Recipe and specification are different documents

A recipe says: melt the wax, add fragrance when hot, stir well, pour when it looks right. Every one of those phrases relies on judgement. You have that judgement, built from dozens of small batches, and that is why your candles come out well. Nobody else has it, and even you do not have it reliably at 10 kg when the batch is too large to read by eye.

A specification replaces judgement with a measured target and a range. It does not tell a skilled maker anything new. It tells a less experienced person, or a tired expert at the end of a long day, exactly what counts as right and how to check it.

Notice that the specification column does not invent anything. Where the wax maker states a figure, such as the 80–90°C addition temperature and 2-minute stir for CSI 400, or the 58–60°C pour for CSI 468, it uses it. Where no figure is published, it says 'validated', meaning a number you established through your own tests and wrote down.

The triggers that mean it is time

Batch weight is a poor trigger on its own. A maker pouring 15 kg alone, carefully, in one short session may be more consistent than two people pouring 5 kg while talking. Watch for these instead.

A second person makes or pours the candle

This is the strongest trigger. The moment someone else weighs, scents or pours, your judgement is no longer in the batch. Two people who both believe they are following the recipe will interpret 'stir well' differently. Write the specification before their first batch, not after their first mistake.

The batch needs more than one melt or a long session

At 10 kg, a single melt fills 50 candles of 200 g exactly, 48 after 5% wastage. Beyond that, you are making several melts that must match each other, or pouring one melt for long enough that the wax drifts. Either way, you need a written standard that each melt and each pitcher can be checked against.

A buyer expects identical reorders

A retailer, a corporate client or a subscriber who reorders will compare this month's candle with last month's. Without a specification, you cannot promise they match, and you cannot prove it if they complain.

You cannot explain a difference

If two batches have come out differently and you do not know why, you are already past the point where a recipe is enough. A specification with records would have told you.

What to specify, and what to leave alone

Over-specifying is a real risk. A thirty-page document for a candle nobody reads is worse than a one-page recipe everybody follows. Specify only the steps that change the finished candle, and keep everything else as plain instructions.

Things like the order in which jars are labelled, where the stirring sticks live or who cleans the melter can stay as ordinary instructions. They matter for tidiness, not for the candle.

How to write the first specification

You do not need a template or software. You need one page per product, written from evidence. The fastest way to produce it is to make one carefully observed batch at the size you intend to run, and write down what you actually did.

Take a 10 kg batch of your bestseller. Before you start, write the targets you believe in: 9.2 kg wax and 0.8 kg oil at 8%, addition at the wax maker's stated temperature, your usual stir, your usual pour temperature. During the batch, record the actual figure beside each target. Pour-number the candles and keep a few from across the run. After cure, burn them against a reference candle.

Where the actual figures and the burn results agree with your targets, the targets become the specification. Where they did not, for example the stir time you believed in left the floor samples stronger than the top, you adjust the target and repeat that part. After two or three batches, the page is a specification rather than a recipe, and every number on it has been checked.

What it costs, and what it saves

Writing a specification costs a few careful batches and some patience. It needs almost no equipment beyond what you already use: a scale, a separate thermometer and a way to number candles. What it saves is harder to see until you need it: the batch that would have gone out weak, the employee who would have poured too cool, the argument with a buyer you can now settle with a record.

There is also a quieter benefit. Writing targets down forces you to notice which steps you had never actually measured. Many makers discover, on their first observed batch, that they had no idea what temperature their fragrance went in at, or that their fill weights varied more than they believed. Finding that out on your own terms, before a buyer does, is worth far more than the hour it takes to fill in the page.

Where this page fits

CSI already has guides that cover parts of this question. This page takes the production-floor angle; these go deeper on the rest:

Three things to buy for this job

Frequently asked questions