The spoonful that passed the test

An original exposition about testing a recipe for an invented community lunch.

A week before our community lunch, we cooked six portions of vegetable stew. The tasting group agreed that the seasoning was balanced and the sauce had the right thickness. On the day of the lunch, we followed the same recipe for sixty people. By serving time, the stew near the bottom of the large pot was thick, while the liquid at the top was still thin.

At first, we said that the trial had failed. That was not quite fair. The small batch had answered the question we had actually asked: whether the proportions of vegetables, liquid and seasoning produced a pleasant flavour. It had not shown us how easily ten times that amount could be mixed in unfamiliar equipment.

For the next event, we carried out two different tests. A small pan allowed us to adjust the seasoning without wasting many ingredients. We also filled the large cooking pot halfway and practised stirring it. This revealed that our usual spoon did not reach the bottom and that the mixture needed to be stirred for longer. Each trial was useful because it examined a different uncertainty.

Testing still has limits. A full-sized trial would use almost as much time and food as the event itself, while the preferences of a few tasters may not represent every guest. Repeating tests without a clear purpose can also delay a decision rather than improve it. A useful trial does not promise that nothing will go wrong; it reduces a particular risk that has been identified in advance.

How does the writer explain the value and limitations of small-scale recipe trials?