AigeoRadar · AI Context Benchmark

AI Context Benchmark

Question → sufficient layer(s) → retrieved slice → measured cost → evidence

The Best GEO WordPress Plugin in 2026

https://aigeoradar.com/blog/the-best-geo-wordpress-plugin-in-2026-aigeoradar-connector-explained-simply

5 questions Answer LLM: gpt-4o-mini ai_context_benchmark_v1

Of 5 questions: lowest measured retrieval cost among sufficient layers — HTML 0 · Schema 1 · AIPM 4 · multi-layer 0 · unanswered 0. Descriptive counts only; no overall ranking.

Of 5 questions: lowest measured retrieval cost among sufficient layers — HTML 0 · Schema 1 · AIPM 4 · multi-layer 0 · unanswered 0. Descriptive counts only; no overall ranking. Each question shows: sufficient layer(s) → retrieved slice → measured input tokens → evidence. Readers interpret; the report does not rank formats. Semantic Redundancy 28% — moderate field overlap; trim duplicated purpose/summary/keyFacts wording.

Gold mode · Mixed independent + AIPM consistency

Most Understanding/Retrieval gold comes from HTML. Some questions remain AIPM self-consistency checks (tagged) and are excluded from Understanding when possible. Matching a machine card against its own fields is not evidence that that layer outperforms HTML.

Execution provenance

Methodology: AI Context Benchmark Methodology v1.0
Planner Spec: v1.0
Execution Protocol: v1.0
Question pack: universal_v1
Score version: aipm_benchmark_score_v10
Engine: aipm_benchmark_v4
Answer LLM: openai / gpt-4o-mini (single provider this lab)
Model snapshot: 2026-07
Run id: fa2439f1-5668-49b5-a719-9435aab5a649

Question outcomes

Per question: which layers were sufficient, and what was the lowest measured retrieval cost among them. No overall winner.

Lowest cost: HTML

Lowest cost: Schema

Lowest cost: AIPM

Multi-layer

Unanswered

Manifest design

Semantic Redundancy · 28%

Semantic Redundancy 28% — moderate field overlap; trim duplicated purpose/summary/keyFacts wording.

→Reduce repeated wording across purpose, abstract, keyFacts, and sections.

Field A	Field B	Overlap
`purpose`	`abstract`	28%

Structural size (secondary)

Structural size is secondary. A larger AIPM is not a failure if per-question retrieval stays tiny — check Semantic Redundancy instead.

HTML · 8,842 chars

Schema · 15,457 chars

AIPM · 12,386 chars

Routing helper (secondary)

Illustrative card-first vs HTML-always simulation — prefer per-question measured cost above. Not a ranking.

Card-first retrieval would cost ~8% more tokens than HTML-always on this pack — machine card is heavier here.

Orientation pack

5 questions · all layers scored

HTML 1/5
Schema 2/5
AIPM 5/5

Depth pack

0 questions · all layers scored

HTML 0/0
Schema 0/0
AIPM 0/0

Full-context pack metrics (secondary)

These measure the whole file fed to the model this run — not the minimum slice needed per question.

HTML

1/5 matched

21,222 full-pack tokens

Schema

3/5 matched

29,318 full-pack tokens

AIPM

5/5 matched

32,168 full-pack tokens

Full-pack resource table (secondary)

Whole-file context fed this run. Prefer Minimal Retrieval Cost on each question.

Metric	HTML	Schema	AIPM
Coverage (matched)	1/5	3/5	5/5
Context size	8,842 chars	15,457 chars	12,386 chars
Total tokens	21,222	29,318	32,168
Tokens / correct answer	21,222	9,773	6,434
Est. cost / correct answer	$0.003251	$0.001479	$0.000987
Matched per 1k tokens	0.047	0.102	0.155
Median latency	913 ms	654 ms	715 ms
Est. cost (USD)	$0.00325	$0.00444	$0.00494

Six independent scores

Answer Efficiency is the primary cost lens. Accuracy axes remain for research — no combined total or winner.

Answer Efficiency

Matched answers per 1k tokens (and cost per match). The primary efficiency axis — not raw accuracy.

HTML 30.3
Schema 65.8
AIPM 100

Understanding

Can this layer convey what the page is about — using independent HTML gold?

HTML 100
Schema 100
AIPM 100

Retrieval

Can this layer surface shared facts (location, contact, hours, pricing, FAQ, CTA)?

HTML —
Schema —
AIPM —

Evidence

Answer quality vs independent gold (score strength). Partial credit counts; UNKNOWN scores zero unless gold is UNKNOWN.

HTML 44.7
Schema 60.4
AIPM 91.1

Metadata

Language, page kind, and freshness from page signals.

HTML 0
Schema 50
AIPM 100

Compression

Information delivered per token and context size. Higher means more matched answers for less context cost.

HTML 30.3
Schema 65.8
AIPM 100

Coverage map

AIPM matched 5 question(s) (alone on Q2, Q3, Q4); HTML matched 1. Read this as complementary coverage — AIPM orients agents cheaply; HTML supplies depth when the sidecar cannot. HTML 1/5 matched (≈21,222 tok/match). Schema 3/5 matched (≈9,773 tok/match). AIPM 5/5 matched (≈6,434 tok/match). Figures are descriptive per layer — not a ranking. See per-question chains for sufficient layers, slice size, and measured cost.

AIPM matched

Q1, Q2, Q3, Q4, Q5

Alone: Q2, Q3, Q4

AIPM insufficient

—

HTML/Schema needed or all layers missed

Unanswered by all

—

HTML layer

Visible page text after stripping AIPM sidecars and JSON-LD. Measures what prose alone can answer.

Context fed: 8,842 chars

Tokens: 21,222 · median 913 ms

Matched this pack: 1/5

Stronger on

Understanding (100)

Weaker on

Metadata (0) · Compression (30.3) · Answer Efficiency (30.3)

Schema layer

JSON-LD structured data with minimal page chrome. Measures what schema markup can answer.

Context fed: 15,457 chars

Tokens: 29,318 · median 654 ms

Matched this pack: 3/5

Stronger on

Understanding (100)

Weaker on

—

AIPM layer

AI Page Manifest (.ai.json) only. Measures what the machine layer can answer without HTML.

Context fed: 12,386 chars

Tokens: 32,168 · median 715 ms

Matched this pack: 5/5

Stronger on

Understanding (100) · Metadata (100) · Evidence (91.1) · Compression (100) · Answer Efficiency (100)

Weaker on

—

When to use which layer

AIPM complements HTML — it does not replace full-page prose.

Scenario	Recommended	Why
Fast orientation (title, purpose, brand, intent)	Compare machine card → HTML fallback on this run	Machine-card pack: 32,168 tok · $0.00494. Machine-card pack is heavier than HTML on this run — densify before relying on card-first retrieval.
Deep content / research (prose facts, process detail)	HTML (with optional machine orientation)	HTML pack: 21,222 tok · $0.00325. Use when depth needs body prose.
Structured entity pulls (org, location, typed fields)	Schema.org	Schema pack: 29,318 tok · $0.00444. Dense JSON-LD tends to score well here.

Findings

—Of 5 questions: lowest measured retrieval cost among sufficient layers — HTML 0 · Schema 1 · AIPM 4 · multi-layer 0 · unanswered 0. Descriptive counts only; no overall ranking.
—Semantic Redundancy 28% — moderate field overlap; trim duplicated purpose/summary/keyFacts wording.
—Manifest design: Reduce repeated wording across purpose, abstract, keyFacts, and sections.
—Structural size is secondary. A larger AIPM is not a failure if per-question retrieval stays tiny — check Semantic Redundancy instead.
—AIPM matched 5 question(s) (alone on Q2, Q3, Q4); HTML matched 1. Read this as complementary coverage — AIPM orients agents cheaply; HTML supplies depth when the sidecar cannot. HTML 1/5 matched (≈21,222 tok/match). Schema 3/5 matched (≈9,773 tok/match). AIPM 5/5 matched (≈6,434 tok/match). Figures are descriptive per layer — not a ranking. See per-question chains for sufficient layers, slice size, and measured cost.

Question-by-question layer analysis

Which layer(s) could answer; which need more or different context; minimum context fed this run.

Q1 · Understanding · orientation

What is the page title?

Gold: Best GEO WordPress Plugin 2026: AigeoRadar Connector — AigeoRadar

Gold source: title · html_independent

Sufficient: HTML, Schema, AIPM

Estimated retrieval cost · lowest cost SCHEMA · 3 input tokens

Layer	Planner fields	Input tokens	Rounds	Confidence	OK?
HTML	`html.title`	17 est	—	—	yes
Schema	`jsonld.name`	3 est	—	—	yes
AIPM	`title`	13 est	—	—	yes

HTML

sufficient

The page title is "Best GEO WordPress Plugin 2026: AigeoRadar Connector — AigeoRadar."

score 93 · 8,842 chars context · 2,104 in-tokens

HTML answered using 8842 chars of layer context (minimum fed this run).

Schema

sufficient

The page title is "Best GEO WordPress Plugin 2026: AigeoRadar Connector — AigeoRadar."

score 93 · 15,457 chars context · 2,920 in-tokens

Schema answered using 15457 chars of layer context (minimum fed this run).

AIPM

sufficient

The page title is "Best GEO WordPress Plugin 2026: AigeoRadar Connector".

score 93 · 12,386 chars context · 3,189 in-tokens

AIPM answered using 12386 chars of layer context (minimum fed this run).

Q2 · Understanding · orientation

List the main keywords for this page.

Gold: AI visibility, GEO, generative engine optimization, ChatGPT, llms.txt

Gold source: keywords · unknown

Sufficient: AIPM · Partial: HTML · Needs more / other context: Schema

Estimated retrieval cost

Layer	Planner fields	Input tokens	Rounds	Confidence	OK?
HTML	`html.lead`	— est	—	—	no
Schema	No typed Schema field for this question.	— est	—	—	no
AIPM	No minimal AIPM slice identified for this question.	— est	—	—	yes

HTML

partial

AigeoRadar, WordPress plugin, GEO, AI, llms.txt, entity.json, AI crawler, discovery package, AIPM, SEO, machine-readable signals.

score 41 · 8,842 chars context · 2,106 in-tokens

HTML was only partially aligned with gold from 8842 chars of context.

Schema

insufficient

UNKNOWN

score 0 · 15,457 chars context · 2,922 in-tokens

Schema did not answer from 15457 chars of context — additional or different layer context needed.

AIPM

sufficient

The main keywords for this page are "AI visibility," "GEO," "generative engine optimization," "ChatGPT," and "llms.txt."

score 92 · 12,386 chars context · 3,191 in-tokens

AIPM answered using 12386 chars of layer context (minimum fed this run).

Q3 · Understanding · orientation

Who is the intended audience declared in AIPM?

Gold: B2B, agency, SMB

Gold source: audience · aipm_consistency

Sufficient: AIPM

Estimated retrieval cost · lowest cost AIPM · 6 input tokens

Layer	Planner fields	Input tokens	Rounds	Confidence	OK?
HTML	`html.lead`	— est	—	—	no
Schema	No typed Schema field for this question.	— est	—	—	no
AIPM	`audience`	6 est	—	—	yes

HTML

skipped

The intended audience is WordPress site owners, agencies managing client GEO, and founders wanting facts stated in files AI can quote.

score 0 · 0 chars context · 0 in-tokens

Skipped (AIPM-only).

Schema

skipped

UNKNOWN.

score 0 · 0 chars context · 0 in-tokens

Skipped (AIPM-only).

AIPM

sufficient

The intended audience includes B2B, agency, and SMB.

score 86 · 12,386 chars context · 3,189 in-tokens

AIPM answered using 12386 chars of layer context (minimum fed this run).

Q4 · Metadata · orientation

What page type is declared in the machine layer?

Gold: blog

Gold source: pageType · aipm_consistency

Sufficient: AIPM

Estimated retrieval cost · lowest cost AIPM · 1 input tokens

Layer	Planner fields	Input tokens	Rounds	Confidence	OK?
HTML	`html.lead`	— est	—	—	no
Schema	`jsonld.@type`	— est	—	—	no
AIPM	`pageType`	1 est	—	—	yes

HTML

skipped

This is a product page for the AigeoRadar Connector, a WordPress GEO plugin.

score 0 · 0 chars context · 0 in-tokens

Skipped (AIPM-only).

Schema

skipped

The page type is a BlogPosting.

score 0 · 0 chars context · 0 in-tokens

Skipped (AIPM-only).

AIPM

sufficient

The page type is blog.

score 88 · 12,386 chars context · 3,189 in-tokens

AIPM answered using 12386 chars of layer context (minimum fed this run).

Q5 · Metadata · orientation

What language is this page in?

Gold: en

Gold source: inLanguage · html_independent

Sufficient: Schema, AIPM · Needs more / other context: HTML

Estimated retrieval cost · lowest cost AIPM · 1 input tokens

Layer	Planner fields	Input tokens	Rounds	Confidence	OK?
HTML	`html.lang`	— est	—	—	no
Schema	`jsonld.inLanguage`	1 est	—	—	yes
AIPM	`inLanguage`	1 est	—	—	yes

HTML

insufficient

UNKNOWN.

score 0 · 8,842 chars context · 2,105 in-tokens

HTML did not answer from 8842 chars of context — additional or different layer context needed.

Schema

sufficient

The page is in English.

score 88 · 15,457 chars context · 2,921 in-tokens

Schema answered using 15457 chars of layer context (minimum fed this run).

AIPM

sufficient

The page is in English.

score 88 · 12,386 chars context · 3,190 in-tokens

AIPM answered using 12386 chars of layer context (minimum fed this run).

Methodology

AI Context Benchmark: Planner → Slice → LLM with measured API input tokens.
Reports describe sufficient layers, slice size, cost, and evidence — they do not declare a winning format.
Layers under test today: HTML, Schema.org, AIPM (extensible to RSS, Markdown, PDF, …).
Engine aipm_benchmark_v1 · Mon, Jul 27, 2026 11:58 PM · aipm_benchmark_score_v10

Per-question context chain across layers.

Post on X LinkedIn WhatsApp

About the lab →

Question outcomes

Semantic Redundancy · 28%

Structural size (secondary)

Routing helper (secondary)

Orientation pack

Depth pack

Full-context pack metrics (secondary)

Full-pack resource table (secondary)

Six independent scores

Answer Efficiency

Understanding

Retrieval

Evidence

Metadata

Compression

Coverage map

HTML layer

Schema layer

AIPM layer

When to use which layer

Findings

Question-by-question layer analysis

What is the page title?

List the main keywords for this page.

Who is the intended audience declared in AIPM?

What page type is declared in the machine layer?

What language is this page in?

Methodology

Share