How Do I Select and Approve a Soy Wax for Long-Term Commercial Candle Manufacturing Based on Fragrance Performance, Wick Behaviour, Appearance, Indian Heat Stability, Process Consistency, Landed Cost, Supplier Reliability and Batch-to-Batch QC?

Five gates, eight criteria, one written approval

Every question in this hundred-post series is one part of a single decision: which soy wax your business makes candles from, from which supplier, in which pack size, under which checks. The eight criteria in the question — fragrance performance, wick behaviour, appearance, Indian heat stability, process consistency, landed cost, supplier reliability and batch-to-batch QC — are the right eight. The mistake is to judge them all at once, from samples, in one afternoon. They are answered at different scales and over different lengths of time, so they belong in a sequence.

The sequence is five gates. Gate 1, performance: does the wax make a safe, good candle with your jars and oils — throw, wick, appearance, process tolerance? Gate 2, Indian heat: does it hold up in your storage, your courier routes and your hottest season? Gate 3, landed cost: what does a saleable candle cost in it, with fragrance, wick, rejects and freight? Gate 4, procurement: can you buy it in packs, quantities and timings that suit your cash and space? Gate 5, supplier and QC: is the supplier reliable over deliveries, is there a backup, and does every new lot get checked? Each gate maps to one twenty-post cluster of this series, and each has a hub post that sets out its method.

The output is not a feeling about a wax. It is a short set of documents: an Approved Material Specification for the wax, an entry on your Approved Supplier List, an incoming QC plan for every delivery, a retained-sample rule, a named backup wax, a list of requalification triggers and a review date. With those in place, a new employee, a new season or a supplier stock-out does not reopen the whole question.

The eight criteria, how each is tested and where it is decided

Each criterion has a test, a pass rule written before testing, and a gate where it is decided. The right-hand column points to the posts in this series that go deepest on each. None of the pass rules relies on a published performance figure, because CSI publishes none for throw, adhesion, frosting, hardness, heat survival or lot consistency; every one is judged against a control candle or your own logged evidence.

The candidates: what CSI publishes for each

Every comparison in this series uses CSI's container soy waxes as candidates to test, never as declared winners. Here is everything their product pages state that the system can use, re-checked on 1 October 2026. Adjectives on product pages — "glass adhesion", "low frosting", "stable in Indian climate", "strong throw" — are claims for Gate 1 and Gate 2 to test, not evidence.

Three CSI waxes sit outside the container-candidate set. Premium Coconut Soy Wax (₹460.00 for 500 g, ₹920.00 for 1 kg) is described on its page as soy with virgin coconut oil; Soy Butter Wax (₹548.70 / ₹1,097.40) as a blend of soy pellets, soy chunks and beeswax — so not 100% soy; and Soy Pillar Wax is a pillar and mould wax, not a container wax. None publishes a melt point, pour temperature, ceiling or shelf life. Soy Wax 464 Pellets were out of stock in every size on the 1 October re-check and are not offered.

Gate 1 — performance: screen four, take two forward

Start with a screen on a standard test candle — set up once and reused — so the four candidates are compared under identical conditions. The example standard is a 200 g candle in a White Matte Glass Jar with Lavender Fragrance Oil at 8% (16 g oil + 184 g wax on CSI's finished-weight convention) on an Eco Candle Wicks Thin C1 wick: 3 candles per candidate and 3 control candles in your current or reference wax, 15 in all. All four candidates' published ceilings that exist (Chunks 13%, CSI 468 13%, CSI 464 10%) cover 8%; Soy Pearl's is not published, so 8% is a labelled test point.

Take forward the two candidates that are no worse than control on every safety line and closest on the quality lines. Then test them in your real products with contrasting oils — the example adds Woody Oud to Lavender — 4 candles per oil per finalist (three burned, one stored), 8 per finalist. If a finalist tunnels or overfeeds on the shared wick, give both the same optimisation plan before deciding — fixed first, optimised second. If you are approving a second source for existing products rather than a new standard wax, SKU-by-SKU equivalence replaces this step. Two related questions from the first cluster sit here too: three waxes each best at a different thing and the qualification test for a standard wax.

Gate 2 — Indian heat: test your conditions, not a promise

CSI publishes no heat-survival data, and no wax should be approved for Indian summer on a product-page adjective. For each finalist, pour 2 extra candles per oil for heat and storage, and test them against conditions you have actually measured: a cheap min/max thermometer or data logger in your storeroom and in a parcel on your real courier route. If you need a controlled exposure, use a labelled example condition — for instance a chamber or closed car at about 40–45°C for six hours, an example to replace with your measured route — and always expose a control candle alongside. Test the candle in its retail packaging as well as bare. In October you are testing storage and autumn routes; the summer test must be scheduled for April–May if you ship through the hot months, and the wax's approval should say so.

Gate 3 — landed cost per saleable candle

Price per kilogram is an input. The decision is made on cost per saleable candle: (wax + oil + wick + jar) ÷ (1 − reject rate) + rework labour + freight. The table shows the four candidates at their 5 kg bag rates with each example oil at its 1 kg rate, the ₹5.90 wick and the ₹70.80 jar, and an example 3% reject rate — a placeholder for the rate your own pilot measures. Freight and rework labour are your figures; CSI does not publish delivery charges.

The spread across the four waxes is ₹40.25 per candle with Lavender. Two things can move that ranking after testing: a wax that reaches its best at a lower load saves ₹10.44 of Lavender per percentage point per candle, and a wax that produces more rejects loses its price advantage fast — the 5% defects question and cost at each wax's optimal load work through both. The cost cluster's hub sets out the full method, and the true commercial cost adds shipping and labour.

Gate 4 — procurement: packs, stock and the pilot

A wax that passes Gates 1–3 must also be buyable in a way that suits your cash, space and lead times. On CSI's listings: container wax bags stop at 10 kg; Natural Soy Pearl Wax goes to 50 kg (limited stock — confirm on WhatsApp); CSI lists no 25 kg carton and no 100 kg pack; larger quantities are quoted on WhatsApp with no published discount. Price per kilo moves differently by wax: Chunks' 10 kg bag (₹599.90 a kilo) is dearer than two 5 kg bags; CSI 464 is flat from 1 kg; CSI 468 steps from ₹650.00 to ₹637.00 at 5 kg; Soy Pearl from ₹450.00 to ₹424.80 at 5 kg.

Gate 4 also contains the pilot: one 10 kg bag of the winning wax at your normal batch size, about 52 candles of 200 g, with the first, middle and last pour pulled and burned beside bench candles. The pilot bag costs ₹5,074.00 for CSI 464, ₹4,248.00 for Soy Pearl, ₹6,370.00 for CSI 468, or ₹5,990.00 for Chunks as two 5 kg bags. The procurement cluster's hub sets pack sizes, reorder points and safety stock size your buffer, and what to validate before a 500 kg order covers very large commitments.

Gate 5 — supplier reliability and batch-to-batch QC

Gate 5 turns a good wax into a controlled raw material. It has four parts, each owned by a post in this cluster.

The whole programme on one page: candles, cost and calendar

Here is the full programme for the worked example, from screen to approval. Candle counts are labelled working practice. Stage 2 and 3 costs depend on which two waxes reach the final; the range shows the cheapest and dearest pair. Pilot wax is mostly saleable stock if it passes.

The cheapest finalist pair by test-wax cost is CSI 464 and Soy Pearl; the dearest is Chunks and CSI 468. That is a statement about the price of the test bags, not about which waxes deserve to reach the final — the screen decides that. On the calendar: started today, 1 October 2026, Gate 1 and most of Gate 2 can finish before Diwali on Sunday 8 November, but the summer half of Gate 2 cannot, and Gate 5's second-delivery check will land in the festive weeks. The practical plan is to approve for production with "summer pending" marked on the specification, buy festive wax after the pilot in 10 kg bags, and complete the summer checks in April–May before the hot shipping months. The November–February wedding season is a good period to accumulate delivery and lot evidence.

The decision rule — and what you buy at the end

When a CSI wax is approved, the first production purchase follows its price structure: CSI 464 in whichever bag size suits your storage, since every size from 1 kg is ₹507.40 a kilo; CSI 468 and Soy Pearl in 5 kg or 10 kg bags, where their per-kilo rates drop (₹637.00 and ₹424.80); Chunks in 5 kg bags at ₹2,995.00 rather than 10 kg. Soy Pearl's 50 kg bag (₹21,240.00) costs the same per kilo as its 5 kg and 10 kg bags, so take it only for handling convenience, and only after Gate 5 has seen two consistent deliveries. Quantities beyond the listed packs are quoted on WhatsApp.

The series map: every cluster and where it sits in the system

This post is the capstone of the hundred-question series. Each gate above is one cluster; each cluster has a hub that sets its method. Start with the hub for the gate you are at, then follow its siblings for your exact situation.

Frequently asked questions