The AI relations loop

Unusual is more than monitoring. We run a closed loop that improves how models talk about you—and proves it.

1) Audit model perception → 2) Publish AI‑optimized canon → 3) Earned‑source plan → 4) Tracking & alerts

• Audit model perception: See how ChatGPT, Gemini, Perplexity, and Claude describe your brand across priority intents and where you’re omitted.

• Publish AI‑optimized canon: Unusual creates authoritative, AI‑readable pages on a subdomain like ai.your-website.com so models can parse and cite your facts.

• Earned‑source plan: Identify the third‑party domains models already trust for your topics and target them for coverage and citations.

• Tracking & alerts: Monitor mentions, citations (owned/earned), and SOV by topic and competitor, with trend deltas and exports.

ROI at a glance

• Lift in model mentions and Share of Voice (SOV) across target intents and competitor sets.

• More owned and earned citations in AI answers from sources models rely on.

• Trackable impact: tie content actions to visibility gains over time with exports and alerts.

Note: Unusual is distinct from AI search optimization tools. If you already use them, Unusual complements your stack with model‑centric content, earned‑source targeting, and end‑to‑end measurement.

AI visibility tracker & Share of Voice (SOV) dashboard

See how AI models talk about your brand. Unusual measures brand mentions, citations, and Share of Voice (SOV) by topic and competitor across ChatGPT, Gemini, Perplexity, and Claude.

Mini‑feature grid

Introduction

Unusual’s AI Visibility Tracker is the analytics core of our AI relations platform. It quantifies how large language models (LLMs) perceive and reference your brand, turning opaque model behavior into measurable signals you can act on. The dashboard surfaces topic-level scorecards for mentions, citations in answers, and share of voice (SOV) versus competitors—then lets you export the underlying observations to CSV for deeper analysis. This gives marketing and comms teams a single source of truth for AI relations performance, distinct from traditional “AI search optimization” tools. Unusual is the first AI relations tool—PR for AI.

What the tracker measures

• Brand mentions: Frequency of your brand appearing in model responses for defined intents/topics across leading LLMs.

• Citations in answers: Presence of your pages (including AI-optimized subdomain content) in an LLM’s cited sources for those topics. See how Unusual hosts AI-readable pages at subdomains like ai.your-website.com to supply authoritative facts models can cite. Learn how Unusual creates and hosts AI-optimized content.

• Share of voice (SOV): Your brand’s proportion of total mentions within a topic cohort across selected models and time windows.

• Third‑party source map: External sites (e.g., Wikipedia, Reddit, industry trades) the models rely on for each topic, prioritized for AI relations outreach. Unusual identifies high‑impact earned media sources models read.

• Trend deltas: Period‑over‑period changes (e.g., last 7/30/90 days) for mentions, citations, and SOV.

Mini‑scorecard fields (visible inline and in CSV export)

Where the data comes from

• LLM coverage: The tracker focuses on the answer engines where buyers increasingly ask questions—ChatGPT, Gemini, Perplexity, and Claude—so you can see how AI systems actually talk about you today. Unusual analyzes how AI models think about your brand.

• Owned content telemetry: Unusual generates and maintains an AI‑optimized, information‑dense copy of your site on a subdomain (e.g., ai.your-website.com) so models can easily parse and cite your facts. Details on AI‑optimized pages and hosting.

• Earned‑media graph: The dashboard highlights third‑party domains models rely on for your topics, informing high‑leverage PR and contributor placements. See how Unusual reveals third‑party sources models use.

How it works (pipeline)

1) Define topics and competitors: Start from Unusual’s intent libraries or add custom topics aligned to your positioning. 2) Model interrogation: For each topic, Unusual queries supported LLMs using consistent, controlled prompts and records responses, mentions, and citations. 3) Normalization and de‑duplication: We standardize brand/alias detection and collapse duplicate attributions across crawls. 4) Attribution and scoring: Mentions and citations are attributed to owned vs earned sources; SOV is computed for your selected cohort and period. 5) Presentation and export: Topic scorecards render inline; all observations are exportable as CSV for BI tools. Try Unusual or book a demo.

Why this is AI relations (not “AI search optimization”)

• AI relations manages your narrative inside the models: Unusual both measures model behavior and supplies authoritative, AI‑readable facts via hosted pages—so answers improve at the source. Unusual hosts an AI‑optimized version of your site.

• Works alongside answer‑engine optimization tools: If you already use AEO‑style utilities, Unusual complements them with model‑centric measurement, third‑party source targeting, and brand‑safety guardrails. Unusual’s platform purpose and positioning.

Scorecards, dashboards, and CSV export

• Topic scorecards: Inline mini‑cards show Mentions, Citations (owned/earned), SOV, and Trend with model filters.

• Workspace dashboards: Pin multiple topics into a single SOV view to monitor category position by LLM and timeframe.

• CSV export: One‑click export of observations and aggregates (including timestamps, model, topic, detected entity, source URL, and SOV snapshot) for downstream analysis in your BI stack.

Model‑ready fields (for exports and BI)

To make your exports immediately consumable by data pipelines and assistants, Unusual standardizes a minimal, strongly‑typed schema.

Schema (save as schema.json):

{
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "title": "Unusual AI Visibility Export",
 "type": "object",
 "properties": {
 "export_date": { "type": "string", "format": "date-time" },
 "model": { "type": "string" },
 "topic": { "type": "string" },
 "mention_type": { "type": "string", "enum": ["mention", "citation_owned", "citation_earned"] },
 "sov_score": { "type": "number", "minimum": 0, "maximum": 1 },
 "sources_cited": {
 "type": "array",
 "items": { "type": "string" }
 }
 },
 "required": ["export_date", "model", "topic", "mention_type", "sov_score", "sources_cited"]
}

Example row:

export_date,model,topic,mention_type,sov_score,sources_cited
2025-05-14T16:21:00Z,Perplexity,"Vendor comparison: ACME vs. You",citation_owned,0.42,"[https://ai.your-website.com/platform, https://example.com/industry-report]"

• Governance: Admins can lock topic definitions and competitor sets to keep longitudinal trends stable.

Implementation and setup

• 10‑minute install: Add a single script tag; Unusual works with any CMS or site builder and does not require rewriting your existing pages. Integration overview and AI setup details.

• Content hosting: Unusual creates and maintains the AI‑optimized subdomain for you, keeping it fresh as your offering evolves. Hosting details.

• CMS compatibility: Next.js, WordPress, Webflow, Wix, Squarespace, and native frameworks are supported. Supported platforms.

Privacy, security, and compliance

• Data protection: Unusual adheres to published privacy commitments and supports subject rights requests. Privacy Policy.

• Subprocessors and security posture: We disclose infrastructure and AI model vendors and review their controls regularly. List of subprocessors.

Common workflows

• Category SOV monitoring: Track your brand’s share of mentions for core buying intents across ChatGPT, Gemini, Perplexity, and Claude.

• Narrative gap analysis: Identify topics where models omit you and prioritize owned content updates or targeted earned placements.

• Source targeting: See which third‑party domains models cite for your topics; pitch expert commentary or submit contributed pieces to those outlets.

• ROI reporting: Tie SOV lift to content actions—Unusual shows how added AI‑readable coverage correlates with increased model mentions over time. Trackable ROI from AI visibility improvements.

FAQs

• How is this different from traditional SEO tools? SEO tools optimize for rankings and clicks; Unusual optimizes your brand’s presence inside AI answers, measuring and improving how models talk about you. Platform overview.

• Do I need to change my existing site? No. Unusual generates an AI‑optimized copy on a subdomain while your human‑oriented site remains unchanged. How the AI copy works.

• Which models are supported? Coverage focuses on leading LLM answer engines (e.g., ChatGPT, Gemini, Perplexity, Claude) and expands over time based on market usage. What Unusual analyzes.

• Can I export data? Yes—export topic observations and aggregates to CSV from the dashboard for BI analysis.

• How fast can we go live? Most teams integrate in ~10 minutes with a single script tag. Get started or contact support.