I Tested Three Soy Waxes and One Has the Best Hot Throw, One Has the Smoothest Tops and One Survives Heat Best — How Do I Choose Which One Should Become My Commercial Wax?

You do not pick the best wax. You pick the best wax for your business

When one wax throws best, another has the smoothest tops and a third survives heat best, the result is telling you that no wax is best at everything in your candle — which is normal. The mistake is to treat it as a tie and pick by gut, or to pick the throw winner because throw is what customers mention in reviews. The right move is to decide, in writing, how much each of those qualities is worth to your business, and to let the arithmetic choose. A shop that ships across India in May and a boutique selling clear-glass candles over the counter in Bengaluru should not end up with the same answer from the same test sheet.

The method has four layers, and their order matters. First, gates: any wax that fails a safety line, or fails a line your business cannot live without, is out regardless of its other wins. Second, fixability: a weakness that a process change can remove — rough tops are the classic case — is not the same as a weakness that cannot be fixed without changing the formula. Third, weights: a weighted score across the surviving lines, with weights set before you look at the scores. Fourth, cost per saleable candle: the price difference between the finalists, including the rework and rejects each one causes.

Layer one: the gates no wax gets past on its other wins

A gate is a pass/fail line. If a wax fails it, its score on every other line is irrelevant. There are two kinds. Safety gates apply to everyone: no wax may be worse than the others on flame height, soot, mushrooming or jar reading at any stage of the full-life burn. CSI publishes no numeric flame-height or jar-temperature limits; the standard is "no worse than the best wax in the comparison, within your jar supplier's guidance and the recognised candle fire-safety standard you work to" (ASTM F2417 is a commonly cited US standard; read it for its own requirements). A hotter jar is never traded for more throw.

Business gates depend on how you sell. If you ship across India through summer, surviving your actual route is a gate, not a weighted line — a candle that arrives sweating or slumped is a return, not a lower score. If you sell only over the counter in an air-conditioned shop, heat survival is a weighted line, not a gate. If your brand promise is a strong throw, a throw floor — the minimum blind score you will accept — is a gate. Write your business gates down with the threshold you will judge them by, and remember that heat results come from your own logged route temperatures and labelled test conditions: CSI publishes no heat-survival data for any wax, and simulating hot courier conditions covers how to run that test.

Apply the gates to your three waxes now. In the online-shipping profile, Waxes A and B must pass the route test to stay in; if either sweats or slumps after your simulated transit, it is out, and Wax C may win by default. In the throw-led profile, B and C must clear your throw floor. Only waxes that pass every gate go on to the next layer.

Layer two: is the losing line fixable?

Before you weight anything, ask of each wax's weakest line: can one process change, made once and kept, remove it? Weaknesses differ in how fixable they are, and an unfixed weakness should count against a wax more than a fixable one. Tops are commonly influenced by pour and jar temperature, cooling rate and draughts, so the wax that lost on tops may close the gap with one pouring trial — change the wax or optimise pouring first sets that trial out. Throw can sometimes be raised inside the wax's published ceiling or by a wick change that brings the pool to full width — but only after the pool is right, and never by trading jar heat. Heat survival is the hardest: in general practice it moves with wax choice, blend and load, so a wax that fails your route is rarely rescued by process alone.

A fix that works once is a lead, not a result. Each fix goes into the confirmation round below with three candles, beside one candle made the old way, so you can see that it worked and that it did not break another line. A load step that rescues throw but sends a candle sweating in warm storage has not fixed anything. And a fix that needs a skilled hand every time — a heat gun on every top, a pour temperature held to one degree — belongs in the process-tolerance line, which is why process tolerance deserves a weight of its own.

Layer three: the weighted decision matrix

The matrix multiplies each wax's score on each line by the weight that line carries for your business, then adds them up. The scores below are illustrative: the only figures that come from the question are the three 5s — A best on throw, B best on tops, C best on heat. Every other score is a placeholder chosen to show how the arithmetic behaves. Replace all of them with your own sheet; this page publishes no test results for any wax.

The same sheet produces three different commercial waxes. With shipping-heavy weights, Wax C wins at 3.70. With clear-glass weights, Wax B wins at 3.95. With throw-led weights, Wax A wins at 3.75. None of those answers is wrong; each is right for a different business. That is the whole point of writing the weights down first: it stops the decision becoming a debate about which quality matters most "in general", when the only question that matters is which quality earns or loses you money.

Layer four: cost per saleable candle settles the close calls

When two waxes are close on the weighted score, the cheaper one per saleable candle wins. That cost has three parts: the material cost at the load that passed, the rework labour the wax needs, and the rejects it produces. The table shows the material cost of a 200 g candle at 8% in each of the three CSI container waxes most often tested together, with Sandalwood at its 1 kg rate of ₹5.03 a gram, the ₹5.90 wick and the ₹70.80 jar. It excludes labour, packaging, freight and GST. It does not tell you which of them is your A, B or C — your sheet does.

Two things follow. First, the material spread between these three waxes is ₹23.85 a candle at production prices — smaller than many makers expect, because the oil (₹80.43) and the jar (₹70.80) dominate. Second, the cost of a weakness can be bigger than the wax price difference. On a candle costing ₹250.49, cutting cosmetic rejects from 6% to 2% is worth ₹10.88 per saleable candle — close to half the ₹23.85 gap between the dearest and cheapest of the three, and more than the ₹6.99 gap between Chunks and CSI 468. So a wax that wins on tops can justify a higher price if, and only if, your own reject records show the saving. Including rejects in the true wax cost and putting a value on production consistency do this with full monthly volumes.

The fix-and-confirm round, priced

Before a wax becomes your commercial standard, confirm it. For each finalist, pour four candles: one made exactly as in the first test, and three with its single fix applied. Same oil lot, same load, same wick, same jar, same cure age. If the fixed candles beat the as-made candle on the weak line and are no worse on every other line, re-score that wax with the fix and recompute the matrix. Each wax needs 736 g, or 772.8 g with a 5% pour-loss allowance (working assumption); the 12 candles need 201.6 g of oil with the same allowance.

If only two waxes survived the gates, drop the third line and its share of the oil and jars. If a fix needs a different wick, add a 10-pack of the alternative — for example Eco Thick C2 at ₹94.40 — as a construction to bracket against; CSI publishes no wick-size charts and no CSI wick is claimed as a size step from another.

The decision rule, ending in a commercial order

Apply the layers in order and stop at the first that decides it.

The order then follows the winner. If it is CSI 464, order one 5 kg bag at ₹2,537.00 for a pilot of 27 candles, then 10 kg bags at ₹5,074.00 — the price per kilo is the same, so buy the size your monthly use empties within a few weeks. If it is Chunks, buy 5 kg bags at ₹2,995.00: two of them are ₹9.00 cheaper than the 10 kg bag. If it is CSI 468, pilot on 5 kg at ₹3,185.00 and move to the 10 kg bag at ₹6,370.00, which saves ₹0.00 against two 5 kg bags. Keep 200–300 g of the bag you approved as a retained sample and write the order date on it: CSI publishes no lot numbers, so your retained sample is the reference for every future bag. Writing the qualification test as a standard turns the whole decision into a document your team can follow.

Where the three lines go deeper

Each of the three wins in this question has its own post in the series. For the heat line, one wax throws better and another survives heat better covers the extra tests that pair needs, and the shipping-heat cluster hub covers route-based testing. For the tops line, process first or wax first is the fix trial. For the throw line, cold and hot throw before a big order sets out the throw evaluation itself. And if the decision has to hold across a whole range rather than one oil, testing one wax across your collection comes before this matrix, not after it.

Frequently asked questions