How a ranking gets made.
theRANKS scores every product across five independent dimensions, combines them into a single composite, and publishes the result - unaltered. This page is the receipt: what we measure, how the math works, and where independence ends and judgment begins.
One number, five dimensions.
Every ranking is a composite score across five independent dimensions - applied to aggregated market and review data, normalised against the category, and combined into a single number from 0 to 100. That number is the rank.
Each ranking, scored five ways.
Demand
What the market is actually reaching for - purchase intent, aggregate search interest, and category share-of-attention.
- Data
- Search-volume curves, marketplace purchase intent signals, social interest indices.
- Why
- A product nobody wants is rarely the right answer, even when it wins every spec sheet. Demand grounds the ranking in the market rather than in a lab.
Trust
Weighted review credibility - verified-purchase ratio, reviewer authority, recency, and the depth of long-tail evidence.
- Data
- Aggregated review platforms; verified-purchase filtering; reviewer history weighting; recency decay.
- Why
- A 4.8 score from twelve buyers is not the same as a 4.6 score from twelve thousand. Trust separates noise from signal.
Value
Price measured against category norms and performance tier - not absolute cheapness.
- Data
- Live category price distribution; performance-tier benchmarking; price-history curves.
- Why
- The cheapest option rarely wins. The right price for the right tier wins. Value is what you actually paid relative to what you got.
Quality
A composite of aggregate ratings, third-party editorial scores, and specification benchmarks against category leaders.
- Data
- Editorial review consensus; specification deltas vs the category frontier; aggregate satisfaction.
- Why
- Demand and trust tell us what the market thinks; quality tells us whether the market is right.
Durability
Long-tail signals on whether a product survives daily use - return rates, warranty data, and product lifecycle behaviour.
- Data
- Return-rate analyses; warranty terms and claim depth; long-tail review behaviour past 12 months.
- Why
- A product is only as good as the second year of owning it. Durability is the dimension most often missing elsewhere.
How a score is computed.
Gather
Aggregated market and review data is collected continuously from the platforms where products are actually sold and discussed - never from a single source, never from a survey panel, never from a paid placement.
Normalise & weight
Each dimension is computed independently and normalised against its category, so a $80 air fryer and a $400 espresso machine can both score on Value without one collapsing the other. Weights are stable per category and published with the model.
Composite
The five normalised dimensions are combined into a single 0–100 composite that is the published rank score. No manual adjustments. No editorial overrides. The composite recalculates whenever new market or review data arrives.
Where the algorithm ends, judgment continues.
The composite score is the backbone of every ranking on theRANKS - applied identically to every product, computed from the same five dimensions, recalculated whenever new market or review data arrives. For a subset of categories, we also conduct first-party hands-on assessment when the gap between the top contenders comes down to a judgment call the data alone can’t settle. We do not label any specific product as “hands-on tested” without a verified per-item signal that it was, and we never claim hands-on as a site-wide practice. The algorithmic composite is the universal, auditable backbone; hands-on is an additive input where it earns its place.
The scoring model receives no revenue signal.
theRANKS earns affiliate commissions from qualifying purchases through links on rankings pages. Affiliate relationships have zero influence on dimension scores or ranking order - the scoring model receives no revenue signal. The ranking is a direct output of the algorithmic composite and would be identical if no affiliate programme existed.
Brands cannot pay for placement, position, score adjustments, or removal. We have no sponsored picks, no paid recommendations, and no undisclosed commercial relationships with any product or manufacturer that appears in a ranking. When commerce and the model disagree, the model wins. That is the only posture under which a ranking can be a ranking and not an advertisement.
The honest limits.
- We don’t accept paid placement, sponsored picks, or paid score adjustments.
- We don’t manually adjust scores to favour any product or brand.
- We don’t claim a product was hands-on tested without a verified per-item signal.
- We don’t republish rankings without recomputing the composite from current data.
- We don’t tier products into editorial labels the model didn’t produce.
- We don’t carry undisclosed commercial relationships with any product or manufacturer in a ranking.
Rankings move when the data does.
Composite scores recalculate whenever new market or review data arrives. The Updateddate on a ranking reflects the latest recalculation, not a fixed publication date. Order changes between visits are signals - not errors. A ranking that hasn't moved isn't stale; it's confirmed.
Frequently asked questions.
How does theRANKS decide the order?
A single composite score, computed from five independent dimensions - demand, trust, value, quality, durability - applied identically to every product in a category. The product with the highest composite ranks #1.
Does theRANKS get paid for these rankings?
theRANKS earns affiliate commissions from qualifying purchases through links on rankings pages. Affiliate relationships have zero influence on dimension scores or ranking order - the scoring model receives no revenue signal, and the ranking would be identical if no affiliate programme existed.
How often do rankings update?
Composite scores recalculate whenever new market or review data arrives. The "Updated" date on a ranking reflects the latest recalculation, not a fixed publication date. Order changes between visits are signals, not errors.
Why a composite - why not just a single metric?
No single number tells the whole story. A product can dominate quality and collapse on value; another can ace durability and underperform on demand. The composite forces a product to earn its position on more than one front before it can rank.
What if I disagree with a ranking?
Send us the category and the product. We re-pull the data, recompute the composite, and either the ranking holds or it moves - based on the data, not the appeal. The scoring model is the appeal mechanism.
Read a ranking next.
The methodology is the receipt. The rankings are where it lands.
// Our Editorial Team
The five-dimension framework is applied by humans, not a model.
theRANKS is an editorial team operation. Every ranking and insight on the site is assigned to a named senior editor or category editor who has signed off on the methodology application, fact-checked the citations, and approved the piece for publication. Editors are matched to articles by category expertise - electronics goes to the electronics editor, kitchen to the kitchen editor, and so on - so the review is informed by genuine category fluency. The full roster is below; bios and category coverage live on the Editorial Team page.
Kyle Cunningham
Senior Editor, Consumer Electronics
Read bio →Louie Cepeda
Editor, Outdoor & Sports
Read bio →Nicole Russo
Editor, Health & Wellness, Beauty
Read bio →Tammy Cunningham
Senior Editor, Home & Kitchen
Read bio →Nicholas Maher
Editor, Adult & Intimate Wellness
Read bio →
// Our Review Process
How a ranking gets published.
- Category selection. Editor identifies the buyer-stage question and the candidate product set worth evaluating.
- Five-dimension scoring. Editor researches market data, structured review signal, pricing history, and product specifications. Applies the demand, trust, value, quality, and durability scoring framework to each candidate.
- Hands-on review where applicable. For products the editor has personally used, a hands-on tested signal is set on the article. We never claim hands-on testing where we haven't done it.
- Editorial sign-off. Reviewer signs off on methodology adherence, factual citations, and category fit. Sign-off is timestamped to the article and emitted as Schema.org Article.reviewedBy on the published page.
- Publication. The ranking goes live with the editor's byline, the score breakdown, and the full reasoning visible to readers.
- Scheduled refresh. Rankings are refreshed against current market data on scheduled cycles so scores stay current. Products that are discontinued are rotated out; new product launches trigger fresh analysis.