Companies post 73% of their job openings within 30 days of approving the budget for them. That makes a job posting something rarer than an intent signal: it is a public disclosure that money has already been committed to a problem.
It is also the signal most often used badly. Analysis of a million B2B software purchases found that job postings on their own correlate with buying at just +7%. Teams that monitor hiring volume and treat every spike as an opportunity are, on that evidence, chasing noise. The value is in reading postings, not counting them.
Start with what actually correlates
| Signal | Correlation with purchase |
|---|---|
| AI tool adoption | +46% |
| Headcount growth of 10%+ in 90 days | +38% |
| Recent purchases in adjacent categories | +38% |
| Job postings alone | +7% |
Notice that headcount growth scores more than five times higher than job postings. These sound like the same thing and are not. Postings are intentions, and a large share of them are never filled. Headcount growth is the confirmed outcome — evidence that the budget survived contact with reality.
The practical rule that follows: use postings for timing and context, and use headcount change for qualification.
Read the posting, do not count it
The tools named in the description
This is the most underused free data source in B2B. Job descriptions that name software by brand — “experience with Salesforce, Outreach and a modern data enrichment stack” — tell you exactly what a company runs, and often what it is about to run.
It beats conventional technographic detection on two counts. Website-based detection sees what is installed on the marketing site; a job description reflects the internal systems people actually work in. And because a company writes the requirement before it hires, the posting reveals adoption earlier than any scanner can detect it.
Two patterns worth watching for specifically:
- A competitor named as a requirement — they are a customer, and the renewal date is a matter of research rather than guesswork.
- A category named without a product — “must be comfortable building enrichment workflows” with no tool specified usually means they are evaluating, not committed.
Volume thresholds that mean something
| Pattern | What it indicates |
|---|---|
| 1–2 postings of a role | Backfill. Usually nothing |
| 5+ SDR postings in 30 days | A funded strategic initiative |
| 10+ SDR postings in 30 days | Full outbound build-out, often alongside a new VP Sales or CRO |
| Engineering headcount +20% quarter over quarter | Infrastructure purchasing almost always follows |
The threshold matters more than the trend. A company posting two engineering roles is replacing people. A company growing engineering headcount by a fifth in a quarter is buying compute, CI/CD capacity, monitoring and security tooling, in that rough order, and the purchases are already scheduled.
The first-ever role
The single highest-signal posting is a role a company has never had before. A mid-market SaaS business creating its first Head of Revenue Operations is announcing that it intends to professionalise its go-to-market motion — and that the person hired will arrive with a mandate and a budget to buy things.
The same applies to a first CISO, a first Head of Data, a first VP of Customer Success. Each creates a buyer where none existed, with roughly 90 days to establish their stack before the organisation’s attention moves on.
What each hire predicts
| What they are hiring | What they are about to buy |
|---|---|
| SDRs and BDRs at volume | Sales tooling, contact data, sequencing infrastructure |
| Engineers at volume | Cloud, observability, developer tooling |
| Security roles | Compliance and security tooling |
| A first CISO or security VP | A full vendor review inside 90 days |
| A first RevOps lead | CRM overhaul, data quality, reporting |
| Recruiters and sourcers | ATS, sourcing tools, candidate data |
The posting text usually names the gap outright. Job descriptions state what the hire will be responsible for fixing, which is a more candid account of a company’s problems than anything on its website.
The ghost job problem
Any serious use of this data has to account for a fact the vendors selling it rarely lead with: 40% of hiring managers reported their company posted a fake job in 2024.
Companies post roles they have no intention of filling — to appear to be growing, to build a pipeline of candidates for later, to pressure existing staff, or because nobody removed a listing that closed months ago. Every one of those produces a signal that looks identical to a real one.
Three filters remove most of the noise:
- Check the posting age. A role live for more than 60 days is either unfillable or not real. Fresh postings are the ones with budget behind them.
- Confirm against headcount. If a company posted eight roles last quarter and headcount is flat, disregard the postings entirely. This is why the +38% signal beats the +7% one.
- Require specificity. Detailed, tool-naming descriptions with a named hiring manager are written by someone who needs the hire. Generic evergreen listings are not.
Timing: you are early, and that is the point
A job posting typically appears 60 to 90 days before the company begins vendor research. That is unusually early, and it changes what the outreach should say.
At that stage there is no evaluation to win. Pitching a product to a company that has not yet defined its requirements is arriving with an answer to a question nobody has asked. What works instead is arriving with the shape of the problem — what teams in the same position usually discover in month three, what the new hire will need in place before they can deliver.
The hiring spike itself decays as a signal within 30 to 45 days, but the buying window it opens runs far longer. Being useful early and still present at month three beats being fast and forgotten, which is the failure mode covered in more depth in our guide to the 2026 buying signal trigger stack.
The same feed works for recruiting
Job posting data is one of the few sources that serves two functions from a single feed, which is why it belongs in both a sales and a recruiting workflow rather than being licensed twice.
| The same posting tells sales | …and tells recruiting |
|---|---|
| Budget approved for this problem | An active requirement to place against |
| Which tools the team runs | Which skills to screen for |
| A new buyer is arriving in 90 days | A hiring manager with an open role now |
| The department is expanding | Where talent is moving, and from where |
| A category is being evaluated | Which competitors are staffing up |
For a recruiting team, a company posting ten SDR roles is a client with ten placements available. For the sales team at the same firm, it is an account building an outbound function and about to buy data. Neither view is more correct; they are the same event read from two directions.
Agencies that run both motions have a structural advantage here, because the qualification work done for one side is already done for the other.
A filter that works
- Ingest postings continuously, not on a weekly report. The freshness window is short.
- Discard anything over 60 days old or lacking specifics.
- Require a threshold, not a posting — five or more of a role, or a first-ever role, or headcount growth above 10% in 90 days.
- Stack with a second signal before acting. One signal runs at roughly a 20% true-positive rate; two inside 30 days reaches 50–60%.
- Extract the named tools from the description and store them as technographic data. This compounds across every posting you process.
Frequently asked questions
Are job postings a reliable buying signal?
Not on their own — they correlate with purchase at only +7%. They become reliable when combined with headcount growth (+38%) or a second signal type inside a 30-day window. Treat a posting as context that sharpens another signal rather than as a trigger in itself.
How do we spot a ghost job?
Age is the strongest tell — anything live beyond 60 days is suspect. After that, look for generic descriptions with no named tools and no named hiring manager, and cross-check whether headcount actually moved. With 40% of hiring managers admitting to a fake posting, filtering is not optional.
Should we reach out as soon as a role is posted?
Yes, but not with a pitch. You are typically 60–90 days ahead of vendor research, so there is no evaluation to win yet. Lead with what companies in the same position usually run into, and plan to still be in the conversation when requirements are actually written.
Is job data better than intent data?
It is earlier and less commoditised. Aggregate intent is sold to every competitor at once; job postings are public, so the advantage comes from reading them more carefully rather than from exclusive access. The tool names inside descriptions are where most of that advantage sits.
Which hire is the strongest single signal?
A first-of-its-kind senior role. A first CISO, first RevOps lead or first Head of Data creates a buyer with a mandate, a budget and roughly 90 days to establish their stack before priorities shift elsewhere.
The takeaway
A job posting is a company telling you, in public and in writing, what it has decided to fix and how much it cares. Very little else in B2B data is that candid or that early.
The mistake is treating it as a volume signal. Counting postings gets you a +7% correlation and a lot of wasted outreach against roles nobody intends to fill. Reading them — the tools named, the thresholds crossed, the role that never existed before — is where the value is, and it costs nothing but attention.
For the wider signal framework this fits into, see B2B buying signals in 2026. For the targeting layer underneath it, start with what sales intelligence actually is and how lookalike company search works.
Request a demo and we will run a live search against your best-fit account profile. Or explore the Recruitment Bundle, whose Job Search module doubles as a sales intent feed, and read more in our AI for Sales & Recruiting topic hub.