Judging a blend against the last one

The useful question is never whether a blend is good. It is whether it is the same, and those require completely different arrangements.

Against what, not how good

Consistency is a comparison, so you need something to compare against: a retained sample from a batch you were happy with, stored sealed and cold so that it has aged as little as possible.

Without a retained sample you are comparing today's blend against your memory of it, and memory converges on whatever you have been tasting most recently. That is how a line drifts a long way in small steps with nobody noticing.

Setting up so the answer means something

Taste blind, with the samples coded by somebody else or by yourself the day before. Knowing which is the new batch determines what you find.

Smell dry first, then in warm water, then in a neutral carrier if the blend is used in fat. A blend can match dry and diverge in use, usually because a component's particle size changed and it now releases at a different rate.

Do it at the same time of day, not after coffee, and stop after three or four samples.

What to record

A verdict against the reference, in a form you will still understand in a year. Same, slightly stronger, slightly weaker, or different in a named direction. Plus the batch number, the date and who tasted.

Numbers are tempting and mostly false at this scale. A recorded direction is more honest and more useful than a score.

When your own palate is not enough

Your palate adapts to your own product, which means you are systematically the worst available judge of whether it has drifted. Two or three other people, tasting blind, catch what you cannot.

They do not need to be expert. They need to be consistent and to be asked the right question, which is whether these two are the same rather than which one they prefer.

References