Buy a full case and the price per unit drops automaticallyRequest a sample
AccountCart0
All ContainersBottlesJarsCaps & ClosuresPumps & SprayersTubs & PailsJugsDroppersTinsTubesAccessories
Industries
Personal CareShop all 934 items →
AromatherapyBathBath →Body careBody care →Color cosmeticsColor cosmetics →DeodorantDeodorant →FragranceFragrance →GroomingHair careLip careLip care →NailNail →Oral careOral care →Skin and faceSkin and face →Soap and washSoapmaking →Sun and protectionSun and protection →
Learn:BathBody careColor cosmeticsDeodorantFragranceLip careNailOral careSkin and faceSoapmakingSun and protectionEverything we have written →
Learn
Beekeeping100Hot sauce making90Jam and preserves96Pickles and ferments64Sauces and condiments36Syrups36Oils and vinegars34Nut butters and spreads33Dairy and alternatives33Coffee and tea36Confection31Resin and casting27Supplement liquids23Laboratory17Dispensing13Photography and darkroom14Industrial chemical14Prepared and deli15Body care14Color cosmetics13Bath14Pet health15Toiletry decanting14Oral care14Capsules and tablets14Nail14Fragrance15Deodorant17Fossils16Janitorial16Baking and dry goods16Seed storage17Specimens16Paint and ink18Automotive fluids18Wax melts17Diffusers17Detailing15Sun and protection18Dish and hand16Soapmaking91Candle making93Beverages92Cleaning products92Lip care89Minerals and rocks93Powders91Skin and face91Spices and seasoning91Findings and components16
All 1,895 pages we have written →
Coffee and tea
All Containers›Shop by use case›Coffee and tea›What goes wrong›The batch is inconsistentShop all coffee and tea →
What goes wrong

The batch is inconsistent

By Dwight Davis, Founder

Updated

On this page7
Two batches that should match and do not. Three causes, and the roast log tells you which.

The profile drifted

Ambient temperature, machine temperature after several batches, or a different batch size. Comparing the logged profiles of the good batch and the bad one usually shows the difference immediately, which is the entire argument for logging the curve rather than the settings.

The green lot changed

Moisture, density and size all vary by lot, and the same profile gives a different result. If the change coincides with a new sack, this is the likely story and the answer is to adjust the profile rather than conclude the roaster is wrong.

A blend component changed

For a house blend this is the most common cause and the easiest to miss, because the recipe did not change and one of its inputs did. Recording the lots that went into each blend batch is what makes this visible.

Or somebody else roasted it

Which is a reason the profile has to be written down rather than held by one person. A roast that depends on an individual’s judgment is a roast that changes when they are away, and that is a business risk rather than a quality one.

Taste against the retained sample

Not against memory. Keeping a small sealed sample from each roast makes the comparison real, and it is the only way to catch slow drift, which is invisible from one batch to the next.

What a wholesale buyer will ask

Whether next month tastes like this month. In coffee that question is asked more often than in almost any other property here, because cafes build drinks around a specific coffee and notice immediately when it moves.

References

21 CFR 101, Food Labeling, which carries the general requirements that apply here in the absence of a standard of identity: the statement of identity, net quantity, ingredient list and the name and place of business of the responsible party. law.cornell.edu
Written from federal regulation and from the published physics and chemistry of staling, not from a laboratory. There is no standard of identity for coffee or tea, so most of what governs a product here is the general requirement that labels be truthful and claims be substantiated. Anything here is a starting point for a conversation with your state department of agriculture, not a determination that you comply.