Our principles
We score from the label, not from taste. Best In Aisle audits ingredient lists against a published, category-specific rubric. We do not physically test products. A score reflects ingredient quality and sourcing transparency as documented on the label and in publicly available information — nothing more, nothing less.
Every product is judged against standards built for its category. Butter and Greek yogurt are not the same kind of product, so they are not graded on the same scale. Each category has its own profile specifying which ingredients are positive signals, which are acceptable, which are red flags, and which are hard disqualifiers that trigger a Poor tier regardless of anything else.
Weights differ by category. A simple, two-ingredient product like butter is graded almost entirely on Ingredient Integrity and sourcing transparency. A complex product like whole grain bread has a longer list of meaningful quality signals, so Product Quality carries more weight. The weights for every category are published in the scoring database.
Scores can change. If a brand reformulates, gets acquired, or if we discover a sourcing error, we re-score and update the record with a new last verified date. We do not quietly change scores — any change is dated and logged.
Scoring dimensions
Every product is scored across four dimensions. The weighted average of those four scores determines the overall numeric score, which maps to a tier. Weights vary by category — see below.
What is actually in the product? We examine every ingredient against category-specific standards: positive signals (ingredients that indicate quality), acceptable fillers, red flags (problematic but not disqualifying), and hard disqualifiers that cap a product at Poor regardless of anything else.
Rewarded
- Short, recognizable ingredient lists
- Category-appropriate primary ingredients (e.g., 100% whole grain as first ingredient in bread)
- Absence of hard disqualifiers: artificial dyes, partially hydrogenated oils, HFCS, potassium bromate
- Sourcing signals: organic, grass-fed, single-origin, pasture-raised
Penalized
- Filler ingredients padded into the list
- Vague terms like 'natural flavors' without justification
- Red-flag additives: carrageenan, azodicarbonamide, DATEM, TBHQ
- Ingredient count above category maximum
How was this product made, and how far is it from its whole-food source? We assess the manufacturing processes documented on the label and in publicly available sourcing information — not speculation.
Rewarded
- Minimal, traditional processing appropriate to the category
- Cold-pressed, stone-milled, slow-fermented, cultured, or hand-harvested processes
- No chemical extraction, bleaching, deodorization, or bromination
- Process transparency (e.g., harvest date on olive oil, strain process on Greek yogurt)
Penalized
- Chemical refining processes (hexane extraction, bleaching, alkali treatment)
- Ultra-processing that strips whole-food character
- Processes that allow disallowed ingredients (e.g., partial hydrogenation)
Does the brand tell you what you need to know to make an informed decision? This dimension rewards verifiable disclosure and penalizes vague or misleading claims. It also sets the ceiling on Confidence — a product with low transparency cannot receive a High confidence rating regardless of score.
Rewarded
- Third-party certifications (USDA Organic, Certified Humane, Non-GMO Project, PDO/DOP)
- Specific sourcing disclosed: farm, country, region
- Harvest or production dates
- Lab testing results made publicly available
Penalized
- Misleading label claims (e.g., 'made with whole grain' when refined flour is primary)
- Welfare terms not backed by certification (e.g., 'natural' on eggs)
- No sourcing information disclosed at all
Does this product do its job well, beyond just its ingredient list? This dimension captures category-specific quality markers that matter for the product's intended use — protein density in Greek yogurt, fiber per slice in bread, polyphenol indicators in olive oil.
Rewarded
- Meets or exceeds category-specific nutritional benchmarks
- Certifications or testing that verify quality claims
- Category-relevant markers: protein, fiber, fat profile, probiotic count
Penalized
- Falls below category minimums (e.g., less than 10g protein per serving in Greek yogurt)
- Quality claims without any supporting evidence
How weights work by category
The four dimensions sum to 100% in every category, but their proportions differ based on what actually determines quality in that product type. Here are three examples from live categories.
Butter
Butter is a two-to-four ingredient product. What matters most is what's in it (cream quality, sourcing) and whether the brand proves it.Greek Yogurt
Straining process and live culture integrity matter as much as the ingredient list. Processing Level is weighted higher than in simpler categories.Whole Grain Bread
Bread has a long ingredient list with many opportunities for corner-cutting. Product Quality (fiber per slice, grain integrity) is weighted meaningfully alongside ingredient purity.The five tiers
A product's weighted score maps to a tier. Tier boundaries are category-specific — the minimum requirements for Excellent in butter are different from those in bread. The table below describes what each tier means in plain terms.
| Tier | What it means |
|---|---|
| Excellent | Meets all required criteria for the category. No red-flag ingredients. Strong sourcing transparency. Represents the best available at mainstream grocery stores. |
| Very Good | Solid ingredient list with no hard disqualifiers. Minor compromises — a less-transparent sourcing story, a filler or two — but nothing that undermines the product's core quality. |
| Good | Meets the baseline for the category. No hard disqualifiers. Acceptable for everyday use, but falls short on sourcing, processing, or ingredient quality compared to top-tier options. |
| Below Average | Present in the category but with meaningful quality compromises. May include red-flag ingredients, minimal transparency, or processing practices that undercut the label's implied quality. |
| Poor | Contains one or more hard disqualifiers for the category: artificial dyes, partially hydrogenated oils, HFCS, or practices that make the product unsuitable regardless of other attributes. |
Hard disqualifiers are absolute. A product containing partially hydrogenated oils, artificial dyes, or other category-level disqualifiers scores Poor regardless of its performance in other dimensions.
Confidence levels
Every score carries a Confidence level: High, Medium, or Low. This is not a softening of the score — it tells you how much publicly available evidence backs it up. Confidence is determined by a fixed rule applied after scoring, not by the scoring model itself.
Full ingredient label verified, strong sourcing documentation, third-party certifications confirmed.
Ingredient list verified but sourcing details are limited or unverified. Score is reliable; context is partial.
Limited publicly available information. Score is based on available evidence but should be treated as provisional.
The confidence cap rule: A product with an Evidence & Transparency score below 0.25 is capped at Low confidence regardless of tier. A score between 0.25 and 0.50 is capped at Medium if it would otherwise be High. This rule is enforced deterministically — it is not a judgment call.
What we don't score
BIA does not score taste, price, sustainability packaging beyond category-specific packaging standards, or any attribute that cannot be verified from the product label and publicly available sourcing documentation.
We do not physically test products in a lab. We cannot verify a brand's manufacturing claims that go beyond what is disclosed on the label or through third-party certifications. When claims are unverifiable, they are treated as absent for scoring purposes, which is reflected in the Evidence & Transparency dimension.
We also distinguish between a confirmed label change and an unverified taste complaint. If a product's ingredient list has not changed, we note it. If a documented label change exists, we flag it with dated evidence. We do not treat Reddit sentiment as evidence of reformulation.
Rubric changelog
When the rubric changes, we update this log and re-score any categories affected by the change. We do not silently update scores.