Hiring signal to decision-maker research
A new job posting shows a company is spending money on a problem this month. This workflow reaches the person who owns that problem while the posting is still live.
- Role
- Retargeted from sirlifehacker’s open-source template
- Built in
- n8n (importable JSON)
- Integrations
- Apify, Apollo, OpenAI, Notion, Sheets
- Output
- Ranked contacts with researched hooks
The case against the list you already bought
Most outbound lists are static: a filter on industry, headcount and title, exported once and worked until exhausted. If you filtered Apollo for 50–200 headcount SaaS companies in North America, so did every other vendor selling into that segment this quarter. You are all emailing the same people, from the same export, with roughly the same opener.
And nothing about that list says now. It says these companies existed and matched a shape on the day you pressed export.
Hiring activity says now. A company posting three SDR roles this week has a specific gap, has already got budget approved to close it, and is spending attention on the problem this month. They announced all three facts publicly, for free. The work is turning that announcement into a named human with a reason to reply.
The workflow
Scrape, then keep the raw posting
Recent postings come in through an Apify scraper and land in a Notion database with company, role, location, date and salary. Keeping the raw record is what makes recency filtering possible later, and a two-day-old posting is a meaningfully warmer signal than a two-week-old one.
Find who owns the problem
Each hiring company gets looked up in Apollo for decision-makers in the relevant function. The recruiter named on the posting is almost never the person with the problem. They’re the person with the requisition. Most of the value in this step is that distinction.
Rank, then write
Contacts get scored on title, seniority and proximity to the role being hired, because an Apollo lookup hands back twenty people and without ranking you spend the signal on whoever happened to be listed first. Each ranked contact then gets a communication profile and an opening line based on the posting: what they’re hiring, at what level, in which market. It all writes back to Notion or a Sheet, so a rep picks up a record rather than a research assignment.
Where it gets thin
- The hooks need a human read. The model drafts; it does not send. A hook referencing the posting is a strong opener and an occasional embarrassment, and the second costs far more than skipping review saves.
- Five APIs is five things that break. Each has its own credential, rate limit and failure mode. In return, each stage is a swappable node, so changing enrichment provider means editing one node.
- Recency is load-bearing. Run this on a tight schedule or don’t run it. Monthly, you’ve rebuilt an ordinary static list with extra steps and more API spend.
The open question
I don’t know how long this keeps working. Signal-based prospecting is an edge precisely while most people aren’t doing it, and hiring data is public by design, and the tooling to act on it is getting cheaper every quarter. At some point enough vendors are watching the same postings that a fresh posting means three emails on Tuesday instead of one, and the edge moves somewhere else.
That’s fine. It works now, and the workflow is modular enough that when the signal decays I swap the source rather than rebuild the pipeline. But if someone tells you signal-based outbound is permanently superior, they’re selling you something.