How Do I Track Wax, Fragrance, Wick and Jar Variance Separately During a 5,000-Unit Candle Production Run?

The 5,000-unit run and its standards

Your figure is the title's: 5,000 units. The rest is a labelled worked example you can replace line by line. The candle is a 160 g fill in the clear 200 ml Salsa jar at 8% on the total-weight basis (12.8 g Fresh Strawberries + 147.2 g Premium Coconut Soy Wax CSI 468), with one Eco C2 wick, on the assumption that this formula and wick have already passed burn testing in this jar. The run makes 1,000 good candles a week, about 167 a day on a six-day week, for 5 production weeks, with delivery in week 6. Ten QC retains are poured in week 1 and logged on their own line, outside the weekly table.

Two details in that sheet matter for variance later. The jars are 53 × 100 + 2 × 5 rather than 54 × 100 (₹1,59,300.00), which is the cheapest way to 5,310, and the 5-pack price is already in it, so a tiny planned price premium exists from day one; the standard jar price is still ₹29.50, because that is what the bulk of the jars cost. And the wax and oil lines leave very little over (0.728 kg and 22 g), which is normal when you buy to the plan and means any usage variance will become a shortage. The weekly table exists to see that coming.

Usage variance and price variance are different questions

Mixing them hides both. Suppose wax costs you more this week. If the reason is that the floor used 2.8% more wax, the fix is on the floor. If the reason is that a top-up was bought as 1 kg bags at ₹650.00 instead of ₹637.00, the floor did nothing wrong and the fix is in purchasing. One number cannot tell you which.

Value usage at the standard price, always. If you value the extra 1 kg of oil at whatever the top-up happened to cost, a purchasing choice leaks into the production number and the two questions blur again. Price variance then picks up the difference between what was paid and the standard, separately.

On CSI prices, price variance is mostly a pack-size story

Because CSI publishes its pack prices and most lines here are flat per unit, price variance in a run like this comes almost entirely from pack choice. Eco C2 wicks are ₹9.44 each in every pack from 10 to 1,000, so wick price variance is nil whatever you buy. Fresh Strawberries is ₹3.54 a gram at 500 g and at 1 kg, so a 500 g top-up costs the standard; only 100 g bottles (₹3.78 a gram) or smaller would add price variance. CSI 468 has a small step (₹650.00 a kg in 1 kg bags against ₹637.00 from 5 kg). The clear Salsa jar has the biggest step: ₹40.12 in 5s against ₹29.50 in 50s and 100s. If you buy from several suppliers, price variance also captures their differences, and that is where it earns its column. For a deeper look at cost per gram and per candle across every ingredient, the cost-per-gram post has the method.

The weekly table, material by material

The weekly figures come straight from the per-batch log: wax and oil issued minus returned, pieces counted out of opened packs, good candles counted at QC. How to keep that log is the post on tracking consumption after every batch; this table rolls it up by material and week. Standard per week = 1,000 good candles × the standard per good candle.

Three things stand out. Usage variance is ₹10,087.68 and price variance ₹316.40; price is 3.0% of the total, which is what you would expect on mostly flat pack prices. Oil is only 8% of the weight of wax and oil, yet its usage variance (₹3,787.80) is close to the wax's (₹4,204.20), because every gram costs about five and a half times as much. And the pieces are small in rupees but not in consequence: 56 jars over standard is a shortage if you bought to the plan, and a jar shortage stops a run as surely as a wax shortage.

Reading the pattern across materials

The heatmap above is the weekly table at a glance. The cause usually reveals itself in which cells move together.

Week 4: every material up at once

Wax 2.8%, oil 2.7%, wicks 4.2%, jars 3.0%. When all four rise in the same week, the run made more candles than it kept: rejects. The QC notes for week 4 show a batch of wicks set crooked by a new helper, rejected at the centring check, wax dug out, jars cleaned where possible. Wicks are up most, because every rejected candle consumed one and none can be recovered; jars are up a little less, because some jars were cleaned and reused. The fix is a wick-centring check on the first ten candles of every session, not a bigger BOM. Incoming checks on the wicks themselves are in the post on spare wicks and the incoming wick check.

Week 3: oil alone

Oil 4.9% while wax is 0.8% and pieces are flat. When one material moves alone, the cause is in how that material is handled: here, oil added on top of the weighed charge in two melts "because the scent seemed weak in the pot". That is a change to the approved formula, not a variance to absorb. The practices to audit when oil runs over formula, from tare habits to load-basis mistakes, are in the post on fragrance consumption above formula.

Week 2: jars alone

Jars 2.3%, everything else normal. A carton of 22 jars was dropped in handling. That shows up as usage variance (jars consumed without candles) and, because the replacements were bought as 5-packs, as price variance as well. The usage line belongs to whoever moves cartons; the price line belongs to whoever bought 5-packs instead of holding a few spare from the 100s. Handling breakage on arrival is a different job, covered in the post on 15 jars arriving broken.

Do not net a good week against a bad one. Wax in week 2 was -0.2% and week 5 -0.1%, and on paper they cancel a little of week 4. They do not cancel the cause. Each week's cells are read on their own; the five-week totals are for re-forecasting and for the next quotation, not for excusing a bad week.

Turning variance into top-ups before they become stops

The purchase sheet left very little spare, so any cumulative usage variance above the planned loss becomes a shortage at the end of the run. The rule used here: when a material's cumulative usage variance passes +1%, re-forecast the rest of the run at the cumulative rate and top up the gap. The re-forecast method itself (remaining requirement at the actual rate minus stock on hand) is the cluster hub, the post on running short mid-production; here is how the variance table feeds it.

Every top-up except the week-2 jars was sized from the table and bought before the line ran out, so the run never stopped. The week-2 jars were replaced like-for-like (20 for the 22 broken) instead of re-forecast; the table said 65 more were needed, which is why jars came round again in week 4. In total the top-ups cost ₹14,923.20, including the week-2 jars. Two of them could have been cheaper: the wax as 5 kg + 3 × 1 kg (₹5,135.00) instead of 8 × 1 kg, a price variance of ₹39.00 instead of ₹104.00; and the jars by holding a few spares from the opening 100s rather than buying 5-packs on the day.

Argued against our own interest: when the table says top up, buy what the re-forecast says, not a round "safety" quantity. A 10 kg bag of CSI 468 for a 7.7 kg gap would leave a couple of kilos you may not use soon, for no per-kilo saving over 5 kg; a second 1 kg of oil "just in case" is ₹3,540.00 of a client's scent sitting on your shelf. Re-check after the next week instead.

What must not be skipped, and its calendar time

A top-up is very likely a different lot from the opening stock. Before the 1.5 kg of oil or the 8 kg of wax goes into the order, pour a few candles from the old and new stock side by side, compare colour and cold scent once set, and burn them after a cure; many makers allow at least 48–72 hours for soy before judging a burn (common practice; the Fresh Strawberries page states no cure time in the text we captured). That is three to four calendar days, which is why the re-forecast happens a week before the shortage, not the day of it. Lot checks are covered in the post on a new lot of the same fragrance and the post on managing multiple wax lots.

Commercial protection (practical judgement, not legal advice): the client pays per candle, so every rupee of variance is yours. If a variance traces to a client change (a different fill, a heavier load "for more scent", a second colour), it is not variance; it is a change, and it should be re-quoted in writing before production continues. Keep the weekly table with the order file: it is your evidence for the next quotation, which is where the five-week totals finally get used. The post-order review is the post on comparing estimated and actual consumption after 5,000 candles, and turning these figures into standing loss percentages is the post on incorporating a historical variance into future BOMs.

The weekly routine takes about ten minutes once the batch log exists:

The four materials and where their variance comes from

Chosen on page statements, price and stock on 7 October 2026. Each card says where that material's price variance can and cannot come from at CSI prices, which is what decides whether its line needs a price column at all.

The honest summary: at CSI's published prices, a disciplined run has almost no price variance; nearly all of the money is usage. Track usage closely, and keep price variance as a short column that only fills when someone buys a small pack in a hurry.

Frequently asked questions