I Made Three Fragrance Variants Using the Same Formula, but Their Product Lives Differ by Several Weeks — Should I Change the Reed Count for Each?
Aktie
What does that look like with numbers? Worked example at 100 g and six reeds: lemon 28 days, lavender 40, vanilla 62, 34 days apart. With a 40–50-day band: lemon at four reeds and a 110 g fill (42 days), lavender unchanged, vanilla at eight reeds (50 days). Illustrative, not CSI test data.
Is there a catch? Reeds move throw as well as life. Fewer reeds on the fast variant can make it quieter; more on the slow one can make it louder. Blind 3 m sniffs run alongside the weighing, and a count only passes if the scent still meets your line.
What does it cost? If you use unsold bottles from the batch, only reeds: ₹1,436.00 for the ladder. Mixing fresh adds three packs of 120 ML bottles, ₹4,622.00 in all.
- Straight answer: measure, then set reeds per variant
- Your three variants, restated
- Life = fill ÷ grams per day, done properly
- The cause table: why lives differ
- The reed ladder: 4, 6 and 8 per variant
- The worked table: choosing each variant's count
- Life is not the only thing reeds change
- What CSI sells for this, and the production run
- What to buy, and what to skip
- Frequently asked questions
Your three variants, restated
Your case, restated. You made three fragrance variants using the same formula, and their product lives differ by several weeks. The title gives no oils, fill or reed count, so this page works a labelled example: Citrus Lemon, Lavender and Dark Vanilla, each 20 g oil + 80 g base (20% by weight), 100 g in a 120 ml Gold bottle, six fibre reeds, one room. Measured over days 7–21 (illustrative), they lose 3.6, 2.5 and 1.6 g a day, so lives of 28, 40 and 62 days: 34 days, about 4.9 weeks, between the first and last. Put your own rates in; the arithmetic is the same.
Why fix it at all? A range where one scent is gone in four weeks and another lasts nine makes any life claim on the box wrong for two of the three. It also makes the fast variant look poor value and the slow one look faint. The aim is not identical lives, which is unrealistic, but lives inside a band you can honestly describe.
Why a lemon empties faster than a heavy scent is the lemon-versus-sandalwood consumption post; this page takes the difference as given and fixes it with reeds and fill. Whether the three should also differ in load is the one-standard-formula-or-per-oil-loads post.
Life = fill ÷ grams per day, done properly
Use the right window. Most diffusers lose weight fastest in the first days, as reeds saturate and the lightest materials leave, then settle. A life projected from week one is too short. Average grams per day from about day 7 to day 21, and treat the projection as an estimate that the confirmation run checks. Projecting life from weight loss without waiting for empty bottles is set out in the live guide on estimating life from weight loss.
Use three bottles, not one. Take the median of three per variant, and note the spread. If two bottles of one variant differ by more than the gap you are trying to close, fix filling and reed handling before touching counts.
Use one room. Lives in a warm, airy room are shorter than in a cool, still one, and Indian summers shorten them further; the live guide on summer-versus-winter lives covers seasonal formulas. For this decision, all that matters is that the three variants share a room, so the comparison is fair.
| Variant | g/day at 6 reeds | Life at 6 reeds | Rate needed for 50 days | Rate needed for 40 days |
|---|---|---|---|---|
| Lemon | 3.6 | 28 d | 2.0 g/day | 2.5 g/day |
| Lavender | 2.5 | 40 d | 2.0 g/day | 2.5 g/day |
| Vanilla | 1.6 | 62 d | 2.0 g/day | 2.5 g/day |
So for a 40–50-day band at 100 g, each variant needs to lose between 2.0 and 2.5 g a day. Lemon must slow down, vanilla must speed up, lavender is already on the edge. Reeds are the obvious lever because the reed page itself says more reeds diffuse faster and fewer last longer; how much, for your oils, only the ladder tells you.
The cause table: why lives differ
| Cause | What you'd see | Discriminating test | Fix |
|---|---|---|---|
| Oil · volatility (composition tendency) | Citrus fastest, vanilla slowest, steady across bottles | Same rate in all three bottles; blank shows the carrier's share | Per-variant reeds or fill (this page) |
| Oil · viscosity and colour | Dark, thick variant climbs slowly, reeds darken | Day 0 climb watch; fresh-reed swap mid-test | Reeds help only if the climb is fine |
| Oil · strength per gram | Slow variant also faint at 3 m | 3 m score alongside the rate | More reeds may fix both; if not, load or oil |
| Oil · lot or supplier | Lives shift between batches of one variant | Retained sample vs new lot, same test | Lot check before you re-set reeds |
| Load | A variant mixed at a different % by mistake | Batch records; weigh a retained bottle | Re-mix; same load |
| Base | Different base lot for one variant | Batch records | One lot per test |
| Reeds | Different counts or packs between variants | Count reeds in finished bottles | One pack; counts per spec |
| Bottle and fill | Fills vary by several grams | Weigh nine unsold bottles of each | Fill by weight |
| Room and climate | Variants tested in different rooms or months | Same room, same weeks | Retest side by side |
| Measurement · nose fatigue, bottle sniffing, first-week rate | Life judged by "when it stopped smelling", or by week one | Use weight, days 7–21; blind sniffs at 3 m | Fix the method first |
Rule out the process rows first. If unsold bottles of one variant weigh differently, or two batches give different rates, no reed count will hold steady.
The reed ladder: 4, 6 and 8 per variant
| Variant | 4 reeds | 6 reeds | 8 reeds | Plus |
|---|---|---|---|---|
| Lemon | 3 bottles | 3 bottles | 3 bottles | one sealed retained bottle |
| Lavender | 3 bottles | 3 bottles | 3 bottles | one sealed retained bottle |
| Vanilla | 3 bottles | 3 bottles | 3 bottles | one sealed retained bottle |
That is 27 bottles and 162 reeds. Use unsold filled bottles from the batch if you have nine of each; otherwise mix fresh: 180 g of each oil and 2,160 g of base, rested 24–48 hours. Six reeds is the control because it is your current formula.
The worked table: choosing each variant's count
| Variant | 4 reeds | 6 reeds | 8 reeds | Chosen | Why |
|---|---|---|---|---|---|
| Lemon | 2.6 g/d · 38 d | 3.6 g/d · 28 d | 4.4 g/d · 23 d | 4 reeds + 110 g fill · 42 d | Four reeds alone falls just short; a larger fill closes it |
| Lavender | 1.9 g/d · 53 d | 2.5 g/d · 40 d | 3.1 g/d · 32 d | 6 reeds · 40 d | Already in the band; no change |
| Vanilla | 1.2 g/d · 83 d | 1.6 g/d · 62 d | 2.0 g/d · 50 d | 8 reeds · 50 d | Eight reeds reaches the top of the band |
Lemon's fill. At four reeds lemon still loses 2.6 g a day, so 100 g lasts 38 days. Raising the fill to 110 g (22 g oil + 88 g base, the same 20%) gives 42 days. A fill of 117 g would reach 45 days, but that is close to the brim of a 120 ml bottle once reeds go in; check headroom with water first. Keeping six reeds at 110 g would give only 31 days, so the reed change does most of the work. Three reeds might work on paper, but a lemon with three reeds may no longer smell like the product you launched.
Vanilla's ceiling. If eight reeds had left vanilla above the band, ten reeds might have brought it to about 43 days (a projection, not a measurement). Beyond the reed page's "6-8 reeds" guidance you are outside what the page describes, so test before you print it on a spec. After the change, the spread falls from 34 days to 10 across the three.
Life is not the only thing reeds change
The reed page itself says more reeds give "stronger throw and faster diffusion" and fewer give "a subtler scent and longer-lasting base". That trade is real: lemon at four reeds will usually smell less in week one than at six, and vanilla at eight may finally reach across the room. That is why the ladder records blind 3 m scores beside the weights, and why a count only passes if the scent still meets your line. A long-lasting variant nobody can smell is not a success; the post on a woody diffuser that lasts three months but cannot be smelt deals with that end of the trade.
Confirmation run. Make three bottles of each variant at its chosen set-up and run them for 35 days (9 bottles, 54 reeds), weighing every 3 days and sniffing at 3 m weekly. If the projected lives hold, write each variant's reed count, fill and band into its own document, as in the formula-specification post; production sheets per scent are the production-formula post. What you say to customers about life should come from several bottles and rooms, which the live guide on product-life claims covers.
What CSI sells for this, and the production run
| Step | Need | Buy | ₹ |
|---|---|---|---|
| Ladder, from unsold bottles | 162 reeds | Fiber Reeds Black 100 + 50 + 25 | ₹1,436.00 |
| Ladder, mixing fresh: bottles | 27 | Gold 120 ML × 3 packs | ₹3,186.00 |
| Ladder kit, mixing fresh | ₹4,622.00 | ||
| Oils, if fresh and CSI's | 180 g each | Citrus Lemon / Lavender / Dark Vanilla 2 × 100 gm | ₹600.00 / ₹1,080.00 / ₹1,580.00 |
| Confirmation | 54 reeds | left over from the ladder packs | — |
| Scale | 1 | Candle Making Weighing Scale | ₹354.00 |
The production run with per-variant counts
Worked example: 30 bottles of each variant. Reeds: lemon 4, lavender 6, vanilla 8, so 540 in all, exactly what a uniform six would use (540); 500 + 50 (₹4,400.00). Per bottle that is ₹32.00 of reeds for lemon, ₹48.00 for lavender and ₹64.00 for vanilla at the 50-pack rate, a ₹32.00 gap you may want reflected in pricing. Bottles: nine packs of 120 ML (₹9,558.00). Lemon at 110 g needs 660 g of oil and 2,640 g of base; lavender and vanilla stay at 600 g of oil and 2,400 g of base each. Hardware ₹13,958.00, ₹155.09 a bottle, before oils and base, excluding labour, freight, packaging and GST. Bundle reeds per variant at packing so the right count goes in the right box.
What to buy, and what to skip
Chosen on page statements, price and stock on 7 October 2026, not on performance. Oils are listed only for the case where your variants use CSI oils.
| Pack | Price | Per reed | Use |
|---|---|---|---|
| Pack of 25 | ₹236.00 | ₹9.44 | with 100 + 50 for the ladder |
| Pack of 50 | ₹400.00 | ₹8.00 | the run: with a 500 for 540 |
| Pack of 100 | ₹800.00 | ₹8.00 | ladder: 100 + 50 + 25 for 162 |
| Pack of 500 | ₹4,000.00 | ₹8.00 | production, after confirmation |
| Pack | Price | Per reed | Use |
|---|---|---|---|
| Pack of 25 | ₹265.00 | ₹10.60 | a colour code per variant, if you want one |
| Pack of 50 | ₹450.00 | ₹9.00 | |
| Pack of 100 | ₹900.00 | ₹9.00 | |
| Pack of 500 | ₹4,500.00 | ₹9.00 | dearer than black per reed |
| Pack | Price | Per bottle | Use |
|---|---|---|---|
| 50 ML Pack of 10 | ₹826.00 | ₹82.60 | a different product; don't mix sizes in the ladder |
| 120 ML Pack of 10 | ₹1,062.00 | ₹106.20 | ladder if mixing fresh: three packs (₹3,186.00) |
| Item | Price | — | Use |
|---|---|---|---|
| Default Title | ₹354.00 | — | weigh every bottle, every 3 days |
| Size | Price | Per gram | 100 g bottles at 20% | Use |
|---|---|---|---|---|
| 15 gm | ₹81.42 | ₹5.43 | 0 | too little |
| 50 gm | ₹170.00 | ₹3.40 | 2 | too little for nine bottles |
| 100 gm | ₹300.00 | ₹3.00 | 5 | ladder if fresh: 2 × 100 gm (₹600.00) |
| 500 gm | ₹1,420.00 | ₹2.84 | 25 | run: 660 g at 110 g fills |
| 1 kg | ₹2,832.00 | ₹2.83 | 50 | barely cheaper per gram |
| Size | Price | Per gram | 100 g bottles at 20% | Use |
|---|---|---|---|---|
| 15 gm | ₹130.00 | ₹8.67 | 0 | too little |
| 50 gm | ₹410.00 | ₹8.20 | 2 | too little for nine bottles |
| 100 gm | ₹790.00 | ₹7.90 | 5 | ladder if fresh: 2 × 100 gm (₹1,580.00) |
| 500 gm | ₹3,840.00 | ₹7.68 | 25 | run of 30: 600 g, so 500 + 100 gm |
| 1 kg | ₹7,646.00 | ₹7.65 | 50 | barely cheaper per gram |
Skip for now: a new base for one variant (a different formula, and none of our bases is positioned on its page as a slower base for citrus); 50 ML bottles for a 120 ml product's ladder; the 500-pack of reeds before confirmation; more oil to "slow down" or "speed up" a variant. CSI HyperDiffusion Base is out of stock on 7 October 2026 and isn't offered.
CandleMakingSuppliesIndia sells the reeds, bottles, scale and oils named here. This page still says the ladder can be run with your own unsold bottles, that per-variant counts don't increase your reed total, and that a variant already in band needs nothing from us.
Every price was checked on the live store on 7 October 2026, with stock checked per size; every item priced is in stock on that date, and HyperDiffusion Base is not offered. Reed-page counts and claims are quoted and attributed as page wording.
Fragrance behaviour is described as general tendencies, never as the composition of a CSI oil, and no viscosity, density or flash-point figure is given. The variants, rates, lives, band, ladder and rule are worked examples and in-house methods; the run quantities are worked examples. For bulk pricing, GST invoicing, and MSDS and IFRA documentation on request, message us on WhatsApp at +91 7397976926.
Frequently asked questions
- I Made Lavender and Jasmine Reed Diffusers Using the Same 20% Fragrance Oil Formula, but Lavender Smells Strong and Jasmine Barely Smells — Should I Change the Base or Fragrance Oil?
- I Made Five Reed Diffusers with the Same Ready-Made Base, but Only Two Fragrances Have Good Throw — How Do I Find Out Why the Other Three Are Weak?
- My Lemon Reed Diffuser Evaporates Quickly While My Sandalwood Reed Diffuser Barely Loses Liquid — Can Fragrance Oil Composition Affect Consumption Rate?
- I Want to Manufacture Ten Reed Diffuser Fragrances — Should I Use One Standard 20% Formula or Test a Different Fragrance Percentage for Every Oil?
- My Reed Diffuser Base Works with Lavender at 15% but Requires More Testing with Rose at 20% — How Do I Establish Individual Formula Specifications?
- I Want to Make a Reed Diffuser Collection Using One Base — How Do I Screen 20 Fragrance Oils Before Choosing the Five That Perform Best?
- I Am Making Reed Diffusers with Fragrance Oils from Different Suppliers — How Do I Compare Their Diffusion Performance Without Relying on Bottle Strength?
- How Do I Develop a Multi-Fragrance Reed Diffuser Collection When Different Oils Have Different Volatility, Solubility, Wicking Behaviour and Fragrance Throw?
Customer reviews: The six reviews at the top of this page are genuine, published Judge.me reviews left by CSI customers on the product pages named on each card — Cinnamon Vanilla, Citrus Lemon Fragrance Oil, Dark Vanilla Fragrance, Gold Reed Diffuser Bottle Pack of 10 (50 ML and 120 ML), Lavender Fragrance Oil, Lift Base – Professional Grade Reed & Car Diffuser Base (500 ml). They are general product reviews, not assessments of the reed-diffuser fragrance advice on this page. Names, star ratings, dates and products are as recorded by Judge.me and wording is unedited. "Verified buyer" appears only on reviews Judge.me recorded as verified — Anonymous, Avanthi Cheeli, Tina Raha, snehal Chaudhary, Punam Sarma, Sooraj R. Reviews below five stars are included as published, not filtered out.
Figures verified 7 October 2026, CSI live pricing: Fiber Diffuser Reeds (Black) 25 ₹236.00 (₹9.44 a reed), 50 ₹400.00, 100 ₹800.00, 500 ₹4,000.00 (₹8.00 a reed); Fiber Diffuser Reeds (White) 25 ₹265.00 (₹10.60 a reed), 50 ₹450.00, 100 ₹900.00, 500 ₹4,500.00 (₹9.00 a reed); Gold Reed Diffuser Bottle Pack of 10, 50 ML ₹826.00 (₹82.60 each), 120 ML ₹1,062.00 (₹106.20 each); Candle Making Weighing Scale ₹354.00; Citrus Lemon 100 gm ₹300.00, 500 gm ₹1,420.00 (₹2.84 a gram); Lavender 100 gm ₹540.00; Dark Vanilla 100 gm ₹790.00, 500 gm ₹3,840.00; CSI HyperDiffusion Base out of stock. Per unit = pack price ÷ units, two decimals. Totals exclude labour, freight, packaging and GST.
Non-price figures: "three fragrance variants", "the same formula" and lives that "differ by several weeks" are the reader's (title). The three oils, 20% load, 100 g fill, six reeds, grams-per-day rates, lives, the 40–50-day band and the 110 g lemon fill are a worked example, not CSI test data; the ladder, 21- and 35-day lengths, sniff method and decision rule are in-house methods. Reed counts and claims and every quoted line are CSI page wording. Diwali (8 November 2026) and Christmas (25 December 2026) are calendar facts.