I Sell Candles Individually and in Bundles — How Do I Calculate Whether Bundle Discounts Are Reducing My Profit Too Aggressively?

Two questions hide inside one bundle

A bundle does two separate things to your business and they pull in opposite directions. It raises order value, which is genuinely valuable: one pick, one pack, one label, one parcel, one conversation with a customer instead of three. And it lowers realised price per candle, which is the cost of buying that convenience. Whether the bundle is a good idea is simply whether the first effect is bigger than the second, and you cannot answer that by looking at revenue, because a bundle almost always makes revenue per order look better.

So the question has to be asked in contribution — what is left after materials — and it has to be asked against the right alternative. The alternative to a bundle is not nothing. For some buyers the alternative was three separate orders at full price, and for those the bundle is a straight loss equal to the discount. For others the alternative was one candle, or no candle at all, and for those the bundle is a clean gain. Every serious bundle decision is an argument about how the two groups are mixed, and the rest of this page turns that argument into rupees.

The figures throughout use an illustrative single-candle price of ₹649.00 and three illustrative bundle depths of 10%, 15% and 20%. Those are not recommendations and they are not CSI prices — they are round numbers chosen so the method is visible. Swap your own price and your own depth into the same arithmetic and the shape of the answer will not change.

Cost the bundle once, on materials you can buy

The costing unit here is the series convention, stated once and carried throughout: 200 g of wax plus 16 g of fragrance oil at 8% of wax weight, so 216 g goes into each jar. The wax is Premium Coconut Soy Wax CSI 468, whose page states a maximum load of up to 13% of total wax weight along with a melting point of 50–53°C and a pour temperature of 58–60°C. The oil is Fresh Strawberries, whose page states 6–9% of total wax weight and an IFRA certification. Where a wax ceiling and an oil range differ the lower one governs, so 9% is your ceiling on this pairing and the 8% used here sits just under it — which is worth noticing, because a bundle is exactly the moment a maker is tempted to push the load up to make the gift feel stronger.

One structural point before the prices arrive. The bundle's materials are exactly three times one candle's, because nothing in the recipe gets cheaper by being poured three times in a row. That is not a criticism of bundles — it is the reason the whole analysis is about the price side. The only genuine cost savings a bundle creates sit in the two blank rows: one carton instead of three, one outbound parcel instead of three. Those are real and they can be large, and they are still yours to measure, because they are not CSI products and no figure for them would be honest here.

Three depths of discount, three answers

Here is the comparison that matters, run three times. The left-hand column is the same three candles sold one at a time at the illustrative ₹649.00. The others are the same three candles as a bundle at each depth. Materials are identical in every column at ₹790.02, so every rupee of difference is price.

Look at the last two rows together, because they contain the thing that surprises people. Contribution margin on the sale only falls from 59.42% to 49.28% across the whole range — ten points, which does not sound alarming. But contribution per candle falls from ₹385.66 to ₹255.86, which is a third of it, and a third is what your business actually feels. Percentages flatter deep discounts; rupees do not. Always finish the comparison in rupees per candle and rupees per order, and use the percentage only to talk to other people.

The incremental-volume test

Everything above describes one bundle. The decision, though, is about a programme: you are choosing to sell part of your range this way for a season. So convert the discount into the thing it is supposed to buy, which is volume, and ask how much volume it has to buy before it has paid for itself.

The arithmetic is short. Contribution from the singles is ₹1,156.98 per set of three. Contribution from the bundle at 10% off is ₹962.28. Divide the first by the second and you need 1.2023 bundles to match one set of singles — in other words, 20.23% more volume. At 15% off the requirement is 33.77%, and at 20% off it is 50.73%. That last figure is the one to sit with: a fifth off the price has to move half as many sets again before your profit is where it started.

Two cautions about how to use that hurdle. First, it assumes the extra volume costs nothing extra to create; if you are also paying for reach to sell the bundle, the real hurdle is higher, and that advertising line is one of the blanks this page will not fill in. Second, it assumes the bundle's units are additional. They usually are not, entirely — which is the whole of the next section.

Who would have bought anyway

Split your bundle buyers into two groups. The first would have bought three candles regardless — a gifting customer, a repeat buyer stocking up before Diwali. For them the bundle changes nothing except that you charged less, so each of those orders costs you the full discount: ₹194.70, ₹292.05 or ₹389.40 depending on depth. The second group would have bought one candle, or none, and the bundle is what moved them. For them the bundle is a large gain: at 10% off, a bundle returns ₹962.28 against the ₹385.66 a single would have returned.

So there is a break-even mix, and it is computable. Assume, as an assumption and not as measured data, that a buyer moved by the bundle would otherwise have bought exactly one candle. Then the share of bundle orders that must come from that group is the discount divided by the gap between the two alternatives — ₹771.32, the difference between three singles and one. That gives 25.24% at a 10% bundle, 37.86% at 15% and 50.49% at 20%.

That is a testable claim rather than a philosophical one, which is the point of writing it as a percentage. Run the bundle for a month and compare units shipped and contribution banked against the month before. If total contribution rose, the mix was on the right side of the line. If units rose and contribution did not, you have found your answer and it is the expensive one. The measurement is cheap and the assumption is not.

What genuinely gets cheaper at three

It would be convenient if buying materials for bundles brought the per-candle cost down enough to fund the discount. Mostly it does not, and this page will be precise rather than encouraging about it. On Premium Coconut Soy Wax CSI 468 the rate does fall in the larger packs: ₹650.00 a kilogram at 500 g and 1 kg, ₹637.00 a kilogram at 5 kg and 10 kg. That takes 200 g of wax from ₹130.00 to ₹127.40, so ₹2.60 a candle and ₹7.80 a bundle. Set that against ₹194.70 handed back at the shallowest depth and the proportions are clear.

On the fragrance the ladder is worth checking rather than assuming. Fresh Strawberries runs ₹7.55 a gram at 15 gm, ₹4.25 at 50 gm and ₹3.78 at 100 gm, then ₹3.54 at both 500 gm and 1 kg — so the last step up buys you nothing per gram at all, only fewer reorders. The vessel behaves the same way: the Maroon/ Deep Red Matte Glass Jar is ₹70.80 a jar in the Pack of 5 and ₹70.80 a jar in the Pack of 10. CSI publishes no quantity-discount ladder across the store, and the honest summary for a bundle maker is that the shopping list is not where a bundle discount gets paid for.

Freight is the exception and it moves in your favour twice. On the way in, four separate 1 kg packs of Premium Coconut Soy Wax CSI 468 carry a stored packed weight of 1.5 kg each and bill at ₹180.00 a parcel on Standard Shipping, so ₹720.00 in total, against ₹630.00 for the single 5 kg pack at its stored 6.0 kg — a difference of ₹90.00, on the published bands as at September 2026. On the way out, three candles in one carton replace three parcels, and that saving is usually the largest real benefit a bundle produces. It is also entirely yours: CSI publishes rates for delivering to you, not for your outbound parcels, so that row stays blank on this page and gets your own figure in your sheet.

Setting a bundle discount you can defend

One last framing that helps at festival time. Diwali 2026 falls on 8 November 2026, and with a 14 to 21 day cure plus transit, gift sets for it want to be poured by about mid-October. A festive bundle is the easiest bundle to sell and therefore the one least in need of a deep discount — demand is doing the work that the discount would otherwise have to do. If you are going to run 20% off at any point in the year, running it in the quiet weeks after the season rather than during it will cost you far less contribution for the same lift in volume.

Frequently asked questions