Predictions grounded in real data
CitePulse deploys hundreds of autonomous agents to simulate AI search behaviour. Every prediction is anchored to a live knowledge graph built from your real citation data, competitor activity, and ABS statistical enrichment.
Live knowledge graph
Audi Perth — real citation data
This is what CitePulse sees: your brand, competitors, keywords, and AI platforms mapped in a live knowledge graph. Drag to rotate. Scroll to zoom. Click nodes for details.
Click a node
Explore relationships and citation data
BMW Perth added FAQ pages and gained 3 citations on Claude. Consider FAQ schema for your service pages.
Audi Perth has the highest citation confidence scores on Perplexity — 8 of 15 prompts.
Your visibility dropped 23% on Gemini this week, correlating with Mercedes Perth indexing new content.
The automotive sector is seeing an 18% increase in AI citations this quarter — momentum is in your favour.
How it works
From raw citations to grounded predictions
Four stages transform raw citation data into the cumulative knowledge that makes CitePulse predictions trustworthy — not hallucinated.
Citations + social mentions
Verified citations from 5 AI platforms flow in continuously, alongside social mentions from Reddit and Twitter via SearXNG.
ChatGPT, Claude, Gemini, Perplexity, Grok + Reddit, Twitter
LLM entity extraction
Graphiti auto-extracts 6 entity types — Company, Platform, Trend, Strategy, Metric, Technology — and maps their relationships.
6 entity types · temporal edges · semantic embeddings
ABS statistical data
Australian Bureau of Statistics data enriches every prediction: 2,200+ SA2 regions with population, industry density, and employment growth.
2,200+ SA2 regions · population · industry · employment
Prediction context injection
Cumulative knowledge is injected into prediction agents via hybrid retrieval — semantic search, keyword matching, and graph traversal combined.
Semantic + keyword + graph traversal retrieval
Why it matters
Grounded predictions vs. generic guesses
Without a knowledge graph, predictions are generic. With CitePulse, every forecast is anchored to your specific citation patterns, competitor data, and market context.
“Your visibility may increase over the next 14 days.”
“Consider adding more content to improve citations.”
“Competitors may be affecting your rankings.”
“ChatGPT citation rate predicted to reach 41% in 14 days (78% confidence).”
“Gemini visibility dropping — BMW Perth FAQ pages indexed 3 days ago.”
“Add FAQ schema to /services — est. +15% citation rate.”
Connected suite
CitePulse doesn't work alone
Every Outercite product feeds into and draws from the same knowledge graph — creating a compounding intelligence advantage.
Vericite™
Feeds verified citations into the graph — only dual-model confirmed data enters.
Citescout™
Competitor data enriches relationship mapping — who cites whom, and why.
Citezens™
Agents query the graph for context before every analysis and prediction.
Citerank™
Authority scores derived from graph centrality and citation patterns.
See CitePulse in action
Deploy your agent team and watch CitePulse build a knowledge graph unique to your business — grounding every prediction in real data.