SpikePanel

Turn social noise into market signals.

Productivity

Artificial Intelligence

Visit WebsiteSee on Product HuntTwitter ⧉

Hi everyone, maker here. If social media is part of how you follow US stocks, you know the problem: on X and Reddit, stock discussion runs to thousands of posts a day, repetitive and noisy, impossible to keep up with. SpikePanel filters that down to a few minutes of reading. Open it and you know: • Which stocks people are discussing on X and Reddit right now, and which conversations are heating up • Why a stock moved today: the earnings, policy, or industry event behind it, in one line • Today's headlines, condensed from thousands of posts into one page • Which themes are running (memory chips, nuclear, optics...) and how an event spreads along the supply chain to related stocks The filter has three layers, all built in-house. We first map the circles where US stock conversation actually happens, so accounts earn their place by network position, not follower count. Then our propagation models separate original information from echo, organic spread from coordinated hype, and catch topics before they peak. Behind it all sits a supply chain knowledge graph where every relationship must be confirmed across multiple independent sources and cross-validated by our algorithms. Keyword tools count mentions; we model who is worth listening to and how information moves. One rule throughout: report what is being discussed and what happened, never predict prices or push urgency. A noise filter should not add noise of its own. If you follow stocks through social media too, what takes you the most time every day? That is probably the thing we should filter next.