Source-grounded, timestamped, citation-backed. AI-assisted, not AI-invented.
The Liquor Bets Pulse is a source-grounded relevance index. We collect public industry signals, link every item to its original source, timestamp it, and score it with a transparent, deterministic method. AI assists with collection and summarization, but it does not decide what is true and it does not set the score.
The Pulse tracks leading indicators, the signals that tend to precede awards and industry outcomes rather than report them after the fact. These are signals, not guarantees. A strong signal raises the odds that something is building; it does not promise a result. Awards-relevant signals influence commercial relevance and market fit when the evidence supports it, but they never override source quality.
Signal tags describe what kinds of evidence back a topic. They are leading indicators, not predictions. A topic carrying awards signals is one we are watching for an awards outcome, not one we are forecasting to win.
Confidence is published on a 0 to 10 scale. It is more conservative than the headline Pulse Score: it requires multiple independent sources and verified evidence. A topic backed by curation alone cannot reach high confidence.
Curation layers are private industry curation feeds modeled as a source-of-sources rather than as raw sources. They are discovery inputs that help surface what to look at. When a curated item links to an original publisher, that publisher remains the citation layer and is preserved as evidence alongside the curation signal. Curation can lift a topic, but it cannot, on its own, make a topic high-confidence.
The Pulse does not depend on any single feed. Collection draws on direct RSS and trade sources, trade publications, awards pages, PR feeds, search and trend signals such as Google Trends, community signals such as Reddit, and private curation feeds. If any one input goes quiet, scoring continues with lower curation confidence and more weight placed on direct sources, so the index stays operable.
This product uses AI-assisted research to collect, classify, summarize, and score public industry signals. Source links, timestamps, and methodology are provided wherever possible. AI does not create facts, and every published item is grounded in cited source material.
Get the weekly read with sources, timestamps, and methodology. Free.