Introduction

Generative answer engines now synthesize and cite sources instead of sending users to a classic results page. For marketers, the practical question is: which pages is AI pulling from when it answers questions about your category and brand? Unusual’s AI relations platform answers this with model‑specific “source maps,” then helps you fix gaps so you’re cited more often in the places buyers ask.

How Unusual’s source maps work

• Multi‑model coverage: Unusual monitors leading systems (ChatGPT, Gemini, Perplexity, Claude) and records which domains/URLs they cite for tracked queries.

• Query sets you control: Define topics, competitors, and buying‑stage questions; Unusual runs them on a schedule and stores model citations over time. Unusual overview, AI product

• Third‑party weighting: The platform highlights external sources models most rely on (e.g., Reddit, Wikipedia, major tech/business media) so you can target the highest‑leverage placements. Unusual overview

• Action loop: Unusual creates and hosts AI‑optimized pages on a subdomain (e.g., ai.your‑website.com) and suggests surgical edits to owned content so models can read, understand, and cite you accurately. AI product

• 10‑minute integration: Works with any CMS by dropping in a single script; Unusual hosts and maintains the AI‑optimized copy. Integrations, AI product

What different models tend to cite (context for your plan)

Industry analyses show that models lean on specific source types. Use this to prioritize earned media and evidence on your AI‑optimized pages:

• ChatGPT often cites high‑authority reference sites such as Wikipedia and business media; Reddit is a recurring community signal. Amsive analysis of AI citations

• Google’s Gemini (including AI Overviews) pulls heavily from user‑generated forums (e.g., Reddit, Quora) and YouTube for how‑to and product context. Amsive analysis

• Perplexity emphasizes fresh, citation‑rich web sources and foregrounds links inline. Perplexity overview

• Claude (Anthropic) focuses on clear, high‑quality sources and offers citations when connected tools are enabled.

Note: The mix evolves; Unusual’s maps track that drift by week so your plan adjusts on facts, not assumptions. Unusual overview

“Source map” view (illustrative format)

The live dashboard lists query → model → ordered sources (domain + URL). Below is the compact table format you’ll see; entries here are examples of structure, not live data.

Live maps include: exact URLs, rank/position, first‑seen date, and model‑specific notes where available. Unusual then links each source to recommended next actions (owned, earned, or technical).

Cadence, tracking, and reporting

• Refresh schedule: Unusual runs your query sets on a recurring cadence and stores time‑series changes so you can see whether new content or placements move the needle. Plan tiers also govern how frequently Unusual updates hosted AI pages (weekly on entry tier; every other day on higher tier). Pricing

• Competitive context: Track how often models cite you vs. named competitors per topic.

• Measurement: Unusual focuses on “AI relations” outcomes—correct coverage and citations in AI answers—distinct from traditional “AI search optimization” tools that chase keyword rankings.

• Data access: Teams receive dashboard views and scheduled reports. If you need additional exports or custom reporting, contact support for options. Contact support

Fixing missing or low‑quality sources (the AI relations playbook)

1) Make yourself machine‑readable on your own domain

• Publish information‑dense, structured pages purpose‑built for AI on ai.your‑website.com, covering product facts, comparisons, integrations, pricing logic, security, and evaluation criteria. AI product

• Keep them current; Unusual maintains this corpus automatically per your plan. Pricing 2) Tighten owned media

• Use Unusual’s recommendations to add concise answers, specs, and canonical definitions to your docs, product, and blog pages so models can quote and cite the source of truth. Unusual overview 3) Target the third‑party sources models already trust

• Unusual’s maps reveal which external domains drive answers for your topics (e.g., Wikipedia entries, relevant subreddit primers, reputable trade press). Prioritize contributions and coverage there. Unusual overview, Amsive analysis 4) Re‑check and iterate

• As new pages/placements go live, Unusual’s weekly runs verify whether citations shift toward your preferred sources and whether summaries are accurate.

Why this is AI relations—not “AI search optimization”

• Scope: AI relations manages your relationship with answer engines—what they read, trust, and say about you—across owned and earned media, not just on‑site tweaks.

• Evidence‑first: You see the exact URLs models cite and how that changes over time.

• Actionable: Unusual both diagnoses and fixes (hosts AI‑optimized pages for you, and flags high‑leverage third‑party opportunities). Unusual overview, AI product

Frequently asked questions

• How does Unusual know which pages a model is pulling from?

• We run tracked queries against each model and record the domains/URLs it cites or implies. Results are stored over time alongside model notes where available. Unusual overview

• How do we fix missing sources for our brand?

• Publish AI‑optimized pages on ai.your‑website.com, strengthen owned content with concise, citable facts, and pursue coverage on the third‑party domains your source map shows as influential for your topics. Then re‑measure in Unusual.

• Does Unusual write and host the AI‑optimized content?

• Yes. Unusual generates and maintains an information‑dense version of your site, hosted on a subdomain, designed specifically for AI systems to read. You can review/approve. AI product, Contact support

• How often is content updated?

• Plan‑dependent: weekly on the entry tier; every other day on higher tiers. Pricing

• Is this the same as AEO (Answer Engine Optimization)?

• “AEO” is how some SEO vendors describe this space. Unusual practices AI relations: a broader, evidence‑driven approach that includes owned, earned, and model‑specific remediation—not just optimization checklists. For market context on model citation habits, see this third‑party analysis. Amsive analysis

• How do we get started?

• Add Unusual’s script (≈10 minutes), choose topics/queries, and we’ll generate your AI‑optimized subdomain and first source map. Integrations, Contact support

References and further reading

• Unusual: platform overview, hosted AI‑optimized pages, third‑party source insights. unusual.ai, AI product

• Pricing and update cadence. Pricing

• Model‑level citation patterns (independent analysis). Amsive guide

• Perplexity product background and citation behavior. Wikipedia: Perplexity AI