How we rate coffee beans

The heaviest criterion in the category and the one no specification can answer. It is left without a number rather than filled with a proxy.

What we can actually read: Deliberately unscored. A coffee scored out of 60 rather than 100 is a coffee we have not tasted, and the panel says so instead of quietly counting it as zero.

Not scored. No specification answers this, so it is left without a number.

Coffee is a perishable good sold as a shelf-stable one. A roast date is the single most useful thing a bag can carry, and most bags do not carry it.

What we can actually read: Roast date, best-before-only, days from roast to ship and a one-way valve are all published or plainly absent from the bag.

Scored directly from published figures.

3 of the 4 rules below are struck through in this category, because too few makers publish the input. The one that remains scores this criterion out of 10 between it: a struck rule leaves the sum entirely rather than counting as a zero. That result is then worth 20 of the category's 100 points.

• Roast date printed4 of 10

  present

  4 pts

• Not best-before onlynot scored here

  Fewer than one maker in four publishes this here, so its 2 points leave the sum rather than being guessed at.

  absent

  2 pts

• Ships after roastnot scored here

  Fewer than one maker in four publishes this here, so its 3 points leave the sum rather than being guessed at.

  3 or less

  3 pts

  over 3, up to 7

  2 pts

  over 7, up to 14

  1 pt

  over 14

  0.5 pts

• One-way valvenot scored here

  Fewer than one maker in four publishes this here, so its 1 points leave the sum rather than being guessed at.

  Also read under Packaging & Freshness Retention, which asks a different question of the same fact.

  present

  1 pt

10 of 10

How much the roaster will tell you about where the coffee came from, which is both an ethical question and a quality one: a roaster who names the farm usually has a reason to.

What we can actually read: Origin, altitude, varietal, certifications and direct trade are published in descending order of frequency. Almost nobody names a farm.

Scored directly from published figures.

2 of the 7 rules below are struck through in this category, because too few makers publish the input. The 5 that remain score this criterion out of 10 between them: a struck rule leaves the sum entirely rather than counting as a zero. That result is then worth 16 of the category's 100 points.

• Origin type1.5 of 10

  single

  1.5 pts

  blend

  1 pt

• Origin named1.5 of 10

  published at all

  1.5 pts

• Altitude publishednot scored here

  Fewer than one maker in four publishes this here, so its 1.5 points leave the sum rather than being guessed at.

  published at all

  1.5 pts

• Varietals publishednot scored here

  Fewer than one maker in four publishes this here, so its 1.5 points leave the sum rather than being guessed at.

  published at all

  1.5 pts

• Certifications1.5 of 10

  2 or more

  1.5 pts

• Direct trade1.5 of 10

  present

  1.5 pts

• Farm or co-op named1 of 10

  present

  1 pt

10 of 10

Price per kilo against the category median, not price per bag, because bag sizes here run from 283g to a kilo.

What we can actually read: Needs a published price here and on at least three rivals.

Computed from price, not from a rule list

Value is the only criterion here with no thresholds of its own, because it cannot be read off one product. It asks what else the same money buys in this category, so it needs the category. That result is then worth 14 of the category's 100 points.

Points already earned on everything else

out of 10

Multiplied by the category median over this price

median ÷ price

Capped at

10

No price published, or fewer than 3 priced rivals

5 of 10

Half marks is not a verdict, it is a refusal to guess: with nothing to compare against, a product is neither good value nor bad, and scoring it either way would invent a finding. It corrects itself as the category fills.

Whether the same bag tastes the same twice. It cannot be read off a label.

What we can actually read: Deliberately unscored, for the same reason as taste.

Not scored. No specification answers this, so it is left without a number.

A one-way valve lets roasting gas out without letting oxygen in. Without one, a bag is either stale or bulging.

What we can actually read: Packaging type, valve and bag sizes are published.

Scored directly from published figures.

2 of the 3 rules below are struck through in this category, because too few makers publish the input. The one that remains scores this criterion out of 10 between it: a struck rule leaves the sum entirely rather than counting as a zero. That result is then worth 10 of the category's 100 points.

• Packagingnot scored here

  Fewer than one maker in four publishes this here, so its 2.5 points leave the sum rather than being guessed at.

  tin, valve

  2.5 pts

  foil

  2 pts

  bag

  1.5 pts

  box

  1 pt

• One-way valvenot scored here

  Fewer than one maker in four publishes this here, so its 5 points leave the sum rather than being guessed at.

  Also read under Freshness & Roast Date Transparency, which asks a different question of the same fact.

  present

  5 pts

• Bag size2.5 of 10

  250 g or less

  1.5 pts

  over 250 g, up to 500 g

  2 pts

  over 500 g

  2.5 pts

10 of 10