. Fresh Strawberries' wave 1 needed 504 g and bought 600 g, leaving 96 g that wave 2 absorbed. Three practical rules follow. Give the tail a floor, not a share. Pour enough to keep it visible, for example two weeks of shelf, and let wave 2 decide if it earns more. Buy the tail in the size where its ladder flattens. For Fresh Strawberries that is 500 g at ₹3.54. Retire a tail scent on evidence. If it sells below its floor for two cycles, drop it and give its share to a new candidate.

Loads can differ by scent too. If Vanilla Bean passed your trial at 9% and the others at 8%, its grams per candle are 18 × 1.05 = 18.9 g, and its allocation must use that. Never apply one load to a whole range by default. I'm Making Different Candle Fragrances in the Same Soy Wax — Should Every Scent Automatically Use the Same 8% Fragrance Load? and My Lavender Candle Performs Well at 8% but My Vanilla Candle Does Not — Should I Optimise Fragrance Load Separately for Every Fragrance? explain why, and I'm Buying Fragrance Oil in Bulk for Candle Production — How Should I Determine the Optimum Dosage Before Calculating How Many Kilograms to Order? covers setting each scent's dosage before calculating kilos. For comparing the cost of scents that need different loads, see One Fragrance Oil Costs More but Gives Stronger Hot Throw at a Lower Dosage — How Do I Compare Fragrance Oils by Cost per Finished Candle Rather than Price per Kilogram?.

Two earlier steps make the split more reliable. If you are choosing the five from a longer list, I Have 30 Fragrance Samples but Only Want to Launch Five Candles — What Testing Sequence Can Eliminate Poor Performers Before I Waste Wax and Wicks on Full Burn Tests? sets out a testing sequence that removes weak performers before you commit wax and wicks. If each scent needs its own load, I'm Launching 20 Candle Fragrances — Do I Need to Test 6%, 8% and 10% for Every Scent or Is There a Faster Formulation Method? gives a faster way to set loads across many scents than a full ladder for each. And if one scent's share grows large enough to need a 20 kg-plus order on its own, I'm Producing 1,000 × 200 g Candles at 10% Fragrance Load — How Many Kilograms of Fragrance Oil Should I Purchase so I Don't Run Short During Production? covers sizing a big single-scent run batch by batch.

One last check before you order: stock. On the pull date every size used in this example was in stock, but several other oils had sizes out, such as Vetiver's 1 kg and Chocolate Fudge's 100 g. A wave plan depends on being able to buy wave 2. If a scent is critical and its ladder has only one production size, ask CSI on WhatsApp about stock before you commit to waves, or buy that one scent's full allocation up front.

Frequently asked questions