What does the ApexGEO score measure?

The ApexGEO score is a weighted-sum measure of brand visibility across AI search engines, combining citation count, share of voice, sentiment, source authority, and content freshness into a 0–100 figure.

Are the weights here the same as production?

No. The weights on /what-is-apexgeo are illustrative and stable across changes to the real engine. Production scoring adapts per-industry and per-engine.

Which engines does ApexGEO sweep?

Core engines: ChatGPT, Claude, Gemini, Perplexity, Grok, DeepSeek. Expanding to 20+ including NotebookLM, Copilot, Cohere, Mistral, Llama, Qwen.

What does a score of 60 mean?

A score of 60 in the illustrative model means the brand averages strong on freshness and sentiment, moderate on citation count, and below-average on share of voice and source authority — typical for an early-stage brand with credible content but limited backlink coverage.

How often does the score update?

Tracked prompts run daily on tier-1 plans and weekly on the standard plan. The score recomputes whenever new prompt-tracking results land.

Can the score be gamed?

Source authority and share of voice are slow-moving and hard to fake. Citation count and freshness can be improved through legitimate publishing and indexing work, which is the goal.

Methodology · Last reviewed 2026-05-14

The methodology behind the ApexGEO score

ApexGEO measures brand visibility across AI search engines using a five-input weighted-sum model. Move the sliders below to see how each input shapes the score.

What ApexGEO measures

ApexGEO sweeps six core AI engines — ChatGPT, Claude, Gemini, Perplexity, Grok, DeepSeek — with category-relevant prompts and records every brand citation. The resulting input vector feeds five axes: citation count, share of voice, sentiment, source authority, freshness.

ApexGEO score (illustrative)

Citation count35% × 64
How often AI engines cite the brand in responses.
Share of voice25% × 48
How often the brand is cited relative to competitors.
Sentiment15% × 72
Whether citations describe the brand positively.
Source authority15% × 41
Whether AI engines treat the brand's site as a trusted source.
Freshness10% × 87
How recent the brand's tracked content is.

Illustrative weighting — production scoring uses a richer model that adapts per-industry and per-engine.

How the score is computed

Each input is mapped to a 0–100 scale. The score is a weighted sum where the weights sum to 1. Production weights adapt per-industry and per-engine; the illustrative model uses fixed weights so the page teaches the concept without coupling to the live algorithm.

Why each input matters

Citation count — raw discoverability.
Share of voice — category dominance.
Sentiment — citation quality.
Source authority — third-party trust signals.
Freshness — recency.

How the score reacts to changes

In the illustrative model, lifting source authority from 41 to 80 with default weights moves the score from 60 to 66. Lifting citation count from 64 to 90 moves the score from 60 to 69. The slider behavior above demonstrates how the score responds when you reweight inputs.

What the score is not

The ApexGEO score is not a guarantee of citation in any specific AI engine response. It is not a substitute for traditional SEO ranking metrics. It does not measure conversion or revenue attribution. It captures visibility, not outcome.

How to improve it

ApexGEO ships a recommendation engine that ranks fixes by expected score impact. See the ApexGEO self-audit case study for a worked example.