ASTONTO
INDEPENDENT AI RESEARCH

We study how AI systems interpret, source and recommend.

ASTONTO investigates the observable behaviour of large language models and applies that research through evidence-based AI Search Visibility audits, optimisation sprints and monitoring.

Book your free visibility checkExplore AI Search Visibility
Framework: PULSE v1.0•Controlled Black-Box Testing
Fig 1.1 — Observation FieldPULSE v1.0
SOURCE
ENTITY
COMPARISON
CONTEXT
Evaluated Output SignalReliability: High

"Entity A cited in position #1 as primary recommended vendor across 5 repeated test iterations."

Research Scope

Four Primary Observational Dimensions

01
Information Interpretation
How models parse complex buyer queries, implicit intent, and structured entity attributes.
02
Source Selection
Which third-party web domains, trade media, and directories are retrieved, cited, and trusted.
03
Recommendation Behaviour
How corporate alternatives are compared, ranked, and endorsed across buyer evaluation scenarios.
04
Variation Over Time
Observing changes in output consistency across platforms, IP geolocations, languages, and model updates.
Applied Commercial Service

Be the business AI recommends.

ASTONTO measures how ChatGPT, Perplexity, Gemini and Google AI Overviews recommend your company compared with four selected competitors—and develops an evidence-based plan to strengthen your AI Search Visibility.

Methodology Notice: The PULSE Score reflects observed performance during the documented testing period. AI-generated answers change, and future visibility cannot be guaranteed.
Entry Audit Offer

24-Hour AI Visibility Audit

£950 + VAT

Understand how ChatGPT, Perplexity, Gemini and Google AI Overviews recommend your business compared with four competitors.

  • •Commercially relevant buyer-question testing
  • •Four AI platforms (ChatGPT, Perplexity, Gemini, AI Overviews)
  • •Your company and four competitors
  • •PULSE Benchmark and Platform Scores
  • •Appearance, recommendation strength and sentiment
The £950 audit fee is credited in full when you begin a 90-Day Optimisation Sprint within 72 hours of the audit presentation.
90-Day Sprint

90-Day AI Visibility Optimisation Sprint

£6,000 + VAT (or three monthly payments of £2,000 + VAT)

Turn the audit findings into a coordinated programme of technical, content, authority and monitoring improvements.

Discuss a 90-Day Sprint →
Monthly Tracking

AI Visibility Monitoring

From £350 + VAT per month

Ongoing monthly tracking of priority prompts, competitor movements, and AI model update impacts.

Enquire about Monitoring →
Empirical Record Standards

Observational Evidence, Logged Verbatim

Every evaluation generated by ASTONTO relies on controlled multi-run testing. We preserve verbatim output logs, citation URLs, and position orders to ensure complete auditability.

• 100% Verbatim Output Capture

• Multi-Platform Run Logs

• Non-Deterministic Variance Tracking

Sample Audit Record #LOG-2026-0842Reliability: High
Platform:ChatGPT-4o (OpenAI)
Methodology Version:PULSE Method v1.0
Location / Language:London, UK / en-GB
Test Run Iterations:5 Repeated Runs
Exact Evaluated Prompt:

"Which commercial managed IT service providers in London specialize in financial services security?"

Recommendation Class:Strong Endorsement
Sentiment Factor:Positive (1.00)
FAQ Preview

Frequently Asked Questions

What is ASTONTO?+
ASTONTO is an independent AI research company that studies the observable behaviour of large language models and AI recommendation systems.
What does ASTONTO research?+
ASTONTO researches how AI platforms interpret corporate information, resolve entity identity, select citation sources, compare alternatives, and generate brand recommendations across platforms, locations, and languages.
Is ASTONTO an AI Search Visibility agency?+
ASTONTO is an independent research company. AnswerSignal is our applied commercial service that provides empirical AI Search Visibility audits and optimisation sprints for corporate clients.
What is black-box AI evaluation?+
Black-box AI evaluation is the scientific observation of publicly accessible AI outputs under controlled prompt conditions, measuring real-world model responses without claiming internal access to proprietary code or weights.
Does ASTONTO have access to AI model source code or weights?+
No. ASTONTO evaluates observable outputs from public and enterprise interfaces. We do not claim access to proprietary model weights, hidden ranking algorithms, or undisclosed vendor update schedules.
How does ASTONTO preserve research independence?+
ASTONTO enforces strict methodology controls, preserves verbatim output logs, and refuses commercial arrangements that mandate specific research findings or ranking outcomes.
View All 30 FAQs →

See how AI recommendations evaluate your business.

Book a 20-Minute AI Visibility Check or full 24-Hour Audit for your company.

Book your AI Visibility Audit