The Batch You Can’t Repeat Is a Batch You Didn’t Really Make
You poured twelve candles four months ago and they were the best you’ve made. Now you can’t remember whether that was the 8% pour or the 10% one — or which of the two waxes was in the cupboard that week.
It’s a specific and slightly maddening kind of loss. The batch was genuinely good: excellent throw, clean burn, sold out in two weeks. And now, four months later, you’re standing in front of your supplies trying to reconstruct it from memory and getting nowhere. Was that the 8% or the 10%? The CD-14 or the CD-12? Which soy wax? That isn’t carelessness, and it happens to everybody, for a simple reason: the details that matter are boring at the moment they matter and gone by the next morning. Good batch record keeping is one habit — write it down before you clean up — and it’s the difference between a craft that compounds and one that keeps restarting.
Give every batch a number, and put it on the label
The foundation is a batch number. Date plus a sequence is plenty: something like 260810–01.
Two places it needs to exist: in your log, and on the product label. That second one is what turns a record into traceability. When a customer contacts you about a candle six months from now, the number on the base finds the row — the wax, the wick, the fragrance, the load, the date, and what you thought of it at the time.
That matters for ordinary reasons (a customer asks which fragrance a gift was) and for serious ones (something goes wrong with a batch and you need to know exactly which units share that formulation). Batch records are also, in many places, part of what selling consumer products actually requires — which differs by country and is worth establishing properly.
The column that earns its keep is “what happened”
Recording the inputs is the obvious part. The column people leave out is the outcome — and it’s the one that makes the whole log valuable.
Consider three rows from a worked example:
- Sea Salt at 6% → “Poor — weak throw”
- Amber & Oud at 12% → “Failed — FO seeped, over max load”
- Amber & Oud at 8% → “Good”
Three lines, and the question of what load to use for that fragrance is answered permanently. Nobody remembers that after a month. Everybody has it forever once it’s written down.
Notice that two of those three rows are failures. That’s the point. Failed and mediocre batches should be recorded as deliberately as the good ones, because they’re where the boundaries are. A batch that seeped tells you where the ceiling is. A batch with weak throw tells you where the floor is. Only logging successes throws away the expensive half of your information.
And on that middle row specifically: going over a fragrance’s maximum load doesn’t produce a stronger candle. It produces one that seeps. The maximum is set by the fragrance supplier, differs per fragrance and per application, and has to come from them — not from a rule of thumb and not from a spreadsheet.
Cure dates, calculated
The other thing worth having in writing is when something is actually ready.
Both crafts involve waiting — cure time for soap, and a rest for candles before the fragrance settles and behaves. And “about a month” is exactly the kind of thing that becomes “I think that’s probably fine” when a market is coming up and you’re short of stock.
Get Diginesian’s stories in your inbox
Join Medium for free to get updates from this writer.
A date made, a cure period, and a status that reads READY or CURING removes the judgement call. You know the exact day, and you don’t sell something early because you were optimistic on a Thursday.
And now the one that matters most: lye
There’s a feature that would obviously belong in a soap-making spreadsheet, and it should not be there.
A spreadsheet should not calculate your sodium hydroxide.
The amount of lye a soap recipe needs depends on the saponification value of every oil in it. Get it wrong in one direction and the bar is lye-heavy and caustic — it can burn skin. Get it wrong the other way and you have a soft, greasy bar prone to going rancid.
Dedicated lye calculators are free, widely used, actively maintained, and tested against known SAP values. A spreadsheet downloaded from a marketplace is none of those things. If a formula in it is wrong — a mistyped SAP value, a cell reference off by one, an edit made by a user who didn’t realize what they were changing — nobody catches it, and the consequence is a caustic product on someone’s skin.
No template author should be asking anyone to trust a spreadsheet with a caustic chemical. Including, I’d say, template authors who make things they’re proud of.
So the right division of labour is: calculate there, record here. Run the recipe through a proper lye calculator, then record what it told you — the NaOH, the water, the superfat, and the date you verified it. Now the batch is repeatable and the number came from a tool built for the job.
One rule that follows: change an oil, a quantity, or the superfat, and run it through the calculator again before you make it. A recorded lye figure belongs to one exact formulation, not to a family of similar ones.
The pre-selling list you have to research yourself
The last thing worth writing down is what you need to have in place before you sell anything at all — and this is one where no template can give you the answers.
Safety assessments, allergen declarations, hazard labeling, warning labels, ingredient listing, batch numbers, registration, insurance, record retention. All of those exist in some form in most markets, they differ substantially between countries, and they change.
What a checklist can do is make sure you’ve considered each one and can tick it off against your own research for your own market. What it cannot do is tell you the rules where you live. Anyone selling you a template that claims otherwise is selling you a false sense of security — and product liability insurance is worth thinking about regardless.
An important note
To be clear: this is a general perspective on record-keeping. It is not safety, chemical, formulation, regulatory, legal, or business advice, it is not training, and it guarantees no outcome. All recipes, fragrances, quantities and figures mentioned are illustrative examples. Never use a general spreadsheet to calculate sodium hydroxide — soap made from an incorrect lye figure is caustic and can cause chemical burns; always use a dedicated, tested lye calculator and follow the safety data sheet for every material. Fragrance maximum usage rates are set by the fragrance supplier, differ per fragrance and per application, and must be obtained from them. Wick and cure figures are common working targets, not standards. Candles and soap are consumer products subject to safety, labeling, ingredient-declaration, registration and record-keeping requirements that differ substantially by country and change — establish and comply with the ones that apply where you make and sell before selling anything, and consider product liability insurance.
If you want it structured for you
I built a batch log in Google Sheets around this habit — every batch numbered and traceable to the label, with fragrance percentage, units, cost per unit and an outcome column that treats failures as data, plus cure dates that calculate a READY status, and a soap sheet that records your lye calculator’s results rather than producing them:
👉 Candle & Soap Making Batch Log for Google Sheets & Excel
Whether you use mine or a notebook by the pour station, write it down before you clean up. Ten seconds while the wax is still warm buys you a batch you can make again — and the best thing you ever poured is only worth something if you can pour it twice. 🕯️
This reflects my own perspective and describes a record-keeping tool — NOT safety, chemical, formulation or regulatory advice, and it is not training. Never use a general spreadsheet to calculate lye: use a dedicated, tested lye calculator and follow the safety data sheet for every material. Establish the rules where you sell before selling anything. Makers: what’s the batch you lost and never managed to recreate? Tell me in the comments.
