Roasting at home
Logging a roast
Six numbers and one sentence. Without them, every roast is a first roast.
- Skill level
- Intermediate
- How long it takes
- A minute per batch
- Equipment needed
- Home coffee roaster, Digital scale
- Steps
- 7
Why it works this way
The difference between someone who has roasted a hundred times and someone who has roasted their first batch a hundred times is a log. Roasting has more variables than brewing and they interact, so memory is not equal to the task — and the specific failure is subtle: you remember the roast you liked and not the conditions that produced it, so you cannot get back to it. What follows is the smallest record that makes a roast repeatable, and it fits on an index card.
What you need
Required
How to do it
Record the coffee, the batch weight in, and the ambient conditions.
A cold kitchen roasts differently from a warm one, and the same roaster behaves differently on its first batch of the day than on its fourth.
Record the charge temperature or heat setting.
Where you started, in whatever units your roaster gives you.
Note the time the colour turns yellow.
The end of drying, and the first checkpoint you can actually see.
Note the time first crack begins.
The landmark everything after it is measured against.
Note the drop time, and calculate how long you spent after first crack.
Many roasters track the stretch after first crack as a share of the total roast. It is a convention for comparing your own batches; no study read for this site shows that a particular share predicts flavour.
Weigh the roasted coffee and record the weight loss as a percentage.
For one green coffee, weight loss indicates how far the roast went; between different coffees it says much less, because most of the loss is water. It is not a substitute for tasting, and it is the easiest objective number you have.
Taste it after resting, and write one sentence.
In your own words. A log with no sensory line records what you did and not whether it worked.
How to tell it worked
Not a sensory outcome — a record that lets the next roast be a deliberate change rather than another attempt.
If it did not work
What you saw, what to change, and why the change works.
Two roasts with identical numbers tasted different
Change: Look for a variable you are not recording: ambient temperature, batch order, or the green coffee's own moisture.
Green coffee is an agricultural product and changes across a bag and across a season.
Common mistakes
Recording only the total time
Two roasts of the same length can have completely different phase structures and taste nothing alike.
Instead: The landmarks matter more than the total.
Logging without tasting
A record of what you did with no way to tell which was better.
Instead: One sentence after resting.
Diagnose a specific problem
The science behind it
- First crack
The audible cracking sound that marks a key roasting milestone.
- Roast development
How far a roast is taken past first crack, independent of total roast time.
Tools that help
Background reading
Learn next
Where this is taught
Become a Coffee Expert · Level 7
What heat does to a green seed, stage by stage, and what a roaster actually controls.