For B2B SaaS, timing is everything.

For companies whose customers are online, contact data is a commodity.

Traditional Signals are a joke. Super generic to be useful.

That leaves no free lunch. All channels are spammed.

The edge is knowing when and which accounts to double down on, so that you can send them a cake or invite them to your wedding.

Real buying windows are nuanced to each vertical and product.

With LLMs, reasoning models, and context graphs, can we infer these windows from public data alone?

Think about what's available: People data snapshots over time Their engagement patterns Job postings and their context Company data stacked historically

Temporal movement and Context tell a story.

Can these patterns be reasoned through and generalized with LLMs?

Can you predict with measurable confidence that a buying window is open?

This is something we look to tackle this year.