Why FinScry exists
I built FinScry because I wanted to trade around catalyst events, and kept losing the day to research instead of trading.
A lot of tradeable dates have no structured source. The announcement itself is public, but the date sits inside a PDF, in a single sentence, with nothing to query it by. Finding them meant reading several announcements per company, noting the dates by hand, and typing them into a calendar. Every week, before I'd made a single decision.
FinScry is the tool I wanted back then. It watches around 1,900 listed companies and turns what they file into a single calendar, with the announcement each date came from one click away.
How I think about the AI
A language model reads the announcements and pulls the dates out. Rule based extractors cover the cases it can't handle, and there is a manual review process behind both. I'd rather you knew exactly how that works than be impressed by it, which is why the Data Quality page sets out the method and where it falls short.
I think this is a genuinely good use of the technology. It reads thousands of PDFs that nobody has time to read, and it gets checked on the results. That isn't the same as being right every time. Every event links to the announcement it came from, and I'd encourage you to open it.
What it isn't
It isn't a screener, and it doesn't recommend anything. It won't tell you what a date means for a share price, only that the date exists and where the company said so. Announcements are read after they are filed rather than streamed in real time, so if your strategy depends on reacting within seconds of a disclosure, this is the wrong tool.
Who it's for
Brokers, individual traders who have hit the same wall, and anyone who wants another data layer alongside what they already use. What it gives you back is the hours you'd have spent working out when things happen.
Get in touch
Questions, corrections and feature requests all reach me directly. The contact page has the addresses, and a wrong date is always worth reporting.