By Robert Fon / October 21, 2026
Most sales teams still build target account lists from static firmographics: industry code, headcount band, and an estimated revenue range. None of those tell you which companies are expanding right now, which is the moment a prospect is most open to a new vendor conversation. The PredictLeads Job Openings dataset turns live job postings into a structured, queryable signal of company growth. This guide covers what the dataset contains, how its fields work, and how to turn hiring activity into a repeatable sales growth workflow.
TLDR:
- The PredictLeads Job Openings dataset tracks postings from company career pages, ATS integrations, and LinkedIn profiles, then classifies every role with an O*NET occupation code.
- Hiring activity works as a growth signal because companies tend to add headcount when they expect demand and freeze it when they do not.
- Job Openings data is strongest when layered with Technology Detections, Financing Events, and News Events, since any single signal is noisy on its own.
- Every record carries first_seen_at, last_seen_at, and a status field, so you can time outreach to the moment a role appears rather than weeks later.
- PredictLeads has logged 279.2M+ job opening records since 2018 across 2.9M+ websites, with 10.2M active openings at any time and 710,600+ companies currently hiring.
What Is the PredictLeads Job Openings Dataset
The PredictLeads Job Openings dataset is a structured feed of job postings sourced from company career pages, career subpages, ATS integrations, and LinkedIn company profiles, with historical coverage back to 2018. Every posting is matched to a company record, deduplicated, and classified with an O*NET occupation code, so a “Senior Account Executive” posting from one company and a “Senior AE” posting from another land in the same queryable category. The dataset spans 26 job categories, from sales and engineering to operations and support, with seniority tracked from entry level through c_level.
That scale, 279.2M+ historical records across 2.9M+ websites, is what makes hiring data usable as a repeatable sales signal rather than a handful of anecdotal postings.
What’s Inside Every Job Opening Record
Every record in the Job Openings dataset carries the same structured fields, whether the company is a two person startup or a public enterprise. That consistency is what makes the dataset queryable at scale instead of requiring you to parse free text postings yourself.
- Role detail: title, normalized_title, translated_title, description, and categories (26 job categories).
- Classification: onet_data (code, family, occupation_name) and a seniority field spanning levels from not_set and junior through director and c_level.
- Timing: first_seen_at, last_seen_at, last_processed_at, posted_at, and a status field (closed or null).
- Compensation and format: salary (raw) and salary_data (low, high, currency, USD value, time unit), plus contract_types (for example, full time, remote).
- Location and contact: location, location_data, and recruiter_data (name, title, contact) where the posting includes it.
Workplace type (remote, hybrid, onsite) is carried inside contract_types only when a posting states it, and many do not. In a PredictLeads analysis of 677,699 US technical job postings from July 18 to October 16, 2026, 46.3% carried no explicit workplace-type tag at all. Filtering only on that tag will miss a large share of otherwise matching roles, so pairing it with title and description keyword matching gives a more complete picture. Every role is also classified against the Department of Labor’s O*NET occupational taxonomy, the same standard used across US labor market research.
Job Openings Data vs. Firmographic Headcount Estimates
Firmographic headcount data answers one question: how big is this company. It is a useful baseline for segmentation, but it is usually refreshed on a slower cycle and gives you a range, not a timestamp of when the company actually started growing. PredictLeads Job Openings data answers a sharper question: where and what is this company staffing right now. The difference matters for timing an outreach motion, not just qualifying an account.
| Dimension | Firmographic Headcount Data | PredictLeads Job Openings Data |
|---|---|---|
| Update cadence | Refreshed on a slower, periodic cycle | Refreshed on an approximately 36 hour cadence |
| Granularity | Company wide headcount range | Role level: title, O*NET code, seniority, category |
| Timing signal | No timestamp on when growth started | first_seen_at and last_seen_at on every record |
| What it tells you | How big a company is | Where and what a company is actively staffing right now |
| Best use | Firmographic segmentation and list building | Timing outreach and prioritizing expanding accounts |
That timestamped, role level view is also why hiring activity is increasingly treated as a form of alternative data for company intent rather than a simple headcount metric.
Why Hiring Is a Sales Growth Signal
Economists have treated job openings as a leading indicator of growth for decades. The U.S. Bureau of Labor Statistics publishes national job openings totals every month through its Job Openings and Labor Turnover Survey (JOLTS), precisely because employers tend to add roles when they expect demand and pull back when they do not. Research groups such as Indeed Hiring Lab track the same posting activity at a market level as a proxy for labor market momentum.
The same logic holds at the company level, with more precision. A company posting a single replacement role is not a strong signal. A company posting five sales engineer roles in the same quarter, on the same team, is more likely allocating new budget toward go-to-market expansion. That is the distinction PredictLeads’ seniority and onet_data fields let you make. For a deeper breakdown of separating backfill hiring from net-new growth, see our guide to hiring signals as company growth indicators.
Four Ways to Use Job Openings Data for Sales Growth
Turning hiring data into a sales growth motion is a four step process. The first two steps find and score the right accounts; the last two time and personalize the outreach. For a full walkthrough, see our guide to job openings data for sales prospecting.
Steps 1 and 2: Find and Score the Right Accounts
- Define the trigger role. Use onet_codes on the discover/job_openings endpoint to pull every company hiring for a role that signals your buying trigger, such as a first RevOps hire or a demand generation manager.
- Score hiring velocity. Compare a company’s count of active openings against its 90 day found_at history to separate a one off backfill from a genuine team expansion.
Steps 3 and 4: Time and Personalize the Outreach
- Time the outreach to first_seen_at. Webhooks can push new JobOpeningsDataset records as they are found, so a rep can reach out within days of a role appearing instead of weeks later.
- Personalize with the posting itself. Reference the specific skill, tool, or responsibility named in the job description, and use recruiter_data where available to route the message to the right contact.
Stack Job Openings With Other Signals for Higher Precision
A single hiring signal is directional. Three stacked signals are a pattern. Consider a company that posts three account executive openings at the mid_senior level, shows a recent technology detection tied to a CRM platform sourced from a job description, and logged an increases_headcount_by News Event in the same quarter. None of those three facts alone confirms a go to market expansion, but together they are stronger supporting evidence of one than any single record.
This is also where PredictLeads’ breadth matters more than any one dataset. See how hiring data pairs with technographics in Job Openings Data and Technographics, and how the two combine into a single buying signal in Job Postings Data for GTM Buying Signals. PredictLeads exposes Job Openings, Technology Detections, News Events, and Financing Events in one API account, so you query one source instead of reconciling exports from several vendors.
Why Job Openings Data Goes Stale, and How PredictLeads Keeps It Fresh
A target list built from job postings is only as useful as it is current. Roles close, get filled internally, or get reposted on a different board days later, and outreach aimed at a closed role reads as careless rather than well timed. PredictLeads addresses this with an approximately 36 hour refresh cadence and multiple crawls per day on high traffic career pages. Every record’s status field flips to closed (or stays null while open), and last_seen_at records the last confirmed sighting, so closed roles are marked rather than silently dropped. The pace of change is real: PredictLeads detected 6.4M+ new job openings in the most recent month and 57.2M+ in the trailing year.
How to Access the PredictLeads Job Openings Dataset
How you pull Job Openings data depends on whether you consume it through a tool or build it into your own pipeline. Data consumers querying one account at a time typically use the REST API. Data builders loading a full warehouse, or teams that need real time alerts, use flat files, webhooks, or the MCP server instead. All four methods return the same structured fields; they differ in latency and volume. Full parameters are documented at docs.predictleads.com, including the job openings data API guide for common query patterns.
| Delivery Method | How It Works | Best For | Typical Use Case |
|---|---|---|---|
| REST API | Query by company domain or discover endpoints with your API key and token | Real time lookups and enrichment at request time | Checking whether one target account is hiring before a call |
| Flat Files | Scheduled JSONL exports delivered to S3, Google Cloud Storage, or SFTP | Bulk warehouse loads and historical backfills | Loading the full history into Snowflake or BigQuery for modeling |
| Webhooks | Push notifications when a followed company’s JobOpeningsDataset changes | Real time trigger workflows and alerting | Notifying a rep the same day a target account posts a role |
| MCP Server | Natural language queries through mcp.predictleads.com | AI agent workflows and conversational access | Asking an agent “is company X hiring SDRs” directly |
How PredictLeads Delivers Job Openings Data at Scale
PredictLeads has logged 279.2M+ job opening records since 2018 across 2.9M+ websites, with 10.2M active openings at any time (9.6M carrying a full description) and 710,600+ companies currently hiring. That scale is maintained alongside PredictLeads’ other datasets in the same API account: Technology Detections (2B+ detections since 2018, 240,000+ technologies, 99M+ companies with technology data), News Events (10M+ signals since 2016), Financing Events (210,800+ events since 2016), and Connections (371.8M+ relationships since 2019), among others.
All of it is sourced from public information only, SOC 2 Type II certified, and handled under GDPR and CCPA. Explore the dataset on the Job Openings dataset page, or pull it directly via API, flat files, webhooks, and MCP.
Final Thoughts on the Job Openings Dataset
Hiring data will not replace firmographic segmentation or intent signals, and it should not try to. What it adds is timing: a timestamped, role level view of where a company is actively investing, refreshed roughly every 36 hours. Used alongside Technology Detections, News Events, and Financing Events, it turns a static target list into a ranked one. For a wider comparison of vendors in this space, see our review of the best job posting and hiring intent data providers for 2026.
Ready to see this in your own data?
Get 100 free API requests per month – no credit card, no sales call.
Frequently Asked Questions
What is the PredictLeads Job Openings dataset?
The PredictLeads Job Openings dataset is a structured, timestamped feed of job postings collected from company career pages, ATS integrations, and LinkedIn profiles, with history back to 2018. Every posting is normalized and classified with an O*NET occupation code, a seniority level, and one of 26 job categories, so you can query by role instead of parsing free text. As of October 2026, the dataset holds 279.2M+ historical records across 2.9M+ websites, with 710,600+ companies currently hiring. You can pull it by company through the API, in bulk through flat files, or as push notifications through webhooks.
How often is PredictLeads Job Openings data refreshed in 2026?
PredictLeads refreshes job openings on an approximately 36 hour cadence, and high traffic career pages are crawled multiple times a day. Every record carries first_seen_at and last_seen_at timestamps, so you can tell exactly when a role appeared and whether it is still listed. In the trailing 12 months, PredictLeads detected 57.2M+ job openings, including 6.4M+ in the most recent month alone, which shows the pace of change the refresh cadence is built to capture. Webhooks can also push new JobOpeningsDataset records to your system as soon as they are found, rather than waiting for a scheduled pull.
Can I filter PredictLeads job openings by O*NET code or seniority level?
Yes. Every job opening record includes onet_data (code, family, and occupation name) plus a seniority field spanning levels from not_set and junior through director and c_level, so you can target a specific function and rank in one query. The discover/job_openings endpoint accepts onet_codes directly, alongside location and active_only filters, which lets you find every company hiring for a given role without looking up companies one at a time. This is the mechanism GTM teams use to build lists like “companies hiring a Head of Revenue Operations” without manual research. Postings are also grouped into 26 job categories, from sales and marketing to software development and operations, for broader filtering.
How is Job Openings data different from firmographic headcount estimates?
Firmographic headcount data gives you a company wide size range that is typically refreshed on a slower, periodic cycle, so it tells you how big a company is, not what it is doing right now. PredictLeads Job Openings data is role level and timestamped: it shows the specific title, O*NET code, and seniority of every open position, updated on an approximately 36 hour cadence. That granularity is what lets a sales team distinguish a company backfilling one role from one opening five new seats on the same team in the same month. PredictLeads tracks 10.2M active openings at any given time, which is the live layer that headcount estimates alone do not provide.
How do I access PredictLeads Job Openings data via API, flat files, webhooks, or MCP?
PredictLeads offers all four standard delivery methods for Job Openings data. The REST API (GET /companies/{domain}/job_openings or GET /discover/job_openings) suits real time, per company lookups; flat files (typically JSONL, delivered to S3, Google Cloud Storage, or SFTP) suit bulk warehouse loads into Snowflake or BigQuery; webhooks push JobOpeningsDataset events as new postings are found; and the MCP server at mcp.predictleads.com lets AI agents query job openings in natural language. Full endpoint documentation, including filters like categories, seniority, and first_seen_at_from, is at docs.predictleads.com.
Pingback: Top 5 Job Data Providers in October 2026