The uniformity test, and how to run it

Blend uniformity is the measurement that converts "we mixed it well" into evidence, and at small scale it is more approachable than you would assume.

What you are measuring

The variation in composition between samples drawn from different locations in the same batch. That variation is what a customer experiences as differing servings, and in this category it is a potency question rather than a quality one.

A non-uniform blend produces varying nutritional content per scoop, which means the panel is true on average and false for most individual servings.

Sampling properly

A single sample tells you nothing about uniformity, because uniformity is a comparison.

Take samples from multiple locations: top, middle and bottom of the blender, and from different horizontal positions. Use a sampling thief so you draw from depth rather than the surface.

Then sample again from the start, middle and end of the filling run, because a hopper segregates as it empties and filling is a second chance for the batch to become non-uniform.

What to measure

For a flavored product, taste the samples side by side. It is free, fast, and detects segregation of small components better than most alternatives, because flavor and sweetener are exactly what separates first.

For anything where potency matters, assay a marker at a contract laboratory. The variation between those results is your uniformity number.

Both together is the practical answer: taste as a routine screen every batch, assay on first batches and on change.

Setting an acceptance criterion

Write a number into your specification rather than judging by eye. What counts as acceptable depends on how much of the label claim variation would consume.

Then record the result for every batch, pass or fail. A trend across batches is the blend time or a material changing, and it is invisible unless the numbers sit next to each other.

References