What a Job Post Tells You That a Database Cannot
Outcome: a written search definition — roles, seniority, geography, volume threshold and recency — plus the qualifying rule you will gate on in lesson 07.
- Surface
- App and MCP server
- Level
- Intermediate
- Uses
check_credits- Credits
- 0
- Prerequisite
- A defined ICP — see GTM Engineering Foundations, lesson 01
Six things a post reveals
A firmographic record describes a company’s shape. A job post describes its intent — what it has decided to do next and already allocated budget to.
A funded decision
A posted role is money that survived a planning cycle. It is more reliable evidence of committed budget than a funding round, which tells you money exists but not where it is going.
The team taking shape
The title and seniority tell you which function is being built and at what level. A first Head of RevOps means a function that did not exist last quarter.
The stack, in writing
Requirements sections name tools. That is current, self-reported stack data, often more accurate than tech-detection because it reflects what they run internally rather than what loads in a browser.
The problem, in their words
Responsibilities describe what is broken well enough that the copy can be reused. “Own reporting across a fragmented tool set” is a sales message someone else already wrote for you.
Timing
The posting date is a real timestamp. You know how long this has been a priority, and whether it is accelerating.
Location and structure
Where the team is being built, and whether it is remote, hybrid or a new office — which is itself an expansion signal.
Why it leads databases
The sequence for a new function looks like this:
| Week | What happens | Where it is visible |
|---|---|---|
| 0 | Budget approved | Nowhere |
| 1–2 | Role posted | Job boards — you are here |
| 4–12 | Hiring, interviewing | Job boards |
| 12–16 | Person starts | LinkedIn profile |
| 16–24 | Headcount and org data updates | Firmographic databases |
| 20–30 | They buy tools | Tech-detection data |
Selling on tech-detection data means arriving at week 20+, after the decision. The job post is available at week 1, and the hiring manager’s problem is most acute while the role is open — they are doing the work themselves.
This is why the hiring manager, not the eventual hire, is often the better contact. They are absorbing the workload the role is meant to relieve, and they own the budget. Lesson 05 targets the manager one level above the posted role.
Intent versus noise
The presence of a post is weak evidence. The pattern across posts is strong evidence.
| What you see | What it means | Worth working? |
|---|---|---|
| First senior role in a function | New function starting | Yes — highest value |
| 3+ roles in one team within 30 days | Funded build-out | Yes — strongest |
| Same role reposted for 4+ months | Backfill, churn, or an unfillable spec | No |
| Hiring across every function at once | General growth, no specific project | Only with another qualifier |
| Requirements name a competitor’s product | Active user of an alternative | Yes — different message |
| Contract or intern roles only | Not a budget signal | No |
Two rules fall out of this, and they are the whole qualifying logic of the course:
- Count roles per team, not per company. Ten posts across ten functions is a growing company. Three posts in one function is a project.
- Recency beats volume. Three roles posted this month outrank ten posted over the last year.
Write your search definition
Lesson 02 needs five inputs. Decide them now, with real values.
| Input | Guidance | Your answer |
|---|---|---|
| Roles | The titles whose existence implies your problem. Include the manager titles, not just the ICs. | |
| Seniority | Which levels count. A single junior post is rarely a budget signal. | |
| Geography | Where you can actually sell and support. | |
| Volume threshold | Minimum roles in one team to qualify. Two is a reasonable default. | |
| Recency window | How fresh. 30 days is the usual choice — 90 catches more and converts worse. |
The most common mistake is on the roles line. Teams list the job titles of their users and miss the titles whose hiring implies the problem. If you sell data tooling, “Data Engineer” is a user title, but “Head of Analytics” being hired for the first time is the stronger signal — it means the function is being stood up and the tooling decisions are all still open.
Do this now
- Fill in the five rows above with concrete values.
- Write your qualifying rule as one sentence, in the shape: “An account qualifies if it has posted ≥ N roles matching [roles] at [seniority] in [geography] within [window].”
- Check the rule against three accounts you have already won. If your best-fit customers would not have passed it, the rule is too tight — loosen the volume threshold before the role list.
- Run
check_creditsand note the balance. Lessons 02–04 cost roughly 4 credits per company screened, so your expected screening volume times four is the run cost before any people are found.
Check your work
- Your role list includes manager and leadership titles, not only IC titles
- You have a numeric volume threshold and a recency window, not “recently”
- Your qualifying rule fits in one sentence and would have caught past wins
- You know what a single reposted role means and have decided to exclude it
Where this breaks
Job-post signals go stale faster than any other trigger in the library, and the failure is silent. A post that is 60 days old has often been filled — the manager’s problem is solved, and a message referencing the opening reads as sloppy rather than timely. Filter on posting date rather than on “currently listed”, and if your run cadence is slower than monthly, this signal is not the right one for your team. Lesson 08 covers the scheduled version.
Further automation
Once the search definition is stable, the highest-leverage extension is pairing it with the stack data in the requirements text. An account hiring three infra roles and naming a competing tool in the spec is a displacement opportunity with a deadline attached — a much stronger list than either signal alone. Lesson 06 writes to that segment differently, and Signals & ABM covers combining triggers into one weighted score.
Next lesson
02 — Search openings by role and geography, turning the five inputs above into a run against Company job listings.
Reference for this lesson: Is hiring, Company job listings, Headcount growth, Signals.