I Want to Buy Premium Fragrance Oil Samples for Car Freshener Manufacturing — How Do I Compare Their Quality Through Blind Testing?
शेयर करना
I Want to Buy Premium Fragrance Oil Samples for Car Freshener Manufacturing — How Do I Compare Their Quality Through Blind Testing?
10 ml perfume bottles ₹400 per pack of 10 · ₹40.00 each15 gm samples from ₹68.44Stocked in India · ships nationwide
CSI testing guides · fragrance oil sampling
Four correct answers out of twelve is pure guessing. Eight is a result. Everything between is why most sample comparisons prove nothing.
★★★★★
"Fresh and soothing"
Yashvi ShahVerified buyer · Jul 2025
Blue Ocean Fragrance Oil
★★★★★
"Good quality wood fragrance oil with a warm, long-lasting scent and strong performance in candles."
Payal PatelVerified buyer · Mar 2026
Woody Oud Fragrance Oil
★★★★★
"In past five years of testing and searching for the best candle fragrance I would say I found what am looking for strong fragrance in cold throw and hot throw and you can go for the minimum percentage in candle still it will give amazing smell well done and all the best"
Emma BecharaJun 2026
British Rose Fragrance Oil
★★★★★
"Hey, the fragrance is so nice. Can’t wait to use in my candles!! A must buy item!!"
S.A.Jun 2024
Jasmine Fragrance Oil
★★★★★
"Fragrances are pretty decent , quite versatile."
Tina RahaVerified buyer · Jun 2025
Lavender Fragrance Oil
★★★★★
"Quality is good with good fragrance"
Avanthi CheeliVerified buyer · Nov 2025
Citrus Lemon Fragrance Oil
★★★★★
"The fragrance was really amazing"
Akanksha SharmaVerified buyer · Jul 2026
Roasted Coffee Fragrance Oil
★★★★★
"Thankyou so much candle supplier M happy with the smell of dark vanilla M obsessed with the vanilla fragrance💕 I ll definitely shop again☺️"
SUMAN BISHTJul 2025
Dark Vanilla Oud Fragrance Oil
★★★★★
"Good product👍"
Sooraj RVerified buyer · Jul 2026
Reed & Car Diffuser Base
★★★★★
"Very good"
Reena BagalVerified buyer · Oct 2025
Sandalwood Fragrance Oil
★★★★★
"Fresh and soothing"
Yashvi ShahVerified buyer · Jul 2025
Blue Ocean Fragrance Oil
★★★★★
"Good quality wood fragrance oil with a warm, long-lasting scent and strong performance in candles."
Payal PatelVerified buyer · Mar 2026
Woody Oud Fragrance Oil
★★★★★
"In past five years of testing and searching for the best candle fragrance I would say I found what am looking for strong fragrance in cold throw and hot throw and you can go for the minimum percentage in candle still it will give amazing smell well done and all the best"
Emma BecharaJun 2026
British Rose Fragrance Oil
★★★★★
"Hey, the fragrance is so nice. Can’t wait to use in my candles!! A must buy item!!"
S.A.Jun 2024
Jasmine Fragrance Oil
★★★★★
"Fragrances are pretty decent , quite versatile."
Tina RahaVerified buyer · Jun 2025
Lavender Fragrance Oil
★★★★★
"Quality is good with good fragrance"
Avanthi CheeliVerified buyer · Nov 2025
Citrus Lemon Fragrance Oil
★★★★★
"The fragrance was really amazing"
Akanksha SharmaVerified buyer · Jul 2026
Roasted Coffee Fragrance Oil
★★★★★
"Thankyou so much candle supplier M happy with the smell of dark vanilla M obsessed with the vanilla fragrance💕 I ll definitely shop again☺️"
SUMAN BISHTJul 2025
Dark Vanilla Oud Fragrance Oil
★★★★★
"Good product👍"
Sooraj RVerified buyer · Jul 2026
Reed & Car Diffuser Base
★★★★★
"Very good"
Reena BagalVerified buyer · Oct 2025
Sandalwood Fragrance Oil
Reviews collected and published through Judge.me on candlemakingsuppliesindia.store. Wording is unedited.
✓ No oil ranked for quality — nothing is published to rank on✓ Live per-variant stock checked before every link✓ Bulk pricing & GST invoicing on WhatsApp
Car Perfume Guides · Blind Testing
I Want to Buy Premium Fragrance Oil Samples for Car Freshener Manufacturing — How Do I Compare Their Quality Through Blind Testing?
Blind testing is worth doing and it does not measure quality. What it measures is whether two samples can be told apart — and that has an exact arithmetic answer, because in a properly built trial a person who cannot smell any difference will still be right a third of the time by guessing. Twelve trials, four right by chance, eight right before you are allowed to call it a result. Most "we tested both and preferred the expensive one" comparisons sit somewhere around five and prove nothing at all.
8 of 12
correct identifications a twelve-trial triangle test needs before chance is ruled out at the 5% level — guessing alone returns four · the whole head-to-head costs from ₹886.66: two 15 gm samples, one pack of ten coded bottles, one 100 g base pack · CSI live pricing, October 2026
Quick answers — read this first
Can blind testing tell me which oil is higher quality? No. "Quality" has no published referent in this range or anywhere else I could cite — no CSI listing publishes an odour description, a throw figure, a longevity figure or a purity specification, and I have not smelled any of these oils. What a blind test answers is narrower and far more useful: can anyone tell these two apart? If the answer is no, every other argument about them is moot.
What is the test? A triangle trial. Three coded bottles, two filled from one oil and one from the other, presented in random order. The assessor says which is the odd one. Someone who perceives nothing is still right one time in three, which is why single trials are worthless and why the threshold matters.
How many trials, and how many correct? Twelve trials need eight correct to clear the 5% level — the exact one-sided binomial probability of getting eight or more by guessing is 0.0188. Seven of twelve gives 0.0664 and fails. Nine trials need six, fifteen need nine, eighteen need ten.
What does it cost? Six coded bottles at a 2.0 g charge use 12 g of oil in total — 6 g of each — so a single 15 gm sample of each candidate runs the whole twelve-trial test with 9 g spare. A pack of ten 10 ml bottles is ₹400 at ₹40.00 each. With a ₹325 base pack, the cheapest complete head-to-head is ₹886.66.
The short answer
Test discrimination, not quality: a triangle trial asks "which is the odd one", which has a right answer. "Which is better" does not, and no supplier page in this range publishes anything you could rank on.
Know the guess rate before you start: one in three. Four of twelve is noise, eight of twelve is evidence, and the band between them is where most informal comparisons land.
Twelve assessors beat one assessor twelve times: the arithmetic is identical, the fatigue and adaptation are not. Three assessors doing four trials each is the practical compromise.
One 15 gm sample of each oil is enough: six bottles at 2.0 g is 12 g of oil for a complete test. Sampling costs run ₹68.44 to ₹1,203 depending on the listing's lowest rung.
The useful thing about this chart is the right-hand column of percentages. At six trials you need more than four in five correct, which almost nothing clears; at twenty-four you need barely more than half. That is the real cost of a small test — not that it is less accurate, but that it demands a near-perfect score before it will tell you anything, so a genuinely perceptible difference can easily fail to register.
How do I compare premium fragrance oil samples through blind testing?
Run a triangle test and decide the pass mark before you start. Take the two candidates, make up six coded bottles — three of each oil, every one of them 2.0 g of oil in 6.0 g of base for an 8.0 g fill, which is the published 25% working ratio — and rest them the published 24 to 48 hours. Each trial presents three bottles, two from one oil and one from the other, in random order, and the assessor names the odd one out. Because a person who perceives no difference still guesses correctly one time in three, the score has to clear a threshold: across twelve trials you need eight correct, which has a one-sided binomial probability of 0.0188 under pure guessing. Seven of twelve comes out at 0.0664 and does not clear it. Twelve trials consume 12 g of oil in total, 6 g of each candidate, so one 15 gm sample of each is enough with 9 g to spare — the cheapest complete head-to-head in this catalogue is ₹68.44 + ₹93.22 of oil, ₹400 for a pack of ten 10 ml bottles and ₹325 for a base pack, a total of ₹886.66. What the result tells you is whether the two are distinguishable, and nothing more. It does not rank them for quality: no page in this range publishes a purity, odour, throw or longevity specification, nobody at CSI has panelled the range, and I have not smelled any of it. If the pair is distinguishable you can then ask a separate preference question of a larger group; if it is not, the dearer sample has nothing to sell and the decision is already made.
One line: decide the pass mark first — eight of twelve — because a test whose threshold you set after seeing the answers is not a test.
The sampling bench: 10 ml perfume bottles ₹400 per pack of 10 (₹40.00 each, 1,989 packs, Gold or Silver — order one colour only), Attar Bottles 3 ml ₹180 per pack of 10 at ₹18.00 each for retained references, and 15 gm oil samples from ₹68.44.
Blind means the assessor does not know. It does not mean the question was worth asking.
Most sample comparisons fail at the question, not at the method. A maker lines up four oils, sniffs each one, writes "nice / too sharp / best" beside them, and treats that as a test. It is not blind — they know which is the ₹1,203 sample and which is the ₹68.44 one — and the question has no right answer, so there is nothing to be correct or incorrect about. Blinding fixes the first problem. Choosing an answerable question fixes the second, and it has to come first.
There are four distinct things people mean by "comparing samples", and only two of them are testable on a kitchen table. Separating them is the single most valuable thing on this page, because it stops you running an elaborate protocol on a question that cannot produce a result.
Four questions makers call "blind testing" · only two have answers
What each question needs, and what it can conclude
Question
Is there a right answer?
Method
How many people
What you can conclude
Can these two be told apart?
Yes — the odd bottle is a fact
Triangle trial, threshold set in advance
12 trials, ideally 12 people
A statistical statement about perceptibility
Which do people prefer?
No right answer, but a measurable majority
Paired preference, coded, forced choice
Many more — 30+ before a split means much
A preference share, with wide uncertainty at small n
Which is higher quality?
No — no published referent exists
None available
—
Nothing. Do not run it.
Which is closer to a named perfume?
No — needs the original and a trained panel
Not available on a workbench
—
Nothing, and the claim itself should not be made
Does this lot match the last one?
Yes, if you retained a reference
Triangle trial against a retained dated sample
12 trials
Whether two lots are perceptibly different
Does the cheap build pass for the dear one?
Yes
Triangle trial between the two builds
12 trials
Whether the premium is perceptible at your charge
About the third and fourth rows: CSI publishes no purity, odour, composition or performance specification for any fragrance oil — the only page in the range that publishes an ingredient list at all is the CSI Luxury Perfume Fixative Complex, and it contradicts its own dose guidance. So "quality" has nothing to be measured against, and this page does not rank any two oils. On the fourth row, the brand-referencing listings in this range name an olfactory direction only: no match, clone, dupe or equivalence to any perfume is claimed or implied, nothing here describes how any original smells, and comparisons to a branded original are not a test anyone can run on a workbench. If your own marketing will reference a brand, that is a question for a qualified adviser.
The last two rows are where the money is. A triangle trial between your dear build and your cheap build is the test that actually changes a purchase decision, and a triangle trial against a retained dated reference is the only incoming-quality check available to you, because nothing in this range publishes a lot code, a batch number or a variation tolerance. Three millilitre attar bottles at ₹18.00 each make that retention practical; the full case for keeping them is in the retained-sample guide.
The triangle trial and its arithmetic
A triangle trial is three samples, two of them identical. The assessor's job is to identify the one that differs. The reason this design is used rather than a simple "are these two the same" is that it has a known and modest guess rate: with three samples and one odd one, a person with no perception at all will still be correct one time in three. That fixed guess rate is what makes the result computable.
Everything then follows from the binomial distribution. Run n trials, count the correct answers, and ask how likely that many or more would be by guessing alone at one in three. If the answer is under 5%, you conclude the pair is perceptibly different. The thresholds below are the exact ones; they are not rules of thumb and they do not scale linearly, which is why picking a trial count matters.
Exact one-sided binomial thresholds at a one-in-three guess rate
Correct answers needed to call a difference real, by trial count
Trials
Expected by guessing
Correct needed (5% level)
Exact probability by chance
As a share of trials
Oil consumed at 2.0 g a bottle
6
2.0
5
0.0178
83.3% — brutally high
12 g (6 bottles)
9
3.0
6
0.0424
66.7%
12 g (6 bottles)
12
4.0
8
0.0188
66.7%
12 g (6 bottles)
15
5.0
9
0.0308
60.0%
24 g (12 bottles)
18
6.0
10
0.0433
55.6%
24 g (12 bottles)
24
8.0
13
0.0284
54.2%
36 g (18 bottles)
30
10.0
15
0.0435
50.0%
48 g (24 bottles)
Worked out above: each threshold is the smallest number of correct answers whose exact one-sided binomial tail probability at p = 1/3 falls at or below 0.05. At twelve trials, P(8 or more correct) = 0.0188 and P(7 or more) = 0.0664, which is why eight and not seven is the pass mark. Expected-by-guessing is n ÷ 3. Oil consumption assumes six coded bottles can serve twelve trials when they are re-coded and re-randomised between trials — twelve trials at 2.0 g a bottle is 12 g, 6 g of each candidate. Above twelve trials, assume a fresh set each block so the bottles are not opened repeatedly: that is where consumption rises. These are properties of the test design, not measurements of any product.
Two design warnings that the arithmetic hides. First, the binomial assumes independent trials, and one person doing twelve trials in a row is not twelve independent trials: the nose adapts, fatigue sets in, and the later answers are correlated with the earlier ones. Twelve different people doing one trial each is the clean version; three people doing four trials each with a break between is the practical compromise. Second, a fail is not a proof of sameness. Falling short of eight means you have not demonstrated a difference, not that no difference exists — a small test is simply not powerful enough to say so, and the only honest write-up is "not demonstrated at twelve trials".
The mistake: deciding the threshold after you see the score. Six of twelve feels encouraging — it is half, after all — and it is almost exactly what guessing plus a little luck produces; the chance of six or more by pure guessing is 0.1777, better than one in six. If you set the bar after the fact, you will set it at whatever your score reached, and you will buy the expensive oil on the strength of a coin toss. Write the pass mark on the log sheet before the first trial, in ink, and hand it to somebody else. The same discipline applies to stopping: do not run extra trials because you are close, and do not stop early because you are ahead.
Buying the samples: what each ticket costs
A triangle test needs very little oil — 6 g of each candidate — so the cost of sampling is almost entirely the cost of the smallest rung that exists on each listing. That varies enormously across this catalogue, from ₹68.44 to ₹1,203, because some listings carry a 15 gm rung and others start at 50 g or 100 g. The practical consequence is that the price of finding out about a direction can differ by a factor of seventeen between two oils you might reasonably be comparing.
Sampling tickets · the smallest buyable rung on each listing, October 2026
What it costs to get 6 g of a candidate onto the bench
Worked out above: a twelve-trial test needs 6 g of each candidate, so spare = ticket quantity − 6 g. Tickets span ₹68.44 to ₹1,203.00, a factor of 17.6, and the per-gram column shows how little of that is about the oil's production price — Forever Red's 50 g ticket is ₹6.98 a gram, almost identical to Palo Santo's 15 gm at ₹7.08, but costs five times as much because no smaller rung exists. Four of these listings have no rung below 50 g and two none below 100 g, so the ticket is a function of the ladder's floor, not of the oil's value. One per-variant warning: White Royal Oud's 15 gm is unavailable behind a product page showing 50 packs, so always check availability rung by rung rather than trusting the product-level figure.
Three complete twelve-trial head-to-heads, fully costed
Everything the test consumes, including the bottles and the carrier
Head-to-head
Oil tickets
Bottles
Base
Total
Cost per trial
Eucalyptus Spearmint vs Woody Oud
₹68.44 + ₹93.22 = ₹161.66
₹400.00 (pack of 10)
₹325.00 (100 g)
₹886.66
₹73.89
British Rose vs Roasted Coffee
₹94.40 + ₹93.22 = ₹187.62
₹400.00
₹325.00
₹912.62
₹76.05
Santal Noir vs Club De Nuit Intense inspired
₹166.00 + ₹1,203.00 = ₹1,369.00
₹400.00
₹325.00
₹2,094.00
₹174.50
Your cheap build vs your dear build
Oil you already own
₹400.00
₹325.00
₹725.00
₹60.42
This lot vs a retained reference
Ticket for the new lot only
₹400.00
₹325.00
From ₹793.44
From ₹66.12
Worked out above: every row uses one pack of ten 10 ml perfume bottles at ₹400.00 (₹40.00 each; six are used, four spare) and one Pure Performance 100 g base pack at ₹325.00 (₹3.25 a gram; 6 bottles × 6.0 g = 36 g used, ₹117.00 consumed, the rest available for the next test). Totals are the whole purchase, not the consumption, because packs are indivisible. Cost per trial = total ÷ 12. The last row assumes you retained a dated sample of the earlier lot — 3 ml attar bottles are ₹180 per pack of 10, ₹18.00 each, which makes retention cheaper than any other line on this table.
Running twelve trials without leaking the answer
A blind test leaks in four places: the vessel, the fill, the order and the operator. Every one of them is cheap to control and expensive to ignore, because a leak does not produce a wrong answer, it produces a confident one. The sequence below closes all four, and it is written for a maker working alone at a table with a scale.
1
Buy one colour of bottle, not an assortmentThe 10 ml perfume bottle is listed in Gold and Silver at ₹400.00 for ten, and a set that mixes the two hands the assessor a second variable to sort on. Order one colour. Use the same pack, the same cap and the same fill height for all six, and set aside the four spares for re-runs rather than mixing in bottles from an older order.
2
Fill all six by weight in one sittingTare each bottle, add 6.0 g of base, then 2.0 g of oil — an 8.0 g fill at the published 25% working ratio, which leaves headroom in a 10 ml bottle. Three bottles of candidate A, three of candidate B. Write every weight down. A fill that differs by half a gram is visible through the glass, and a visible difference is an answer the assessor did not have to smell.
3
Code with random three-digit numbers, and keep the key away from the tableGenerate six random codes — not 1 to 6, not A1/A2/A3 and B1/B2/B3 — write them on identical labels, and record the key on a sheet that goes in another room. Codes that cluster, or handwriting that changes between the two sets, are the commonest leak of all and the easiest to avoid.
4
Rest everything the published 24 to 48 hours before the first trialEvery base page in the range publishes a 24 to 48 hour rest. Rest all six together, in the same place, for the same duration, and note the times. A freshly mixed bottle beside a rested one is a different formulation, not a different oil, and the assessor will correctly spot a difference that tells you nothing about the two candidates.
5
Randomise which oil is the odd one, trial by trialAcross twelve trials, candidate A should be the odd one in roughly half and candidate B in the other half, decided by a coin rather than by you. If A is always the odd one, an assessor who notices any pattern — including an unconscious one in how you set the bottles down — starts scoring above chance for the wrong reason. Write the intended answer for each trial on the key sheet before the trial, never after.
6
Have somebody else present the triplets, or present them from behind a screenYou know the codes, so you cannot run the trials. If nobody else is available, lay the triplet out, leave the room, and let the assessor record their own answer on a slip they fold over. Give them water and a minute between trials, and stop at four trials per person if you are using fewer than twelve assessors.
7
Count, compare to the pass mark you wrote down, and write the verdict in fullEight or more correct out of twelve: the pair is perceptibly different, p = 0.0188. Seven or fewer: not demonstrated — which is a different statement from "the same". Record the score, the number of assessors, the trials per assessor, the fill weights, the rest period and the date. That paragraph is the output, and it is what makes the purchase decision defensible in three months when a price has moved.
Add a positive control if the result surprises you
If a pair you expected to be obviously different fails to reach eight, run one extra block with a deliberately different pair — say Eucalyptus Spearmint against Roasted Coffee, two listings nobody would confuse. If your assessors cannot score eight on that either, the problem is the protocol, the room, the rest period or the assessors, not the oils. At ₹161.66 of oil it is the cheapest diagnostic in this entire method, and it tells you whether your bench is capable of producing a result at all.
The vessel and three sampling tickets
One piece of packaging and three oils, priced at every rung. The three oils are chosen to span the ladder shapes you will meet — one flat, one descending, one with its cheapest gram at the bottom rung — because that shape is what decides whether the price you paid to sample predicts the price you will pay to produce. Nothing here says how any of them smells.
1
The coded presentation vessel
10 ml perfume bottle (Pack of 10)
1,989 packs. ₹400.00 for ten = ₹40.00 a bottle, listed in Gold and Silver at the same price, and the right vessel for a triangle test for one reason: ten identical bottles in a single pack means six for the test and four spares with no visual variation between them. Order one finish only. An 8.0 g fill — 6.0 g of base plus 2.0 g of oil at the published 25% working ratio — leaves comfortable headroom in a 10 ml bottle. For retaining a dated reference of each lot after the test, the Attar Bottles are cheaper still at ₹18.00, ₹30.00 and ₹36.00 each for 3, 6 and 12 ml in packs of ten — note the handle reads 2 ml against a 3 ml title.
313 packs. Its 15 gm rung at ₹68.44 is the cheapest entry price anywhere in the catalogue, which makes it the natural positive control: the pair you use to prove your bench can produce a result before you spend ₹1,203 finding out that it cannot. Its ladder descends genuinely, from ₹4.5627 a gram at 15 gm to ₹3.0680 at the kilo, a 32.8% fall, with the 500 gm rung saving ₹118 against five 100 gm packs. One per-variant note worth checking before a production order: its 100 gm rung has run thin behind the product-level total before, so confirm the rung you intend to buy rather than the product.
The flat ladder · your sample gram is your production gram
Woody Oud Fragrance Oil
402 packs — which, converted from pack counts, is roughly 159,650 g on the shelf, so depth is not a concern here. Its value in a sampling exercise is the ladder: ₹4.5312 a gram at 100 gm, 500 gm and the kilo, identical to four decimal places. That means a decision you take on a 15 gm sample carries into production at a known price, with no rung to discover and no inversion to trip over. The 15 gm ticket at ₹93.22 is ₹6.2147 a gram, a 37.2% premium over the flat production price — the full price of finding out, and one of the smaller premiums in the range. Payal Patel's review calls it a "Good quality wood fragrance oil with a warm, long-lasting scent and strong performance in candles" — a customer's experience in candles, not a published specification and not a car-perfume result.
Size
Price
Per gram
Premium over the flat price
15 gm
₹93.22
₹6.2147
+37.2%
50 gm
₹271.40
₹5.4280
+19.8%
100 gm
₹453.12
₹4.5312
flat
500 gm
₹2,265.60
₹4.5312
flat
1 kg
₹4,531.20
₹4.5312
flat
₹4.5312 a gram from 100 gm to the kilo — nothing to negotiateBuy Woody Oud
4
No rung below 50 g · and its 50 g is its cheapest gram but one
Forever Red Bath and Body Works Inspired Fragrance Oil
290 packs, and it is on this page as the shape you need to recognise: a brand-referencing listing with no rung below 50 g, so the sampling ticket is ₹349.00 even though the oil's production price is ordinary. Two inversions sit inside its ladder. Its 50 g at ₹6.9800 a gram undercuts its own 100 g at ₹6.9900, and its 500 g at ₹3,499.00 is ₹4 dearer than five 100 g packs — while the kilo at ₹6,499.00 is a genuine ₹491 discount against ten 100 g packs. So the sensible purchase pattern is 50 g to sample, 100 g packs to pilot, skip the 500 g, and go to the kilo only at volume. The title names an olfactory direction and nothing else: no endorsement, affiliation, match or equivalence is claimed, and nothing here describes how the original product smells.
A blind test does not tell you which oil is better. It tells you whether "better" is a conversation worth having.
Nobody at CSI has panelled this range, no listing publishes an odour or purity specification, and this page has not smelled a single oil in it. What we can hand you is a test with a known guess rate, an exact pass mark, and a bill of materials under ₹900. If your assessors cannot clear eight of twelve, the premium you were about to pay for has no customer on the other end of it.
Sampling four directions rather than two? Four 15 gm tickets from ₹68.44 to ₹166.00 plus one ₹400 pack of bottles and one ₹325 base pack runs six pairwise triangle tests out of the same purchase. Retain a 3 ml dated reference of each lot at ₹18.00. For recurring volumes, bulk pricing and GST invoicing are handled on WhatsApp.
One last thing to check before a sample round, because it quietly decides whether the test was worth running. The price you pay for a 15 gm sample is not the price you will pay in production, and the gap between them varies hugely by listing. On a flat ladder the gap is a fixed premium you pay once. On a descending ladder the production gram can be less than half the sample gram. And on a few listings the ladder climbs, so the sample is the cheapest gram you will ever buy — which is a genuinely useful thing to know before you build a costing on a sample price.
Sample gram against production gram · the premium you pay to find out
Whether your sample price predicts your production price
Listing
Sample rung, per gram
Cheapest production gram
Premium on the sample
Ladder shape
Club De Nuit Intense inspired
₹24.0600 (50 g)
₹24.0560 (1 kg)
+0.0%
Flat to the paisa — the sample price is the production price
Gilli Mitti (Patchouli)
₹9.3333 (15 gm)
₹8.9600 (50 gm up)
+4.2%
Flat from 50 gm
Forever Red BBW Inspired
₹6.9800 (50 g)
₹6.4990 (1 kg)
+7.4%
Shallow descent, with a 500 g inversion
Palo Santo
₹7.0800 (15 gm)
₹6.0180 (50 gm up)
+17.6%
Flat from 50 gm
Santal Noir
₹11.0667 (15 g)
₹9.1200 (1 kg)
+21.3%
Descends, but its 500 g is dearer than five 100 g
Saffron Cedarwood
₹6.9227 (15 gm)
₹5.1684 (500 gm)
+33.9%
Descends to 500 gm, then the kilo jumps back up
Woody Oud
₹6.2147 (15 gm)
₹4.5312 (100 gm up)
+37.2%
Flat from 100 gm
Eucalyptus Spearmint
₹4.5627 (15 gm)
₹3.0680 (1 kg)
+48.7%
Genuine descent at every rung
Verbena
₹7.0800 (15 gm)
₹4.7200 (1 kg)
+50.0%
Genuine descent, kilo saves ₹1,180 on ten 100 gm
Roasted Coffee
₹6.2147 (15 gm)
₹3.7380 (1 kg)
+66.3%
Genuine descent
British Rose
₹6.2933 (15 gm)
₹3.0680 (1 kg)
+105.1%
Falls by half — but 7,085 of its packs sit in one rung
Jasmine Coconut Spa Ceylon
₹8.8667 (15 g)
₹7.9800 (50 g)
+11.1%
Climbs above 50 g — its 50 g is its cheapest gram even for production
Worked out above: premium = sample per-gram ÷ cheapest production per-gram − 1. The span runs from 0.0% to +105.1% on these twelve listings. Two shapes to watch. On Jasmine Coconut Spa Ceylon the ladder ascends above 50 g (₹7.9800 → ₹7.9900 → ₹7.9980 → ₹7.9990), so buying bigger costs more per gram and its 50 g is the cheapest gram at any volume. On Saffron Cedarwood and Sage and Cedarwood the 1 kg pack at ₹6,867.60 is ₹1,699.20 dearer than two 500 gm packs, so the cheapest production gram sits at 500 gm, not at the top. A fuller treatment of sampling tickets by product format is in the woody sampling guide.
The practical rule that falls out of this is short. Before you buy a sample, look at the listing's ladder and write down the production gram beside the sample gram. If the two are close, your blind test is settling a decision whose economics you already know. If the production gram is half the sample gram, the test is even better value than it looks. And if the ladder climbs, the sample price is the best price you will ever see on that listing and your costing should use it rather than a hoped-for volume discount.
Set the pass mark before you open the bottles. After that, the test is just counting.
— CandleMakingSuppliesIndia
Why trust this guide
CANDLEMAKINGSUPPLIESINDIA supplies raw materials, not finished car perfumes. This page does not tell you which sample is better, and it will not. No CSI listing publishes an odour description, a purity specification, a throw figure or a longevity figure; nobody at CSI has run a sensory panel on the range; and I have not smelled any of these oils. The test design and the arithmetic are what we can honestly give you, and they are enough to turn a sample round into a decision.
Every price, rung and stock figure comes from live CSI data in October 2026, read per variant rather than per product, and links to the product page where current pricing always takes precedence. The statistical thresholds are exact one-sided binomial tail probabilities at a guess rate of one in three — at twelve trials, P(8 or more) = 0.0188 and P(7 or more) = 0.0664 — and they are properties of the test, not of any product. Stated specifications are quoted only where a page publishes them: the base pages' "Blend with 10-30% fragrance oil", 25–30% working ratio and 24–48 hour rest. The Room & Linen Spray Base publishes a separate and lower 8–15% / 10% recommended / 15% maximum / 12–24 hour rest and is not used here.
Two catalogue notes for anyone sampling widely. Availability must be checked per variant — there are active listings whose individual rungs are unbuyable, including White Royal Oud's 15 gm behind a product showing 50 packs. And several listings report inventory in a repeating 111 / 1000 / 1110 pattern that is plainly placeholder data; treat any unconfirmed quantity as unconfirmed. For bulk pricing, GST invoicing, MSDS or IFRA documentation, message us on WhatsApp at +91 7397976926; documentation requests are answered with documents, never with a summary.
Frequently asked questions
How many correct answers does a blind fragrance comparison need?
In a triangle test the guess rate is one in three, so the pass mark depends on the trial count: 5 of 6, 6 of 9, 8 of 12, 9 of 15, 10 of 18, 13 of 24, each at the 5% level. Twelve trials with eight correct has an exact chance probability of 0.0188; seven correct comes out at 0.0664 and fails. Write the pass mark down before the first trial — a threshold chosen after the score is not a test, and six of twelve, which feels like a result, happens by guessing alone almost one time in six.
How much fragrance oil do I need for a blind test?
Very little. Six coded bottles at a 2.0 g charge — three of each candidate — is 12 g in total, 6 g of each, so a single 15 gm sample covers a complete twelve-trial test with 9 g spare. The bottles and the base cost more than the oil: a pack of ten 10 ml bottles is ₹400.00 and a 100 g base pack is ₹325.00, against ₹161.66 of oil for the cheapest head-to-head. Complete test from ₹886.66.
Can a blind test tell me which fragrance oil is better quality?
No. "Quality" has no measurable referent here — no CSI page publishes a purity, composition, odour or performance specification for any fragrance oil, so there is nothing to score against. What a triangle test answers is whether two samples are perceptibly different, which is a factual question with a right answer. If they are, you can then ask a separate preference question of a much larger group. If they are not, the price difference between them has nothing to sell.
Is one person doing twelve trials the same as twelve people doing one?
Arithmetically yes, practically no. The binomial threshold assumes independent trials, and a single nose adapts within minutes, so the later answers in a long session are correlated with the earlier ones and the test's real error rate is worse than 0.0188. Twelve assessors doing one trial each is the clean design; three doing four each with a break between is the usual compromise. If you must use one assessor, spread the trials across days and note the gaps in the log.
What should I present the samples in?
Ten identical bottles from one pack. The 10 ml perfume bottle is ₹400.00 for ten, ₹40.00 each, and it comes in Gold and Silver — order one colour, because a mixed set gives the assessor a second thing to sort on. Fill all of them by weight in one sitting to the same 8.0 g, code them with random three-digit numbers rather than sequential labels, and rest them together for the published 24 to 48 hours. For keeping a dated reference afterwards, 3 ml attar bottles are ₹18.00 each.
My test failed to reach the pass mark. Does that mean the two oils are the same?
No — it means a difference was not demonstrated at that number of trials, which is a weaker statement. A twelve-trial test needs two thirds correct to clear the bar, so a real but subtle difference can easily fail. Two things to do before concluding anything: run a positive control with a pair nobody would confuse, to prove your bench and assessors can score eight at all; and if the control passes but your pair does not, accept that the difference is too small for your customers to notice either — which is itself a complete answer to a purchasing question.
Eight of twelve, or it did not happen
Six coded bottles, 12 g of oil, one pass mark written down in advance — under ₹900 for an answer you can defend.
10 ml perfume bottles ₹400 per pack of 10 = ₹40.00 each, Gold or Silver, 1,989 packs. Attar Bottles 3 / 6 / 12 ml ₹180 / ₹300 / ₹360 per pack of 10 = ₹18.00 / ₹30.00 / ₹36.00 each. Pure Performance Base 100 g ₹325 (₹3.25/g) or 500 ML ₹1,099. Sampling tickets: Eucalyptus Spearmint 15 gm ₹68.44, Woody Oud ₹93.22, Roasted Coffee ₹93.22, British Rose ₹94.40, Saffron Cedarwood ₹103.84, Palo Santo ₹106.20, Santal Noir ₹166.00, Forever Red 50 g ₹349.00, Royal Oud 100 g ₹699.00, Amber Oud Royal 100 g ₹1,299.00, Club De Nuit Intense inspired 50 g ₹1,203.00. Message us for bulk pricing, GST invoicing and documentation.
About this guide: Written by the CANDLEMAKINGSUPPLIESINDIA team. Product recommendations reflect the CSI range; the triangle-test design, the binomial thresholds and the sample-to-production comparison apply to any supplier's materials.
Customer reviews: The 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; wording is unedited. They describe customers' own experience, largely in candles, and are neither specifications nor blind-test results.
Figures verified October 2026: Statistical thresholds are exact one-sided binomial tail probabilities at p = 1/3: 5 of 6 (0.0178), 6 of 9 (0.0424), 8 of 12 (0.0188), 9 of 15 (0.0308), 10 of 18 (0.0433), 13 of 24 (0.0284), 15 of 30 (0.0435); 7 of 12 is 0.0664 and 6 of 12 is 0.1777; expected by guessing is n ÷ 3. Test consumption: six 10 ml bottles at an 8.0 g fill of 6.0 g base plus 2.0 g oil = 12 g of oil and 36 g of base. Glass: 10 ml perfume bottle ₹400.00 per pack of 10 = ₹40.00 each, Gold and Silver, 1,989 packs; Attar Bottles 3 / 6 / 12 ml ₹180 / ₹300 / ₹360 per pack of 10 (the handle reads 2 ml against a 3 ml title), 298 packs. Base: Pure Performance 100 g ₹325 = ₹3.25/g, 500 ML ₹1,099 = ₹2.198/ml, 119 packs. Sampling tickets and ladders: Eucalyptus Spearmint 15 gm ₹68.44 to 1 kg ₹3,068.00 (313 packs); Woody Oud ₹93.22 to ₹4,531.20, flat at ₹4.5312/g from 100 gm (402); Roasted Coffee ₹93.22 to ₹3,738.00 (435); British Rose ₹94.40 to ₹3,068.00 (7,272, of which 7,085 packs sit in the 500 gm rung); Saffron Cedarwood ₹103.84 to ₹6,867.60 (417); Palo Santo ₹106.20 to ₹6,018.00, flat ₹6.018/g from 50 gm (48); Santal Noir ₹166.00 to ₹9,120.00 (1,537); Forever Red 50 g ₹349.00 / 100 g ₹699.00 / 500 g ₹3,499.00 / 1 kg ₹6,499.00 (290); Sage and Cedarwood 50 gm ₹354.00 to ₹6,867.60 with no rung below 50 gm (184); Royal Oud 100 g ₹699.00 to ₹6,400.00 with no rung below 100 g (23); Amber Oud Royal 100 g ₹1,299.00 to ₹11,640.00 (29); Club De Nuit Intense inspired 50 g ₹1,203.00 / 100 g ₹2,406.00 / 1 kg ₹24,056.00 (1,198); Gilli Mitti 15 gm ₹140.00 to ₹8,960.00, flat ₹8.96/g from 50 gm (128); Verbena ₹106.20 to ₹4,720.00 (475); Jasmine Coconut Spa Ceylon 15 g ₹133.00 to ₹7,999.00, ascending above 50 g (1,540). Sample-to-production premiums span 0.0% to +105.1%. Stated specifications as published: reed and car bases "Blend with 10-30% fragrance oil", 25–30% working ratio, 24–48 hour rest; the spray base's separate 8–15% / 10% / 15% / 12–24 hours, not used here. No CSI oil, base or packaging page publishes an odour description, purity specification, throw, longevity or heat-stability figure, no oil is ranked for quality here, and no comparison to any branded original is claimed or implied. Brand names appear only as exact listed CSI product titles naming an olfactory direction; trademark questions belong with a qualified adviser. Availability is checked per variant — White Royal Oud's 15 gm is unavailable behind a live product — and several listings report inventory in a repeating 111 / 1000 / 1110 pattern that is placeholder data. Prices and availability change; the linked product pages are always authoritative.