ProteinScore logoProteinScore

How we score protein efficiency

One score, one public formula, no secrets. The exact maths behind every ProteinScore badge is below — this is what makes us different from black-box ratings.

The formula

All nutrition inputs are per 100 g, taken from the product's UK label. The score answers a single question: what share of this product's calories come from protein? The more calories that come from protein, the more “protein-efficient” the product is.

proteinKcal = proteing × 4

Protein provides 4 kcal per gram — the Atwater factor printed on every UK nutrition label.

proteinPct = proteinKcal / kcal

The share of the product's calories that come from protein, between 0 and 1.

kcalPerGProtein = kcal / proteing

The supporting metric: how many calories you “spend” for each gram of protein. Lower is better.

efficiencyScore = round(100 × min(proteinPct / 0.5, 1))

100 when at least 50% of calories come from protein; linear below that. A product with 25% of calories from protein scores 50.

Grade: score ≥ 80 → A · ≥ 60 → B · ≥ 40 → C · ≥ 20 → D · else → E

Worked example

Take a real product in our directory: a high-protein pancake mix with 359 kcal and 56 g protein per 100 g (values from the manufacturer's UK label).

proteinKcal = 56 × 4 = 224proteinPct = 224 / 359 = 0.624 (62.4%)kcalPerGProtein = 359 / 56 = 6.4efficiencyScore = round(100 × min(0.624 / 0.5, 1)) = round(100 × 1) = 100
100A

Over half its calories come from protein, so it scores a perfect 100 — grade A. You spend just 6.4 kcal per gram of protein.

Edge cases

Insufficient data
If kcal or protein is zero or missing, all metrics are left empty. The product is excluded from ranked and sorted lists and shows an “Insufficient data” badge instead of a score. We never invent numbers to fill the gap.
Flagged for review
If the data implies more protein calories than total calories (proteinPct > 1), the value is clamped to 1 and the product is flagged for human review — that always means bad source data, so it never appears with a score until corrected.

How we verify the data

  • Manufacturer label first. We take nutrition data from the brand's own label or nutrition page wherever possible — it is the most authoritative source. Every product page links to the source we used.
  • Monthly re-verification. We re-check the nutrition of the most-viewed products every month — labels change and reformulations are the number-one risk to data accuracy.
  • Quarterly audits. We audit the full catalogue each quarter (sample-based once the catalogue grows beyond 500 products).
  • Never auto-published. Nutrition data is never published automatically from third-party databases — every figure is human-verified before it goes live.

Rankings are independent of commission

We may earn a commission if you buy through the links on product pages — it keeps the lights on. But scores and rankings are computed from nutrition data alone. Commission never affects a product's score, its grade, or its position in any listing. Sponsored products, when we introduce them, will be clearly labelled and will never rank higher because of it.

← Back to the directory