Topic
Signal-based lead intelligence
Finding the companies that are about to buy from what they are doing in public, such as hiring, restructuring and leadership changes, rather than from static firmographic filters.
Overview
Most lead lists are built backwards. They start with a filter (industry, headcount, a job title) and hope that some of the companies in it happen to need what you sell. Signal-based lead intelligence starts from the other end: it looks for public evidence that a company is changing in a way that creates a need, then works out who to talk to.
The evidence is usually mundane. A job posting for an FP&A analyst that mentions automating month-end reporting. A press release about a new finance director. A company that posted three related roles in one month. Each of these is weak alone. What makes them useful is scoring them consistently, discounting them as they age, and noticing when several point at the same company.
The pipeline I built for this does exactly that, in seven stages: collect job-posting and public signals, score each one with an LLM against a fixed rubric, apply recency decay, cluster signals by company and boost companies with several, validate each company against the ideal customer profile from its own homepage, find the likely decision-makers, and export a ranked list with the evidence attached. Every rejected row is kept in a per-stage rejection log, which is where threshold tuning starts.
Projects
Where I did this
- Founder 2025 to present
I'm building a system that identifies companies preparing to buy, from public signals such as hiring, leadership changes and transformation announcements, instead of static job-title filters.
Questions
What is signal-based lead intelligence?
It ranks companies by observable evidence that they are about to need something, for example a job posting for a finance systems role or a new CFO announcing a transformation, instead of filtering a database by industry, headcount and job title. Each lead comes with the signal that triggered it, so an outreach message can reference something real.
Why use recency decay?
A hiring signal from last week says much more about current need than one from three months ago. In the pipeline I built, a signal keeps its full score for 14 days, then counts for 75 percent up to 30 days and 45 percent up to 60. Between 60 and 90 days only the strongest signals survive, at 20 percent, and anything older is discarded.
How is this different from intent data vendors?
Intent data vendors sell aggregated topic-interest scores, usually on a monthly subscription. A signal pipeline like this one is narrower, uses only public sources, and shows the specific evidence behind every lead. It is cheaper to run and easier to audit, but it covers fewer companies.