A New Supplier Is 15% Cheaper — How Do I Calculate Whether Testing and Switching Costs Are Worth the Expected Annual Saving?

Where 15% lands on your candle

The title gives the 15%; everything else here is a labelled worked example. The candle is the series default, 200 g at 8% fragrance on a total-weight basis (184 g wax, 16 g oil). Wax and fragrance are at your current prices (worked: ₹600.00 and ₹4,800.00 a kilo); the wick and box at CSI prices; the jar and label are yours.

Wax is 39.7% of this candle, so 15% off wax is ₹16.56 a candle, 6.0% of COGS. Fragrance is 27.6%; 15% off it at the same load is ₹11.52. These are the two materials where a supplier switch can matter. For a jar or a box, the logic is the same but the risks are different (fit, heat, breakage); a jar change is a different post. Before any of this, ask whether a switch is the right lever at all: if every line is already near the best available price, the answer is redesign, as the supplier switch or redesign post shows.

15% on price versus 15% on cost per saleable candle

Rejects are the quiet tax on a new material. The worked rise from 2% to 4% is an assumption, not a forecast; use your own first-batch numbers. The load point is worse: a cheaper wax that needs more fragrance to throw the same can give back almost all of the saving, because fragrance costs eight times what wax does per gram here. That combined comparison is the subject of the dearer wax, less fragrance post from the change-control series; this post only asks whether the saving survives.

The one-off switching cost, itemised

The worked wax test: five of your scents, each poured three times in your current wax (control) and three times in the new wax, all in your jar with the same Eco C2 wick and the same 16 g of oil. Thirty candles, cured and burned to the end. The new-wax arm is priced at CSI 464 1 kg bags as an example of a sample purchase; replace it with the new supplier's sample price.

Two items deserve a word. Dead stock is absent here because the old wax can be used up before the switch; if it cannot (it has a short life, or you need the shelf), add its value. The risk premium is the expected cost of a failure that the test missed: here a 10% chance of one bad month costing ₹40,000.00 in returns, remakes and lost repeat orders. Both figures are judgement; set your own, but never set the risk premium to nothing. How to judge the risk side for a wax customers already love is the excellent-feedback threshold post; the itemised budget for a whole range of 20 candles is the cost of reformulating 20 candles.

Payback, year one, and the volume that makes it worth it

At 1,000 candles a month, a real 15% on wax is worth testing even in the bad case: it pays back within about 6.0 months. At 100 candles a month the realistic saving is ₹1,133.00 a month and the switch takes 12.7 months to repay: year one is a loss. The test cost hardly changes with volume; the saving does. That is why the same supplier offer can be right for one maker and wrong for another.

The decision rule, so you can use your own numbers: switch if (realistic saving per saleable candle × candles a month × your payback window in months) ≥ one-off switching cost, and the controlled test has passed. Six months is our default window (judgement); a candle with excellent reviews deserves a shorter window, because a failure costs more than money. If the new supplier is cheaper but less reliable, add that cost before you decide: the reliable-but-dearer supplier post turns it into a threshold percentage, and the stock-out cost post prices production delays.

The oil case: 15% cheaper and the break-even load

Same worked candle, fragrance at your ₹4,800.00 a kilo (worked), the new supplier at ₹4,080.00. At the same 8% load the saving is ₹11.52 a candle, ₹1,38,240.00 a year at 1,000 a month. But fragrance savings live or die on load. The break-even is a one-line formula: break-even load = current load ÷ (1 − discount) = 8% ÷ 0.85 = 9.41%, about 18.8 g in a 200 g candle. Above that, the cheaper oil costs more per candle.

Each oil is a separate test: a new supplier's lavender tells you nothing about its oud. The worked ladder for one oil is three candles each at 8%, 9% and 10% of the new oil against three control candles of your current oil at 8%:

If all six oils passed at 8%, the six tests would pay back in 2.2 months at 1,000 candles a month. Per oil, the sum is less kind: an oil selling a sixth of your volume repays its own test in 2.2 months. Switch your biggest-selling scents first, and do not test an oil you sell twenty of a month. Which fragrance tests to repeat when the supplier changes is the fragrance supplier switch post, and how to judge whether a pricier, more concentrated oil earns its price is the concentrated oil post; the change-control version with a ₹5,000.00 and a ₹3,500.00 oil is the different-loads comparison post.

The test gate, and keeping the customer's candle identical

Keep the customer's candle identical: same jar, same label, same fill, same scent name, same burn. If the test shows a visible or burn difference, the saving is not 15% of anything; it is the price of a different product. Record the version change in your bill of materials so you can prove the saving later: the before-and-after BOM post.

Test materials at real prices

Four items for the worked tests: a wax sample priced at the production rate, two oils that show how sample sizes can mislead (one flat, one stepped), and the wick that stays identical in both arms. Chosen on page statements, price and stock on 4 October 2026; CSI publishes no throw or burn data, so nothing is ranked on performance.

Frequently asked questions