First-party platform data, live responses, crawl evidence and visible page content.
Evidence before recommendations
AI visibility without the theatre.
A practical AI visibility research and delivery standard grounded in observed answers, cited sources, entity clarity, accessible evidence and honest measurement.
Research standard
Separate facts, estimates and sampled signals.
Third-party traffic, keyword and authority models labelled as estimates.
Prompt tracking and AI answers that vary by model, phrasing, location and time.
Audit reports state the domain, date, scope, sources, crawl or sample size, assumptions, blocked checks and missing first-party data.
Evidence hierarchy
Use the strongest available signal.
First-party
GSC, GA4, Bing, CRM and conversion data.
Direct inspection
Crawls, response tests, rendered output and index checks.
Public authority
Earned mentions, citations, profiles and source quality.
Platform estimates
Modeled keywords, links, traffic and visibility.
Prompt samples
Directional monitoring across supported AI surfaces.
Prioritization
Fix what unlocks the next layer.
- Eligibility: crawlability, indexation, rendering and canonical consistency
- Relevance: page intent, topic coverage, answer clarity and internal discovery
- Trust: entity facts, credentials, visible evidence and review standards
- Authority: source-worthy assets, reclamation, relationships and earned coverage
- Conversion: analytics, events, assisted journeys and qualified outcomes
Priorities consider expected impact, confidence, effort, dependencies and reversibility.
Measurement model
Report the signal and its limitation together.
Impressions, clicks, CTR, indexed coverage, keyword cohorts and Core Web Vitals.
Primary: Google Search Console, Bing Webmaster Tools, crawl dataMentions, citations, cited pages, share of voice, sentiment and competitor inclusion.
Sampled and directional unless a first-party platform report is availableEarned mentions, referring sources, reclaimed links, citation types and source relevance.
Opportunity counts are not placement guaranteesOrganic and AI referral sessions, assisted conversions, goal completions and landing-page performance.
Checkout clicks remain separate from confirmed purchasesWhere data exists, each subscription starts with a 28-day baseline, then reports absolute results, month-over-month movement, work shipped, limitations and next priorities.
Crawler controls
Search inclusion and model training are different permissions.
We check Googlebot, Bingbot, OAI-SearchBot and PerplexityBot access for public discovery. GPTBot training controls are assessed separately from OpenAI search inclusion.
Access guardrails
- Public content is crawlable and indexable intentionally
- Robots rules match the chosen publisher policy
- Canonical and sitemap signals agree
- Structured data matches visible content
- No special “GEO schema” is invented
- llms.txt is not sold as a visibility requirement
Ethical authority
Earned relevance over manufactured links.
Authority work can include credential cleanup, relevant profiles, link reclamation, expert contributions, original research, useful resources, partnerships and digital PR.
- No guaranteed backlink packages
- No automated link schemes
- No fabricated authors or credentials
- Sponsored links receive appropriate qualification
- Publisher independence is disclosed
Platform limitations
What we will never imply.
Answers vary across engines, models, accounts, locations, prompts and time.
A readiness score prioritizes work; it does not guarantee revenue or citations.
We distinguish observed conversions, assisted journeys and unverified influence.
AEO methodology FAQ
What the system does—and refuses to pretend.
The methodology follows durable search foundations, visible evidence and explicit measurement limits.
What does answer-engine optimisation change?
AEO improves how clearly AI systems can retrieve, interpret and support brand facts, comparisons and claims. It still depends on useful, accessible pages and credible source evidence.
What does GEO add?
GEO focuses on how generative systems understand, mention or cite a brand through source-quality content, entity consistency and corroborating authority—not a separate technical shortcut.
Do you add llms.txt or special AI schema?
No tactic is marketed as a visibility shortcut. Structured data describes visible content, and crawler controls are configured only for a clear evidence-access purpose.
How do you handle OAI-SearchBot and GPTBot?
Search inclusion controls and model-training controls are treated separately. Access decisions are verified against the client’s intended publishing policy.
Why is prompt tracking directional?
Answers vary by engine, model, location, time, context and phrasing. Samples help track trends but cannot represent every possible user response.
How do you prioritize fixes?
Observable evidence, commercial impact, confidence, implementation effort and dependencies determine the queue—not issue-count volume.
What makes authority work ethical?
Credential alignment, reclamation, useful contributions and qualified outreach are allowed. Paid link schemes and guaranteed placements are not.
Can the composite score predict revenue?
No. It organizes readiness and priorities. First-party traffic, conversions and revenue are reported separately from sampled visibility indicators.
See the method applied