My Fragrance Oil Worked in a Small Test but Performs Differently in Bulk Production — Why?

Eight things that change when 2 candles become 40

Candle recipes scale linearly on paper: 20 times the wax, 20 times the oil. Heat does not. A larger mass of wax heats and cools more slowly, holds heat unevenly, and takes longer to pour, and each of those is a variable your small test never had. The effects below are tendencies that vary by wax and oil; CSI has no measured scale-up data. What we can say with confidence is that each one is measurable, and that a test candle and a production candle are only comparable if you know how each was made. (If the problem is that some oils fail in your formula while others pass, even in small tests, that is a different question, answered in the hub for this cluster.)

1. Heat-up and hold time

A jug of 300 g melts and is poured within minutes. A pot of several kilograms takes much longer to melt and is often held hot while jars, wicks and labels are set out. Too-cold addition can give poor incorporation; long holding at high temperature tends to drive off lighter top notes and may darken some oils. The fix is not a new number from us: add the oil within your wax page's stated range (CSI 464: pour 80°C–90°C; Eco Soy CSI 400: add fragrance at 85°C), and write down how long the scented wax waited before the first pour. Whether a given add temperature was wrong in the first place is covered in Did I Add Fragrance Oil at the Wrong Temperature?.

2. Temperature spread

In a jug, one reading describes all the wax. In a pot, the bottom near the element, the top and the edges can differ, and the whole pot drifts as you pour. Read after stirring, at mid-depth, and write it down.

3. Mixing

A spoon in a jug reaches every gram. In a pot, oil added at the surface has to reach the bottom and edges. Stir for a fixed time, scrape the bottom and sides, and use the same stir time every batch. A pot that is stirred for 30 seconds one week and 3 minutes the next is two processes.

4. Pour span

Forty jars take time. The first jar gets wax at the add temperature; the last gets wax that has cooled and waited. If your complaints come from the last trays, not the first, this is the prime suspect. Read the temperature at the first and last jar; if the gap is large, pour in smaller portions from a Steel Pouring Pitcher (Large, 600 ml, ₹708.00; the Small is out of stock on 25 September 2026) refilled from a pot kept within the page range.

5. Cooling on the tray

A single test jar cools with air all round it. Forty jars packed on a tray commonly cool unevenly, the centre ones more slowly than the edge ones, and a fan, AC vent or open window across the tray changes that further. Uneven cooling tends to show as surface differences more than scent. Space the jars, keep draughts off, and note which tray position each labelled jar came from.

6. Cure conditions

Your test candle cured on a shelf you walked past every day. Your production batch may cure in cartons in a store room that is warmer in April to June and damper through the monsoon, roughly June to September. Soy commonly needs a cure period before hot throw is judged, and the conditions of that period are part of the process. If production is judged earlier than the test was, or cured somewhere different, you are not comparing like with like.

7 and 8. The wax and the oil bottle

Production uses a new bag of wax and, often, a new bulk bottle of oil. Either could differ from what you tested. The oil check is below; for the wax, pour your twin candle from the same bag as the batch so the wax is a constant within the bridge test.

Is it actually the fragrance oil? The kept-sample check

The only fair way to ask "is the bulk oil different from the sample?" is to pour both in the same melt. Weigh 12 g from the sample bottle and 12 g from the bulk bottle, each into 138 g of the same wax, add both at the same temperature, stir for the same time, pour both into the same jar type with wicks from the same pack, cure them together, and burn them on separate evenings in the same closed room. Before pouring, smell both on paper strips side by side: a clear difference in the bottle is worth a WhatsApp message on its own.

The check costs ₹340.64 in wax, jars and wicks, plus 24 g of your own oil. If you did not keep any of the sample, buy the same size again only if the bulk bottle genuinely smells different; otherwise the process is the likelier cause and the money is better spent on the thermometer. For future orders, keep 20 to 30 g of every approved sample in a sealed, labelled bottle away from light and heat; how to build that into approval is in How to Qualify a New Fragrance Oil Before Adding It to Your Candle Business, and the repeat-order checks for large orders are in What Tests Should I Run Before Ordering 5 Kg of a Candle Fragrance Oil?.

The bridge batch: first, middle and last jar

The bridge batch is a normal production pour with four extra labels and one extra candle. It changes one thing relative to your approved test, the batch size, and records where in the batch each difference appears. You sell the candles that pass, so the real cost of the check is the twin candle and the time spent writing things down.

Oil: 41 × 12 g = 492 g, leaving 8 g in a 500 g bottle. Wax: 5,658 g. Cure T, jar 1, jar 20 and jar 40 together in the production store room, not on your test shelf, so cure conditions are the same for all four.

With Roasted Coffee 500 g instead, the same 41 candles come to ₹8,873.17, ₹216.42 per candle. All figures exclude labour, freight, packaging and GST. The batch is production you intend to sell; what you are really paying for the diagnosis is the twin and an hour of labelling.

The batch log that makes production repeatable

Every scale-up diagnosis above depends on numbers someone wrote down. Keep one line per batch, filled in as you go, not afterwards. A full testing SOP with forms and roles is in How to Build a Fragrance Oil Testing SOP for a Candle Business; this is the minimum log a production batch needs to be comparable with its test.

Where a melter helps, and where it does not

Once batches pass a few kilograms, many makers move from a double boiler to an electric melter, and two reviews on CSI's store describe exactly that: one customer says 10 kg melts "in mere 20-30 minutes", another describes a mini melter helping in a very cold region where her temperature "would drop within seconds". Those are real experiences and real benefits. But a melter does not make a batch behave like a test; it makes a big batch easier to hold at a set point, and holding is one of the variables above.

Read the electric wax melter's page wording closely: a display from 60°C to 100°C and a stated "±10°C deviation". That band is wider than CSI 464's entire 80°C–90°C pour window, so the display alone cannot tell you whether you are inside it: a pen thermometer can. Its melting-speed figure is quoted at 100°C, above that wax's stated pour range; melting speed is not a pour instruction. For jug-sized work such as twin candles and small tests, the Mini Electric Wax Melter (1 litre, ₹2,360.00) is the smaller option; it is not a production pot.

What to buy: tools first, oils to scale with care

Ranked by how often, in our judgement, each is the right purchase for a scale-up problem. The two oils are not replacements for yours; they are examples of the two ends of scale-up risk (a light, top-led oil exposed to long holds, and a dense gourmand whose production differences tend to show in cure and colour), chosen on stated notes, stated load and cure wording, price per gram and stock at every size on 25 September 2026, not on throw, which CSI has not measured. Candle counts assume 150 g at 8%.

Scaling up before Diwali, 8 November 2026

Diwali falls on 8 November 2026, 44 days after these prices were checked, and for many makers it is the largest pour of the year, which is where scale-up problems show. A bridge batch fits: pour it in the first week of October, judge at day 7 and day 14, adjust one step, and run full production in the third week of October with two weeks of cure before dispatch. If the bridge batch passes, it is also your first production run, not a lost week. Oil per 8% candle at the 1 kg price: Cucumber Green Tea ₹39.93, Roasted Coffee ₹44.86; a kilogram makes 83 candles of 150 g. Whether an oil deserves 1 kg of inventory at all is covered in our guide to choosing which oils deserve 1 kg quantities. Dispatch times depend on your location; confirm on WhatsApp before you promise a customer a date.

Frequently asked questions