Methodology

Published in full — a score nobody can check is a score nobody should trust

Scoring v1.0.0 · frozen
AI answers
1,386,091
Brand mentions
4,361,554
Raw responses
76,862
Entities
346,703

What one observation is

PROMPT"Beste 55-Zoll TVs?"ENGINEPerplexityONE ANSWER1. LG C52. Samsung S90F3. Sony Bravia 8= 1 observationLGrank 1Samsungrank 2Sonyrank 3

An answer naming a competitor but not the tracked brand still counts in the denominator — being absent is a result. An answer naming no brand at all is excluded, because nobody lost it.

Two data sources

Search surfaces

default
4 engines1.39M answers
ChatGPT SearchGeminiGoogle AI OverviewsGoogle Search
Brand order
From the export
Demand weighting
Available

Named AI models

18 engines77K answers
GPT-5PerplexityGemini 2.5 FlashClaude Sonnet 4CopilotDeepSeekGrok+11 more
Brand order
Extracted by us
Demand weighting
Not available

Visibility score

The headline number

ReachHow often you appear^0.4ProminenceHow high up^0.3QualityDescribed correctly^0.3×GEOMETRIC MEAN( … )^1 × 100VISIBILITY SCORE0–100Multiplied, not averaged — a zero on any dimension collapses the score
Formula
Reach      = answers naming the brand
             ÷ all answers in scope
Prominence = mean attention, where present
Quality    = mean quality gate, where present

Score = 100 × ( R^0.4 × P^0.3 × Q^0.3 )^1

Quality gate

One verifiable sub-gate today: whether the brand's own domain was among the sources the model used. 1.0 if yes, 0.6 if no. Spec correctness and market availability need retailer and catalogue data this export does not carry.

λ = 1

Presentation only — mathematically cannot change ranking order. Published and frozen because it is the most gameable constant here.

Attention by rank

1 / log₂(rank + 1) — the standard information-retrieval discount, borrowed not invented

0.000.250.500.751.001.0010.6320.5030.4340.3950.3660.3370.328position in the answer

Competitive rating

Every answer naming two or more brands becomes a set of matches

LGSamsungSonyHisense4 brands · 6 matchesLG beats SamsungLG beats SonyLG beats HisenseSamsung beats SonySamsung beats HisenseSony beats Hisense
Bradley–Terry
P(a beats b) = e^θa / (e^θa + e^θb)

θ fitted by MM iteration to reproduce
the observed record, normalised Σθ = 0.

Strength = 1500 + 400θ / ln(10)

No chosen weights

Nothing here is tuned — the rating is whatever best explains the match record. That is why the fit check is published beside it on the Competitive page.

Half-win prior

Each observed pair carries a half-win each way. Without it a brand that never loses has no finite rating and the fit silently degenerates.

Confidence

Bootstrapped over 300 resamples, seeded so results are reproducible

Clear leadReported as a lead
OverlappingReported as tied

Two brands whose intervals overlap are not meaningfully different, and the overview says so in words rather than presenting a ranking the data does not support.

Suppression thresholds

A dash is more honest than a precise-looking number built on nothing

Visibility score
100minimum observations

Below this the score shows — instead of a number.

Competitive rating
300minimum head-to-head matches

Below this the score shows — instead of a number.

Known limitations

Quality is half-built

Two of four intended gates — spec correctness and market availability — need retailer and catalogue data not in these exports.

No time dimension

The primary export carries no date, so this is a snapshot rather than a trend.

Rank inferred on the model source

Reliable for ranked lists, rougher for flowing prose.

Short brands need an allowlist

LG, HP, 3M and GE are two characters. A length filter silently deleted them during development, so the allowlist is explicit and visible.

Entity list is not exhaustive

Retailers and marketplaces are separated from brands by an explicit list — auditable, but incomplete.

Not externally validated

Reproducible and defensible. Whether the scores predict commercial outcomes is still open.

Ingested 2026-08-11T06:07:31Z · /sessions/zealous-beautiful-gates/mnt/categories