My Vanilla Candle Keeps Getting Darker the Longer It Sits — How Do I Determine Whether the Colour Will Eventually Stabilise?
Share
Vanilla oils from ₹106.20 / 15g · 150g clear jars ₹94.40 each 62 of 107 oil pages publish a discolouration position Stocked in India · ships nationwide
CSI stability guides · colour over time
Nobody can tell you when the browning stops. What you can do is find out whether it is slowing down — and that takes eight readings, not an opinion.
★★★★★
"Frangance is sooo good"
Sneha TamangVerified buyer · May 2025
Cinnamon Vanilla
★★★★★
"Very good"
Tina Lynette Lazaro MogantiVerified buyer · Dec 2025
Black Matte Glass Jar
★★★★★
"…M happy with the smell of dark vanilla M obsessed with the vanilla fragrance…"
SUMAN BISHTJul 2025
Dark Vanilla Oud
★★★★★
"All the products are so good"
Jinal PatelVerified buyer · May 2025
Luxury Soy Wax Chunks
★★★★★
"The jars were of good quality"
Soumili GopeVerified buyer · Mar 2026
Clear Glass Jar with Golden Lid
★★★★★
"Quality wick, works very well"
AshwiniSep 2025
Eco Candle Wicks
★★★★★
"Frangance is sooo good"
Sneha TamangVerified buyer · May 2025
Cinnamon Vanilla
★★★★★
"Very good"
Tina Lynette Lazaro MogantiVerified buyer · Dec 2025
Black Matte Glass Jar
★★★★★
"…M happy with the smell of dark vanilla M obsessed with the vanilla fragrance…"
SUMAN BISHTJul 2025
Dark Vanilla Oud
★★★★★
"All the products are so good"
Jinal PatelVerified buyer · May 2025
Luxury Soy Wax Chunks
★★★★★
"The jars were of good quality"
Soumili GopeVerified buyer · Mar 2026
Clear Glass Jar with Golden Lid
★★★★★
"Quality wick, works very well"
AshwiniSep 2025
Eco Candle Wicks
Reviews collected and published through Judge.me on candlemakingsuppliesindia.store. One longer review is shortened with an ellipsis; wording is otherwise unedited.
✓ Vanillin percentages published per oil where stated ✓ 106 fragrance oils listed ✓ Bulk pricing, GST invoicing & documentation on WhatsApp
Candle Business Guides · Discolouration
You cannot determine it from the candle in your hand, and no product page anywhere states an endpoint. You determine it by measuring the same units at fixed intervals and watching whether the change between readings is shrinking.
8 checkpoints
days 0, 7, 14, 30, 60, 90, 120 and 180 · 12 retained units, 6 with the inventory and 6 in a cool dark control · ₹356.43 of held stock per reading on a 150g candle · our method, not an industry standard · CSI live pricing, September 2026
Quick answers — read this first
Will the colour stabilise? Very likely it will slow down, and nobody can honestly tell you when. No CSI page states an endpoint, a rate or a shelf life, and there is no published conversion from a warm week to a month at ambient. "Will it stabilise" is not a lookup — it is a question your own log answers.So what do I actually measure? The change between consecutive readings, not the total change since pour day. If day 60 to day 90 moved less than day 30 to day 60, the process is decelerating in your conditions. If every interval moves about the same, it has not settled yet.
What do I need? 12 units held back — six on the shelf with the batch, six in the coolest darkest place you have — photographed together against the same white card at the same eight checkpoints. On a 150 g candle in a Clear Glass Jar with Golden Lid that is ₹2,851.44 of withheld stock, or ₹356.43 per reading.
Can I sell in the meantime? Yes, if the units pass a physical inspection and burn correctly. What you cannot do is print a colour promise, or a shelf-life figure, on the label before your own log supports it.
The short answer
Measure intervals, not totals: the only observable that answers "is it settling" is whether each checkpoint moved less than the one before. A total shift from pour day tells you nothing about what happens next.
You need a control or you have nothing: six units in a cool dark box are what let you say the shelf did it, the oil did it, or both. Without the paired set, a reading at day 90 cannot be attributed to anything.
Read the page for the material, then stop expecting numbers: Vanilla Bean and Cinnamon Vanilla both state 2.5% vanillin. That is a composition figure. It is not a prediction of shade, speed or endpoint, and nothing on any page is.
In this guide
The schedule is ours, not a standard, and the three 30-day intervals are the part that does the work. Comparing a 7-day interval with a 60-day one tells you nothing; comparing day 30–60 with day 60–90 and day 90–120 tells you whether the change is decelerating in your own conditions.
Page states 2.5% vanillin
Cinnamon Vanilla · 15 gm ₹106.20
The jar you are watching through
Clear Glass Jar 150 gram · ₹94.40 each
Setting up a stability log?
Ask on WhatsApp
Straight answer
How do I find out whether my vanilla candle's colour will eventually stop changing?
By turning the question into intervals. Hold back 12 units from the batch, split them six and six between the shelf your stock actually sits on and the coolest darkest place you have, and photograph all twelve together against the same white reference card at days 0, 7, 14, 30, 60, 90, 120 and 180. Then ignore the total shift since pour day — it is the wrong measurement — and compare the three matched 30-day intervals: day 30 to 60, day 60 to 90, day 90 to 120. If each moves visibly less than the one before, the change is decelerating in your conditions and you can describe the product accordingly. If all three move about the same, it has not settled and any number you print on a label is a guess. What nobody can give you is the answer in advance: Vanilla Bean Fragrance Oil states "Vanillin Content: 2.5%" and Cinnamon Vanilla states the same figure, but a composition percentage is not a rate, and zero CSI pages state a shelf life, a storage instruction or an ageing curve. On a 150 g candle in a Clear Glass Jar with Golden Lid the retained set costs ₹2,851.44 of stock — ₹356.43 per reading, or 2% of a 600-unit batch worth ₹1,42,571.76 in materials. Meanwhile, keep selling: inspect, burn two units, and if the burn is unchanged the colour is a description problem, not a stock write-off.
One line: stop measuring how much it has changed and start measuring whether each 30-day interval changed less than the last — twelve retained units and eight photographs are the whole instrument.
Building the retained set from scratch rather than from stock? Luxury Soy Wax Chunks 1 kg at ₹599, Clear Glass Jar with Golden Lid 150 gram, 10 for ₹944, Cinnamon Vanilla 100 gm at ₹483.80 and Eco Candle Wicks Thin (C1), 20 for ₹118.
Shop containers
The verdict on the stock in hand
The colour is still moving. That is uncomfortable, and it is almost never a reason to stop dispatching.
A candle whose wax is deepening month on month is doing something visible, and the instinct is to treat visible change as deterioration. Separate the two. Deterioration is a burn fault or a physical fault. Take six units out of the carton, inspect them, and burn two for two hours each. If the pool opens to the glass in the same time it did when fresh, the flame is steady, there is no soot and the jar is comfortable to hold, the batch is functionally the product you made — in a darker shade.
What changes is what you are allowed to say. Three sentences are now off the table until your log supports them: "the colour is stable", "the colour will settle by month four", and any printed shelf life. All three are predictions, and you do not have the data for a prediction yet. What you can say immediately, and should, is descriptive: this fragrance contains vanillin and the wax deepens in tone over time. That is the wording the product page itself uses, and it is accurate on day one.
Dispatch decision while the log is still running
What to check on a six-unit sample, and what each finding means for the carton
| Check | Method | Pass looks like | If it fails |
|---|---|---|---|
| Glass integrity | Rotate each jar under a lamp, run a fingertip round the base | No crack, chip or star | Quarantine those units and check the whole carton |
| Wick position | Look straight down into the jar | Wick central, not leaning on the glass | Quarantine — do not ship a wick against glass |
| Burn behaviour | Two units, two hours each, in a still room | Full pool to the glass in the same time as fresh, steady flame, no soot | Re-wick and re-test before any of the batch moves |
| Jar temperature | Touch the outside of the glass at two hours | Warm, comfortable to hold | Drop a wick size; this is a safety check, not a colour check |
| Surface | Look across the top at a low angle | Dry, no droplets or slick | Sweating is a separate stability question and takes priority |
| Colour | Photograph beside a white card and beside your listing image | Anything, as long as the listing matches | Update the listing image and the description — not the candle |
The mistake: waiting for the colour to settle before you ship. Makers hold a whole Diwali run back "until it stops", and because nothing tells them when that is, they hold it through the season and sell nothing. Diwali 2026 falls on 8 November, and stock that misses it is worth a fraction of stock that meets it. Ship the units that pass the physical checks, describe the colour honestly, and run the log on twelve retained units in parallel. The log informs next year's formula and next year's label — it is not a gate on this year's dispatch.
What "stabilise" can honestly mean
The word carries three different meanings and only one of them is measurable, which is why this question feels unanswerable. Pull them apart before you design the test.
Three readings of one word
What each version of "stabilise" would require, and whether you can get it
| Version | What it would mean | Can you establish it? | What it needs |
|---|---|---|---|
| Stops entirely | The wax reaches a final shade and never moves again | No. Proving a change has permanently ceased requires an indefinite observation window | Nothing you can buy. Drop this version of the question |
| Slows down measurably | Each equal-length interval shows less movement than the one before it | Yes. This is the one your log answers | Three matched 30-day intervals on retained units against a control |
| Stops mattering commercially | The shade stays inside the tolerance you have set and described to customers | Yes — and it is the version your customer actually cares about | A written tolerance, a reference photograph, and listing copy that matches |
The third version is the commercially useful one, and it is the version nobody sets up because it requires you to decide something. A tolerance is a sentence like: "this candle is poured cream and deepens towards a light tan within the first few months; units outside this pair of reference photographs are not sold as part of the white range." Once that sentence exists, "will it stabilise" stops being an anxiety and becomes a pass-or-fail against a line you drew. Where that line belongs is your call and nobody else's — not ours, and not a supplier's.
Why total change is the wrong number
Suppose you note the candle is "two shades darker" at day 90. That single figure is compatible with a process that finished in week two and a process that is still running hard, and it cannot distinguish them. The same total can be reached by a big early move and a flat tail, or by a steady march that will keep marching. Only the sequence of interval differences carries the information you want — which is why the schedule has three equal 30-day gaps in the middle rather than an assortment of convenient dates.The eight checkpoints and the log sheet
A log sheet is what stops the test from turning into an impression. One row per unit per checkpoint, and every column below is something you can fill in without equipment. Keep it on paper beside the carton or in one spreadsheet — not in a chat thread, where the photographs get compressed and the dates get lost.
Colour log · one row per retained unit per checkpoint
Every column, why it is there, and what it costs you to record
| Column | What goes in it | Why it earns its place |
|---|---|---|
| Unit ID | S1–S6 for shelf, C1–C6 for control | Without IDs you cannot track one unit's trajectory, only the group's mood |
| Checkpoint day | 0, 7, 14, 30, 60, 90, 120, 180 | Actual day, not "about three months" — intervals only compare if they are real |
| Photograph reference | File name of the shot with the white card in frame | The photograph is the measurement; the note is only an index to it |
| Shade note | Cream / warm cream / ivory / light tan / tan, plus "vs previous: same, slightly darker, clearly darker" | The second half is the interval reading — the only field that answers the question |
| Evenness | Uniform / darker on one face / darker near the top / ring near the surface | Uneven change moves the diagnosis away from the fragrance and towards light or migration |
| Weight | Grams on the same scale, lid on | A unit that has lost mass has lost something; it costs five seconds and it rules things in |
| Surface | Dry / droplets / slick / frosted / cracked | Catches a second symptom developing while you were looking at colour |
| Where it sat | Shelf position or "control box", plus anything that changed | A carton moved to a sunnier wall in April is the kind of thing you forget by September |
| Burn result | Only at day 90 and day 180, on one unit from each arm | Appearance testing alone is not enough — the aged unit has to be burned |
Nine columns, eight checkpoints, twelve units. That is 96 rows over six months, which sounds like a lot until you notice each row takes under a minute and seven of them are the same photograph. The expensive part is not the recording, it is the twelve units of sellable stock sitting in a box — and that is a real cost, which the arithmetic section prices properly.
Building and reading the retained set
The retained sample is the single most useful instrument in this whole subject: units held back from a real batch, stored two ways, so that months later you are comparing rather than remembering. Here is the build.
1
Pull twelve units at random from the batch, not twelve from the top of the cartonTake them from different cartons and different positions — top, middle and bottom of the stack. Units from one position share a history, and a set drawn from one corner of the store answers a question about that corner. Label them S1–S6 and C1–C6 on the base with a marker, and write the batch number and pour date on each.
2
Record the recipe and the oil page's own wording on the sheet headerWax and pack size, oil and exact size bought, load as a percentage, add temperature, pour temperature, wick, jar, cure length. Then copy your oil's published vanillin or discolouration sentence underneath — for Cinnamon Vanilla that is "Vanillin content 2.5%". If the page says nothing, write "page silent". This header is what makes the log usable by somebody other than you.
3
Fix one photographic setup and never change itSame wall, same lamp, same distance, same white card in frame, lids on, jars in the same order. Write the checkpoint day on the card in the shot. Daylight through a window is the worst choice available because it changes hour to hour and season to season. A single lamp in a room with the curtains shut is boring, repeatable and correct.
4
Split the set: six on the real shelf, six in a cool dark closed boxS1–S6 go back onto the shelf the batch actually lives on, at the positions you drew them from. C1–C6 go into a closed carton in the coolest, darkest place you have. The control is not a "better" storage recommendation — it is the reference that lets you separate what the fragrance did from what the shelf did. Without it, a reading at day 120 cannot be attributed to anything.
5
At each checkpoint, photograph all twelve together — then weigh themBring the control units out, shoot all twelve in one frame, put the controls straight back. Weigh each one lid-on on the same scale and write the grams down. Two minutes of extra work per checkpoint gives you a second, independent signal that a colour photograph cannot give: whether anything has left the candle.
6
Burn one from each arm at day 90 and day 180Two hours, still room, no competing scent. Record time to a full melt pool, flame height, soot, mushrooming, and your own throw score at 30 and 120 minutes. This is the step that turns a colour log into a stability programme: a batch that darkened and burns identically is a very different asset from one that darkened and lost its pool.
7
At day 120, line up the three 30-day intervals and make the callPut the day 30, 60, 90 and 120 photographs side by side in one image. Compare gap to gap, not first to last. Then write one sentence in the log: "movement decelerating" or "movement steady" or "no detectable movement since day X". That sentence is the deliverable. It is also the only defensible basis for anything you later print about colour.
Do not try to speed the log up with a warm cupboard. Heat-ageing a sample is a legitimate technique for finding a formulation's practical limit, and it is covered properly in the sweating cluster. What it cannot do is tell you how many real months your warm week represents. There is no published conversion — not on CSI's pages, not anywhere that is honest about it — and a candle held at an elevated temperature is being asked a different question from one sitting on your shelf. If you want to know what six months on your shelf does, the only instrument is six months on your shelf. Run a heat-aged arm as well if you like, but log it as its own arm and never translate its results into calendar time.
Three results and what each one tells you
At day 120 the log will read one of three ways. Each one points at a different next action, and none of them requires you to know why.
Reading the three matched intervals · day 30–60, 60–90, 90–120
What the pattern means, what to do about the batch, and what to change next time
| Pattern across the three intervals | Honest conclusion | The batch in hand | Next formula or purchase |
|---|---|---|---|
| Each interval clearly smaller than the last; shelf and control converging | Change is decelerating in your conditions. Still not a promise that it has stopped | Sell with a descriptive line; photograph the day-120 shade for the listing | No change needed. Keep the oil, keep the jar, keep the log for next year |
| All three intervals about equal; shelf and control both moving | Not settling yet on this schedule. You have no basis for a colour claim | Sell on burn performance; describe the colour as deepening over time | Test 8% against 10% side by side, and price an opaque jar |
| Shelf units moving, control units flat or nearly flat | The shelf is contributing — this is an exposure or temperature finding, not only a fragrance one | Move the remaining stock into closed cartons away from light and heat | Change the storage first; it is cheaper than changing the formula |
| Both arms flat from an early checkpoint onward | The colour moved once and settled in your conditions | Re-shoot the listing image at the settled shade and carry on | Nothing. Record the settled shade as your reference for this recipe |
| One or two units diverging wildly from the rest of their arm | Within-batch inconsistency — points at mixing or position, not the formula | Widen the inspection; sample more units per carton | Fix the pour process: stir time, portioning, and where trays cool |
Notice how often the answer is not about the fragrance. Two of the five rows send you to storage or to process, and only one sends you back to the formula. That is the real value of running a control alongside: it stops you from changing an oil that was never the problem. If the shelf-versus-control split is what your log shows, the light-exposure route in the side-lit candle guide is where to go next.
₹356.43 a reading: the arithmetic
Cost the log properly and it stops feeling like homework. Take a 600-unit run of 150 g candles in Clear Glass Jars with Golden Lid, poured from Luxury Soy Wax Chunks with Cinnamon Vanilla at 8%, on a Thin (C1) Eco wick. Materials only.
One candle, one batch, one retained set · CSI live pricing, September 2026
What the log costs expressed per reading, per unit and as a share of the batch
| Line | Basis | Cost |
|---|---|---|
| Wax in one 150 g candle | 138 g at ₹599/kg = ₹0.599/g | ₹82.66 |
| Oil in one candle | 12 g of Cinnamon Vanilla at 1 kg pricing (₹4.5548/g) | ₹54.66 |
| Jar | Clear Glass Jar with Golden Lid 150 gram, pack of 10 at ₹944 | ₹94.40 |
| Wick | Eco Candle Wicks Thin (C1), 100 at ₹590 | ₹5.90 |
| Materials per candle | Sum of the four lines | ₹237.62 |
| 600-unit batch | 600 × ₹237.62 | ₹1,42,571.76 |
| Six shelf units held back | 6 × ₹237.62 | ₹1,425.72 |
| Six control units held back | 6 × ₹237.62 | ₹1,425.72 |
| Retained set, 12 units | 2% of the batch's materials value | ₹2,851.44 |
| Cost per checkpoint | ₹2,851.44 ÷ 8 readings | ₹356.43 |
| Two burn units consumed at day 90 and 180 | 4 × ₹237.62 across both checkpoints | ₹950.48 |
Worked out above: 150 g candle at an 8% load = 138 g wax + 12 g oil. Wax ₹0.599/g × 138 = ₹82.66 · oil ₹4.5548/g × 12 = ₹54.66 · jar ₹944 ÷ 10 = ₹94.40 · wick ₹590 ÷ 100 = ₹5.90 · total ₹237.62. Batch 600 × ₹237.62 = ₹1,42,571.76. Retained set 12 × ₹237.62 = ₹2,851.44, which is 2.00% of the batch. Per checkpoint ₹2,851.44 ÷ 8 = ₹356.43. Four units burned across the two burn checkpoints = ₹950.48. With Vanilla Bean at ₹7.646/g the oil line would be ₹91.75 instead of ₹54.66 and the per-candle figure ₹274.71.
Eight readings for less than the cost of twelve candles, and ten of the twelve come back into stock at the end if you want them. Set against a batch worth ₹1,42,571.76 in materials, the log is 2% of the batch — and it is the difference between telling a retail buyer "our candles are colour-stable, we think" and handing them a dated photographic series. The second one wins shelf space.
1
Numeric position · 2.5% vanillin stated
Cinnamon Vanilla Fragrance Oil
Worth logging precisely because the page gives you a number rather than a warning: "Vanillin content 2.5%", alongside a stated flashpoint of 270°F (132°C). A stated composition figure is the right starting point for a stability sheet — it is a fact about the oil that will not change between orders, which makes your log comparable across batches. At ₹4,554.80 per kilo it is also the least expensive way to run a vanilla-carrying arm: ₹54.66 of oil in a 150 g candle against ₹91.75 for Vanilla Bean. Two CSI customers keep it short — "Frangance is sooo good" and "Will buy more.. wonderful".
| Size | Price | Per kg | 150g candles @ 8% |
|---|---|---|---|
| 15 gm | ₹106.20 | ₹7,080 | ≈1 |
| 50 gm | ₹271.40 | ₹5,428 | ≈4 |
| 100 gm | ₹483.80 | ₹4,838 | ≈8 |
| 500 gm | ₹2,301 | ₹4,602 | ≈41 |
| 1 kg | ₹4,554.80 | ₹4,554.80 | ≈83 |
₹54.66 of oil per 150g candle at 8% and 1 kg pricing Buy Cinnamon Vanilla
2
The reference vanilla · 2.5% stated, 203°C flash point
Vanilla Bean Fragrance Oil
The straight vanilla in the range, and the other page that publishes a figure: "Vanillin Content: 2.5%" with a measured flashpoint of 203°C. If you are choosing between two vanillas for a range you intend to hold for months, run both arms in the same log — same wax, same jar, same day, one variable — rather than reading the percentages and assuming they behave identically. Two oils can share a stated vanillin figure and still differ in how they read in your wax, because the rest of the composition is not published. That is a paired comparison, and it is cheap: one extra six-unit arm.
| Size | Price | Per kg | 150g candles @ 8% |
|---|---|---|---|
| 15 gm | ₹130 | ₹8,666.67 | ≈1 |
| 50 gm | ₹410 | ₹8,200 | ≈4 |
| 100 gm | ₹790 | ₹7,900 | ≈8 |
| 500 gm | ₹3,840 | ₹7,680 | ≈41 |
| 1 kg | ₹7,646 | ₹7,646 | ≈83 |
₹91.75 of oil per 150g candle at 8% and 1 kg pricing Buy Vanilla Bean
3
Takes the colour question off the customer's table
Black Matte Glass Jar
If the log says the movement is not decelerating, the cheapest honest response is usually to stop showing the wax. A matte black vessel at ₹70.80 each in the pack of 10 costs ₹23.60 less than the 150 gram clear jar and removes the mismatch between your photograph and the parcel entirely. Keep a small clear-glass arm in your retained set anyway, because a log that can no longer see the wax stops being a log. One CSI customer's verdict on these is simply "Very good". Note the capacity is stated in millilitres rather than grams, so weigh one filled jar and re-test the wick before you switch a whole range.
| Size | Price | Per piece | Candles per pack |
|---|---|---|---|
| Pack of 5 | ₹354 | ₹70.80 | ≈5 |
| Pack of 10 | ₹708 | ₹70.80 | ≈10 |
₹70.80 per jar · ₹23.60 less than the 150 gram clear jar Buy Black Matte Jars
The CSI principle
You cannot measure whether something has stopped. You can measure whether it is slowing down — and that is enough to run a business on.
Three equal intervals, twelve units, one white card and the same lamp. Everything else about colour stability is either a number somebody invented or a promise nobody can keep.
Running a log across two vanillas at once? Cinnamon Vanilla 100 gm at ₹483.80 and Vanilla Bean 100 gm at ₹790 cover a six-unit arm each in 150 g candles. Bulk pricing, GST invoicing and documentation are handled directly on WhatsApp.
Order in bulk on WhatsApp
What the log changes about your next purchase
A log that only ends in a conclusion is half a log. It should end in an order.
Every outcome in the table above has a purchasing consequence, and they are not all "buy a different oil". Ranked by how much they cost you to act on, cheapest first: change the storage; change the listing copy; change the load; change the vessel; change the fragrance. Most makers reach for the last one because it feels decisive, and it is the most expensive and least often correct.
Changing storage costs the price of cartons and a different shelf. Changing the listing costs an afternoon with a camera. Changing load from 10% to 8% actually reduces your oil cost — on a 150 g candle with Cinnamon Vanilla, 15 g of oil at 10% is ₹68.32 against ₹54.66 at 8%, so the test pays for itself if the lower load holds up on throw. Changing the vessel to an opaque jar saves ₹23.60 a unit at the same time as solving the appearance question. Only when all four of those have been tried does replacing the fragrance make sense — and at that point the screening method in the white-collection screening guide is how you choose the replacement, and the blank-test guide is how you confirm the old one was the culprit at all.
Three photographs thirty days apart are worth more than every confident number you will ever be told about how candles age.
— CandleMakingSuppliesIndia
Why trust this guide
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished candles. The easy version of this page would tell you that vanilla discolouration settles after roughly two months and sell you a jar. We will not, because we have no data for it and neither does anyone else who says it. What we can do is publish what our own product pages state, tell you plainly where they are silent, and hand you a protocol that produces your own numbers — including the protocol's cost, so you can decide whether it is worth running.
Every price, pack size and specification here comes from live CSI stock in September 2026 and links to the product page, where current pricing always takes precedence. Vanillin figures and flash points are quoted only where the page states them — 2.5% on Vanilla Bean and Cinnamon Vanilla, flash points of 203°C and 270°F (132°C) respectively. No CSI fragrance page states a shelf life, a storage instruction, an ageing rate, a discolouration endpoint or an accelerated-to-real-time conversion, and none is asserted here. The eight-checkpoint schedule, the twelve-unit retained set, the 8% planning load and the 2-hour burn window are our method, labelled as a method throughout.
For help designing a stability log for your own range, for bulk pricing, GST invoicing, MSDS or IFRA documentation, message us on WhatsApp at +91 7397976926.
Frequently asked questions
How long does vanilla discolouration take to stop in a soy candle?
There is no published figure, and any specific answer you are given is invented. No CSI page states a rate, an endpoint or a shelf life, and the honest reason is that it depends on the oil, the load, the wax, the vessel and the storage — five variables that differ between every maker. What you can establish is whether the change is decelerating in your conditions, by comparing three equal 30-day intervals on retained units against a cool dark control. That is a measurement. A month count is not.
How many candles should I hold back for a colour log?
Our method is twelve: six stored with the inventory and six in a cool dark control, drawn at random from different cartons and different stack positions rather than from the top of one box. Twelve is a defensible choice rather than a standard — it gives you six replicates per arm, which is enough that one odd unit does not decide the result. On a 150 g candle at ₹237.62 of materials that is ₹2,851.44, about 2% of a 600-unit batch, or ₹356.43 per reading across eight checkpoints.
Can I speed up the test by storing samples somewhere warm?
You can run a heat-aged arm, and it is useful for finding a formulation's practical limit — but you cannot convert its results into calendar months. There is no published conversion between accelerated storage and real time, on CSI's pages or anywhere honest, and a candle held warm is being asked a different question from one on your shelf. Log the heat-aged arm separately, never translate it into "equivalent to X months", and keep the shelf arm running for the real answer.
Should I put a colour-stability claim or a shelf life on my label?
Not until your own testing supports it. A printed claim is a promise to every customer who buys that batch, and no supplier page gives you a basis for one — zero CSI fragrance pages state a shelf life or a storage instruction. What you can print immediately is descriptive rather than predictive: that the fragrance contains vanillin and the wax deepens in tone over time, which is the wording the product pages themselves use. Add a dated reference photograph to your listing and you have said something true and useful.
Does a higher vanillin percentage mean faster darkening?
It is a reasonable expectation and it is not something CSI's pages state. What they publish is composition: 2.5% on Vanilla Bean and Cinnamon Vanilla, 0.2% on Gardenia, 0% on Sandalwood and its group, and "Present" on Dark Vanilla. No page relates a percentage to a speed or a final shade. If the ranking matters commercially, run the two oils as paired arms in one log — same wax, same jar, same day — and rank them on your own readings.
My candle darkened and the smell changed too — is that the same problem?
Treat them as two findings until your log says otherwise. Colour and perceived scent can move independently, and cold throw in particular is a perception affected by temperature, the room and your own adaptation — which is why the fragrance cluster insists on a blind paired comparison against a control at a matched temperature rather than a sniff test. Record both in the same log, on the same units, at the same checkpoints, and weigh each unit as well: mass loss is the one signal that says something has genuinely left the candle.
Eight readings, one white card
₹356.43 a checkpoint buys you the only honest answer to "will it stabilise" — whether it is slowing down.
Cinnamon Vanilla states 2.5% vanillin at ₹4,554.80/kg (₹54.66 of oil in a 150 g candle); Vanilla Bean states 2.5% with a 203°C flash point at ₹7,646/kg (₹91.75); Clear Glass Jars with Golden Lid 150 gram are ₹94.40 each and Black Matte Glass Jars ₹70.80; Luxury Soy Wax Chunks are ₹599/kg with a stated 8–13% load; Eco Candle Wicks Thin (C1) are ₹5.90 each in the 100-pack. Message us for bulk pricing, GST invoicing and documentation.
Shop fragrance oils Ask on WhatsApp
Continue the read
- White vanilla candle brown at three months — is the fragrance causing it?
- Darker on the side facing the light — exposure or chemistry?
- What to test before choosing fragrances for a white candle collection
- Why do some vanilla fragrance oils turn candle wax yellow?
- French vanilla vs vanilla bean fragrance oil: a full comparison
- How cure time changes the smell of different fragrance families
Editorial standards & sources
About this guide: Written by the CANDLEMAKINGSUPPLIESINDIA team. Product recommendations reflect the CSI range; the checkpoint schedule, the log sheet and the interval-comparison logic apply to any supplier's materials.Customer reviews: The six reviews at the top of this page are genuine, published Judge.me reviews left by CSI customers on their product pages. Names, star ratings, dates and products are as recorded by Judge.me; one longer review is shortened with an ellipsis and otherwise unedited.
Figures verified September 2026: Vanillin figures quoted verbatim from CSI product pages — Vanilla Bean Fragrance Oil "Vanillin Content: 2.5%" and "Flashpoint: 203°C"; Cinnamon Vanilla "Vanillin content 2.5%" and "Flashpoint 270°F (132°C)"; Dark Vanilla "Vanillin Content Present"; Gardenia "Vanillin content 0.2%"; Sandalwood "Vanillin content 0% - no discolouration, safe for white and neutral wax". Across CSI's 107 fragrance-oil pages, 62 publish a vanillin or discolouration position and 45 publish none. Zero pages state a shelf life, an expiry date, a storage instruction, a storage temperature, a discolouration rate, a discolouration endpoint or an accelerated-storage conversion; none of those is asserted anywhere on this page. Wax specifications as published: Luxury Soy Wax Chunks "Recommended fragrance load: 8% to 13% of total wax weight", "10% Recommended", add 80–90°C, pour 70–80°C, cure minimum 48 hours ideally 1–2 weeks before burn testing. Loads are wax formulation limits, not regulatory limits. Prices used in maths: Luxury Soy Wax Chunks 1 kg ₹599 (₹0.599/g); Cinnamon Vanilla 15 gm ₹106.20, 100 gm ₹483.80, 1 kg ₹4,554.80 (₹4.5548/g); Vanilla Bean 100 gm ₹790, 1 kg ₹7,646 (₹7.646/g); Clear Glass Jar with Golden Lid 150 gram pack of 10 ₹944 (₹94.40 each); Black Matte Glass Jar pack of 10 ₹708 (₹70.80 each, capacity stated in ml not grams); Eco Candle Wicks Thin (C1) 20 ₹118 and 100 ₹590 (₹5.90 each). Assumptions stated as method, not standards: 150 g finished candle at 8% = 138 g wax + 12 g oil, and at 10% = 135 g wax + 15 g oil (₹68.32 of Cinnamon Vanilla); eight checkpoints at days 0, 7, 14, 30, 60, 90, 120 and 180; twelve retained units split six shelf and six cool dark control; two-hour burn assessments at days 90 and 180; a 600-unit batch for the stock-at-risk arithmetic. Per-candle materials ₹82.66 + ₹54.66 + ₹94.40 + ₹5.90 = ₹237.62; × 600 = ₹1,42,571.76; retained set 12 × ₹237.62 = ₹2,851.44 = 2.00% of the batch; ÷ 8 readings = ₹356.43; four burn units = ₹950.48; Vanilla Bean variant per candle ₹274.71. Diwali 2026 is 8 November 2026. Prices and availability change — the linked product pages are always authoritative.