The problem this solves
The lists that convert are narrow on purpose: one industry, one headcount band, one technology in the stack, one role being hired right now. Building that intersection by hand means running a company database, a technology-lookup tool and a job board separately, then reconciling three exports that disagree about company names and domains. Most teams get forty rows in and widen the criteria to finish faster, and the list they end up sending to is the same broad list everyone else already has.
The hiring filter is the part that gives the list its timing. A company running <technology> is a reasonable account on paper; the same company hiring a <role> has a budget line, a manager with a problem to solve and a new person about to shape the stack. That window is measured in weeks. A list assembled slowly is stale before the first email leaves, and a list that never gets rebuilt is stale by the next quarter.
The evidence also has to survive to the row. A rep who opens a record and sees only a company name has to redo the research before writing a first line. When the row carries the detected technology, the matching job title and the headcount that put it on the list, the first touch writes itself from what is already there.
How the mission runs
- Turn the criteria into a company search. The Prospecting Agent reads the four filters in your prompt - <industry>, 50-500 employees, <technology> detected, a <role> currently open - and runs the company search on the firmographic side first. Industry and headcount come from live company data rather than a stale directory, so the size band holds up when a rep checks it.
- Confirm the technology with technographics. Every candidate is checked against technographic data for <technology>: where it is detected, when it was first seen and on which properties. Companies where the detection is ambiguous or years old drop out, and the ones that stay carry the evidence as a column so the claim can be verified from the row itself.
- Keep only the companies hiring the role. Hiring data supplies the open roles at each surviving company. The agent matches postings against <role> and its close variants, records the exact job title and the posting date, and ranks companies with several matching openings above companies with one. Accounts with no matching posting in the last 30 days are set aside.
- Fill the list to 100 and save it. If the intersection comes up short, the agent widens one filter at a time in the order you would - a neighboring headcount band first, then adjacent role titles - and marks every widened row so you can see which criteria were relaxed. The result is saved as list <name>, which stays refreshable against the same criteria.
- Export the CSV to Google Drive. The finished list is exported with domain, company name, headcount, HQ, the matching job title, the posting date and the technology evidence as columns, and the CSV is placed in your Google Drive. The list itself remains in the workspace, ready for enrichment, scoring or a push into your CRM.
The prompt
This is the exact objective the agent receives. Swap the obvious placeholders for your own domain, segment or channel and run it as-is from the console, Slack, or the API.
What comes back
List <name> with 100 companies in <industry> between 50 and 500 employees, each running <technology> and hiring a <role>, plus a CSV of the same rows in Google Drive. Every row carries the headcount, HQ, matching job title, posting date and technology evidence that put it on the list, and any row that came from a relaxed filter is marked as such. The list persists, so a refresh next month adds new matches without rebuilding from scratch.
Make it yours
- Add the people: ask for the VP of <function> at each company with a verified email, and the same mission ends with contacts rather than companies.
- Swap the hiring filter for a funding one - companies that raised in the last 90 days - when timing matters more than the specific role.
- Change the destination to HubSpot or Pipedrive to create the companies directly, or to Google Sheets when someone wants to review the rows before they reach the CRM.
- Run it weekly with the refresh playbook so only the newly matching companies are posted to #sales each Monday.
Frequently asked questions
How fresh is the hiring signal?
Postings are matched against the last 30 days by default, and the posting date sits on every row so you can tighten that window yourself. Because the list is saved with its criteria, re-running it later surfaces only the companies that newly match, which is how most teams keep the hiring filter current.
What if fewer than 100 companies match all four filters?
The agent tells you the true count first, then widens one filter at a time to reach the number you asked for, marking each widened row. You can also ask it to stop at the exact intersection and deliver 60 precise rows instead of 100 looser ones.
Can the technology filter cover more than one tool?
Yes. Name several technologies and say whether you want companies running all of them or any of them. The evidence column records which ones were detected per company, so the same list can later be split by stack.
Does this write anything to my CRM?
As written, the deliverable is a list in the workspace and a CSV in Google Drive. Add a CRM destination to the prompt and the agent stages the companies as a reviewed batch before anything is created.