Real catalog example · Apple Vision Pro

How BettaScore works

The evidence

A 68-source catalog snapshot for Apple Vision Pro — these are real excerpts from Hacker News, Reddit, YouTube, and Apple's specs.

68-source catalog snapshot for Apple Vision Pro — real threads, videos, and specs. The 72-point totals are a fixed methodology walkthrough.

AI sorts it

AI reads each one and files it under a product dimension, marked positive, negative, or neutral. That is all it decides — the tallies count every source, not just the cards shown.

AI files each one under a dimension as positive, negative, or neutral. That is all it decides.

Not enough evidence to score yet

With only two qualified sources, Software ecosystem gets BettaScore's product-page Sparse badge and no numeric score. Sparse means evidence coverage is incomplete — not that the product performed badly.

Only two qualified sources: BettaScore shows Sparse and withholds the number. Sparse means incomplete evidence, not a bad product.

A fixed formula scores it

The scoring is arithmetic, not judgment: the same formula turns each column's tally into its score. Neutral evidence stays as context.

positive ÷ (positive + negative) × 100. Neutral stays as context.

Dimensions combine

Every scored dimension — the three above plus passthrough and audio — averages into one public score, weighted by source depth (capped at w3). The arithmetic is printed on the card.

One public score, each dimension's source-depth weight capped at 3.

Every number traces back

Open any dimension and you land on the source the claim came from — the score is a reference you can check, not a verdict.

Tap a dimension and you land on the source the claim came from.

And you can re-weight it

Care more about comfort than price? Soon you can weight that dimension and Your Score recalculates from the same evidence. The public score never moves.

Display quality0+ · 0− · 34 src
Comfort0+ · 0− · 27 src
Price friction0+ · 0− · 29 src
Hacker NewsDisplay+

The displays in this device are crazy.The displays in this device are crazy.

Hacker News+

Vision Pro provides a pleasing and enjoyable display experience with album art.Vision Pro provides a pleasing and enjoyable display experience with album art.

Hacker NewsDisplay

Passthrough has motion blur, pixelation, distortions, limited color and dynamic range.Passthrough has motion blur and distortion.

Spec pageDisplay · neutral

23 million pixels · 90 / 96 / 100 / 120 Hz23 million pixels · 90 / 96 / 100 / 120 Hz

Hacker NewsComfort

The Vision Pro is heavy and needs to be lighter.Vision Pro is heavy.

Reddit

Vision Pro is uncomfortable due to weight and pressure on face.Vision Pro is uncomfortable due to weight and pressure on face.

YouTube+

The Vision Pro is more comfortable due to better weight support.The Vision Pro is more comfortable due to better weight support.

YouTube

The original Vision Pro feels heavy after just 10 to 15 minutes for a lot of people.The original Vision Pro feels heavy after just 10 to 15 minutes for a lot of people.

Hacker NewsPrice

$3,500 is too expensive for what it offers.$3,500 is too expensive for what it offers.

Reddit

Vision Pro lacks sufficient value for its $3,500 price.Vision Pro lacks sufficient value for its $3,500 price.

Hacker NewsPrice+

They've absolutely blown Meta out of the water on their value proposition.Strong value versus Meta.

YouTube+

The Vision Pro is now a worth-getting item after price drop.The Vision Pro is now a worth-getting item after price drop.

Dimension statusSoftware ecosystem
Sparse

2 qualified sources · below publishing threshold

Numeric score withheld until evidence coverage is strong enough.

12 real excerpts are shown from the catalog. The 72-point totals are a fixed methodology walkthrough.

Display:29 pos3 neg2 neutral
One fixed formula · positive ÷ (positive + negative) × 100
Display quality

29 ÷ (29 + 3) × 100 = 91

Comfort

12 ÷ (12 + 13) × 100 = 48

Price friction

9 ÷ (9 + 20) × 100 = 31

Neutral evidence never enters the maths. Same evidence in, same numbers out — every time.

Display quality

29 ÷ (29 + 3) × 100 = 91

Same formula for every dimension — comfort 48, price friction 31.
Public scoreApple Vision Pro
0%
Display quality91 · w3
Passthrough92 · w2
Audio89 · w2
Comfort ×348 · w2
Price friction31 · w2
Software ecosystemSparse

(91×3 + 92×2 + 89×2 + 48×2 + 31×2) ÷ 11 = 72.1

Your Score Coming soon

Comfort weighted ×3

72%↓ −6%

Same evidence, your priorities. The public score stays 72% for everyone.

Source of this claim
Scroll

BettaScore Methodology

How BettaScore works

BettaScore turns public product signals into concise, evidence-backed review scores for emerging technology.

Public-source research

BettaScore reads publicly available product pages, reviews, videos, community discussions, and media coverage. The output is a source-aware summary designed to show what people repeatedly notice about a product.

Sources are grouped by product and evidence dimension, then reduced to recurring signals instead of one-off comments. Strong repeated claims can raise confidence; thin or conflicting coverage keeps the public read more cautious.

Example: from signals to a product page

For a spatial-computing product such as Vision Pro, BettaScore looks for repeated public signals around display quality, passthrough, audio, comfort, controls, software ecosystem, price friction, and everyday use cases. The product page then distills those signals into a visible public rating, summary, evidence dimensions, and source context on one page.

How dimension scores are calculated

Every scored dimension uses one fixed formula: positive ÷ (positive + negative) × 100. The same evidence always produces the same dimension score.

Neutral evidence remains visible as useful context, but it does not directly raise or lower the score. Dimensions without enough valid scored evidence are labelled Sparse and do not contribute a numeric value to the product total.

Score dimensions

Product scores are organized around dimensions such as usability, capability, reliability, design, and market fit. Each product page keeps the visible score tied to the public evidence available for that product.

Source and update policy

BettaScore favors public sources that can be revisited, attributed, and grouped by product evidence. Product pages can change when the public evidence changes, when a new import adds stronger source coverage, or when a previous product interpretation needs correction.

Independence

BettaScore ratings are not merchant offers, affiliate prices, checkout availability, or raw user aggregate ratings. They are editorial-style public research signals intended to help readers decide what to inspect next.

Limits and interpretation

BettaScore should be read as an editorial-style public research score, not as a checkout offer, merchant rating, professional advice, or user aggregate rating.

Product pages can change as public evidence changes. When evaluating a purchase, use BettaScore alongside official specs, current pricing, retailer policies, and your own requirements.

Where to start

Browse the Discover catalog or read more about BettaScore to understand how the platform fits product research workflows.