Signals Database

Signals Database

The structured catalog of the signals that shape which store, product, or offer an AI shopping agent selects. Every record is evidence-typed (spec-fact, reported, or hypothesis) with a confidence level and, where a platform documents its behavior, honest per-platform status.

The whole database is also published in machine-legible form: /signals.json (structured) and /signals.md (Markdown catalog).

32 signals, sorted by funnel stage (discovery, matching, comparison, choice), then impact.
SignalAffectsPlatforms documentedEvidenceConfidenceImpactLast verified
Agent infrastructure reachabilityDiscovery1/5 ChatGPTHypothesisLow3/3
Content equivalence (no cloaking)Discovery1/5 CopilotSpec-factVery high3/3
Crawl directives (robots.txt + sitemap)Discovery1/5 ChatGPTSpec-factVery high3/3
Agent-readiness monitoring and ownershipDiscoveryNone documentedHypothesisLow2/3
llms.txt priority orderingDiscoveryNone documentedSpec-factVery high2/3
Machine-readable page representationDiscoveryNone documentedHypothesisLow2/3
Origin serve-ability for agentsDiscoveryNone documentedHypothesisLow2/3
Track AI-answer presence, not just referralsDiscoveryNone documentedHypothesisLow2/3
Offer markup completeness and truthfulnessMatchingNone documentedSpec-factVery high3/3
Server-rendered, crawlable contentMatchingNone documentedHypothesisLow3/3
Product identity and canonical URLMatchingNone documentedSpec-factVery high2/3
Validation hygiene (use existing validators)MatchingNone documentedSpec-factVery high2/3
Variants and titles use the words shoppers typeMatchingNone documentedHypothesisLow2/3
Competitive price awareness with parityComparisonNone documentedReportedHigh3/3
Delivery ETA is computable per destinationComparisonNone documentedHypothesisLow3/3
Descriptions answer relayed buyer questionsComparisonNone documentedHypothesisLow3/3
Facts per token, complete against peersComparisonNone documentedHypothesisLow3/3
Offer cost legibilityComparisonNone documentedHypothesisLow3/3
Query-conditional rewrites are a targeted experimentComparisonNone documentedReportedHigh2/3
Return terms are fully legibleComparisonNone documentedHypothesisLow2/3
ACP feed and checkout readinessChoice1/5 ChatGPTSpec-factVery high3/3
Agent-payment rail readinessChoiceNone documentedReportedHigh3/3
Audit grid position bias per providerChoiceNone documentedReportedHigh3/3
Defined agent win-rate measurementChoiceNone documentedHypothesisLow3/3
Earn endorsement, never self-label sponsoredChoiceNone documentedReportedHigh3/3
Offer consistency and freshness across surfacesChoiceNone documentedHypothesisLow3/3
Re-test after every major model updateChoiceNone documentedReportedHigh3/3
Review depth: count, recency, distributionChoiceNone documentedHypothesisLow3/3
Guest checkout completabilityChoiceNone documentedHypothesisLow2/3
Price sensitivity varies by modelChoiceNone documentedReportedHigh2/3
Seller identity reachable in two hopsChoiceNone documentedHypothesisLow2/3
UCP profile and order-management readinessChoice1/5 GeminiSpec-factVery high2/3

Platform coverage matrix

Signals as rows, shopping surfaces as columns. Most cells read unknown: which signals each engine actually weights is largely undisclosed, so a cell is filled only where a platform documents its own behavior (5 of 160 cells). This view is generated from the signal records; fixing a cell means fixing a signal, never this table.

Per-platform status for each signal. Unknown is the honest default where a platform has not documented its behavior.
SignalChatGPTGeminiPerplexityCopilotClaude
Agent infrastructure reachabilitySupportedUnknownUnknownUnknownUnknown
Content equivalence (no cloaking)UnknownUnknownUnknownReportedUnknown
Crawl directives (robots.txt + sitemap)SupportedUnknownUnknownUnknownUnknown
Agent-readiness monitoring and ownershipUnknownUnknownUnknownUnknownUnknown
llms.txt priority orderingUnknownUnknownUnknownUnknownUnknown
Machine-readable page representationUnknownUnknownUnknownUnknownUnknown
Origin serve-ability for agentsUnknownUnknownUnknownUnknownUnknown
Track AI-answer presence, not just referralsUnknownUnknownUnknownUnknownUnknown
Offer markup completeness and truthfulnessUnknownUnknownUnknownUnknownUnknown
Server-rendered, crawlable contentUnknownUnknownUnknownUnknownUnknown
Product identity and canonical URLUnknownUnknownUnknownUnknownUnknown
Validation hygiene (use existing validators)UnknownUnknownUnknownUnknownUnknown
Variants and titles use the words shoppers typeUnknownUnknownUnknownUnknownUnknown
Competitive price awareness with parityUnknownUnknownUnknownUnknownUnknown
Delivery ETA is computable per destinationUnknownUnknownUnknownUnknownUnknown
Descriptions answer relayed buyer questionsUnknownUnknownUnknownUnknownUnknown
Facts per token, complete against peersUnknownUnknownUnknownUnknownUnknown
Offer cost legibilityUnknownUnknownUnknownUnknownUnknown
Query-conditional rewrites are a targeted experimentUnknownUnknownUnknownUnknownUnknown
Return terms are fully legibleUnknownUnknownUnknownUnknownUnknown
ACP feed and checkout readinessSupportedUnknownUnknownUnknownUnknown
Agent-payment rail readinessUnknownUnknownUnknownUnknownUnknown
Audit grid position bias per providerUnknownUnknownUnknownUnknownUnknown
Defined agent win-rate measurementUnknownUnknownUnknownUnknownUnknown
Earn endorsement, never self-label sponsoredUnknownUnknownUnknownUnknownUnknown
Offer consistency and freshness across surfacesUnknownUnknownUnknownUnknownUnknown
Re-test after every major model updateUnknownUnknownUnknownUnknownUnknown
Review depth: count, recency, distributionUnknownUnknownUnknownUnknownUnknown
Guest checkout completabilityUnknownUnknownUnknownUnknownUnknown
Price sensitivity varies by modelUnknownUnknownUnknownUnknownUnknown
Seller identity reachable in two hopsUnknownUnknownUnknownUnknownUnknown
UCP profile and order-management readinessUnknownSupportedUnknownUnknownUnknown

[ newsletter ]