AI Visibility Intelligence

Every Intelligence engagement produces six deliverables tailored to your business — or five, depending on Google Business Profile eligibility.

GBP eligibility note: Google Business Profiles are available only to businesses that interact with customers in person during designated operating hours. For clients whose business model does not meet these requirements, the GBP Optimization Checklist is omitted and the engagement produces five deliverables. Eligibility is confirmed during the kickoff conversation.

The hybrid process is our AI Visibility methodology run at its fullest depth. It is not a longer Audit; it is a different process built around three reinforcing analytical layers.

Intelligence is not a longer Audit. It is a different process built around three reinforcing analytical layers.

AI Visibility data collected across four engines. ChatGPT, AI Mode, Perplexity, and Gemini. Data is collected over the first month so that patterns become visible that a one-month snapshot cannot detect: movement in competitor citation, freshness signals shifting in your category, and query patterns evolving as the AI engines mature.

A manual Claude assessment in a standardized 30-prompt format. Claude is one of the major AI engines and one of the engines without a continuous-search consumer surface that AI visibility tools can track. We conduct a structured 30-prompt assessment in Claude directly, capturing how Claude reads your business across discovery, evaluation, and gap queries.

Qualitative pattern reading across several hundred answer texts. This is the analytical core. Our pattern library names recurring patterns observed across prior engagements: content-query mismatch, strong content with thin extraction signals, aged content presented as current, address inconsistency, manufacturer credentials not visible, the asymmetric reading of positional versus substantive credentials, and others. Each engagement applies the library to the answer texts your business appears in, and adds new patterns when the analysis reveals something worth carrying forward.

Intelligence reaches five AI engines — the widest coverage of any Telstar service tier.

Tracked through an AI visibility tool: ChatGPT, AI Mode, Perplexity, and Gemini.

Assessed manually in a standardized 30-prompt protocol: Claude.

For businesses where Claude or Perplexity citation matters — particularly business-to-business contexts where decision-makers use Claude for research, or technical contexts where Perplexity is a primary discovery engine — Intelligence is the right tier. The Audit and Snapshot cover four engines each; Intelligence adds Perplexity and a manual Claude assessment that an audit-tier engagement does not reach.

Intelligence includes up to nine competitors across the engagement. The Audit covers four. The deeper set produces strategic findings the project-scoped diagnostic cannot.

You identify your competitive set during the kickoff conversation — businesses you compete with for buyer attention, plus the category leaders the AI visibility tool identifies as dominant AI Visibility presences in your space. We verify coverage before analysis begins.

Each tracked competitor is analyzed across all five engines through the same recommendation framework — summary form for competitors, full analysis for your domain. Findings name where competitors appear that you do not (opportunities), where you appear that they do not (defensible strengths), where AI engines treat the category as a whole (positioning territory), and where competitor citation patterns shift across the two months.

Six phases across eight to ten weeks.

Phase 1 — Setup (Week 1). A one-hour kickoff conversation to confirm scope, competitive set, priority topics, client-specific goals, and eligibility for the Google Business Profile. We configure the AI visibility tool, begin data collection, and pull your content inventory.

Phase 2 — Data collection and content extraction (Weeks 2–5). Data is collected across the four tracked engines. AI-generated answer texts are extracted, and the content map is built. Your team’s involvement remains light.

Phase 3 — Qualitative analysis (Weeks 6–9). The densest phase. The manual Claude assessment runs here, alongside qualitative reading of several hundred AI-generated answer texts and the five-pass cross-reference framework.

Phase 4 — Synthesis and recommendations (Week 9). Findings are synthesized into the strategic narrative. The tiered set of recommendations is developed: Tier 1 (highest cumulative impact), Tier 2 (optimization potential), Tier 3 (completeness).

Phase 5 — Deliverable production (Weeks 9–10). The five or six deliverables are produced in sequence.

Phase 6 — Delivery and handoff (Week 10). The engagement concludes with an executive briefing (one hour on Zoom), one round of revisions per deliverable based on your feedback, and a handoff to whoever will execute the recommendations.

There are three execution paths once the engagement closes. The right path depends on what the Intelligence analysis revealed and what your internal capacity supports. We present them as peer options, not a hierarchy.

Path A — Your team executes. The Cross-Reference Workbook documents what to publish where; the GBP Checklist (when applicable) documents the local-search tasks; the Source Data Workbook supports ongoing reference. Best when your team has content production capacity and the recommendations fit that skill set.

Path B — Telstar executes a separate implementation engagement. This routes to one or more of our other services: Priority Fixes for technical work, Content Strategy for content planning, or Content Development for writing and publishing.

Path C — Ongoing Partnership for continued analysis and iterative work. It picks up where Intelligence ends, continuing AI Visibility tracking and iterating on the recommendation set as AI engines evolve. Best when you recognize AI Visibility is an ongoing discipline rather than a one-time project.