Greenhouse Job Change Monitor — New, Changed & Closed Posts

Follow public Greenhouse job boards over time. Receive current job posts with baseline, new, changed, unchanged or reopened labels, plus confirmed closed observations. Each run saves a snapshot for the next comparison.

Built and maintained by JAFF Consulting B.V., Netherlands.

Quick start

{
  "boards": ["greenhouse"],
  "stateStoreName": "my-greenhouse-monitor",
  "maxJobsPerBoard": 5000
}

1. Enter company board tokens such as greenhouse, or their Greenhouse board URLs.
2. Choose a snapshot store name. Keep it unchanged for later runs.
3. Run once to establish the baseline, then run again or use an Apify schedule.
4. Export CSV, Excel or JSON, filtering the change column as needed.

The Actor reads the official public Greenhouse Job Board API. It accesses published posts only; it does not access applicant data, apply for jobs or require a Greenhouse API key.

Change labels

Label	Meaning
baseline	A job present during the first successful observation of that board.
new	A post ID absent from the previous successful snapshot.
changed	The title, location, departments, offices, description, URL, language or public metadata changed. changedFields identifies the fields.
unchanged	A current post with the same material fields. A source timestamp change alone is ignored.
reopened	A previously confirmed missing post ID appears again.
closed	The post is missing from two complete, successful observations at least one hour apart.

closed means the post disappeared from the public board. It does not establish whether someone was hired or the role was cancelled. A source that serves an incomplete but internally consistent response twice can still cause a false closure; source truth cannot be guaranteed. Pending missing posts are counted in RUN_SUMMARY and are not yet labeled closed.

Reliable comparisons and limits

Failures, malformed responses, duplicate IDs, total-count mismatches, oversized boards and insufficient spending allowance do not replace that board's previous snapshot. Boards are processed independently. A budget or timeout interruption can leave delivered rows without committing the new baseline; rerunning may repeat change labels. Check RUN_SUMMARY for successful boards.

Run each snapshot store sequentially. Do not overlap runs using the same store name. The Actor checks for a changed snapshot before writing, but storage is not a transactional lock. Use a separate store name for independent monitors. Different stores establish separate baselines. Deleting the named store resets history. Closed posts are retained for 90 days; a post returning after that may be labeled new.

Up to 50 boards and 20,000 posts per board are supported. The Actor compares complete boards; it never truncates a board and treats omitted rows as closures. All current posts are delivered, including unchanged ones. Each output includes the post ID, title, location, description, source URL, first/last observed times and material change fields. A per-run OBSERVATION-<board> record makes the fetched snapshot available in the default key-value store.

The Actor stops safely before the run timeout and after at most 30 minutes. Named snapshot storage persists and consumes your Apify storage allowance. Schedule frequency is your choice; daily runs are a practical starting point. At least one hour between missing observations is required for a closure.

Pricing

$1.50 per 1,000 current job posts processed ($0.0015 each), plus the Actor-start fee shown in the Pricing tab. Baseline and unchanged posts are billable because they are fetched and compared. Confirmed closure records and summary reports have no separate result fee. Repeated runs process and bill the current posts again. The Pricing tab is authoritative; there is no monthly rental fee for this Actor.

Support

Open an Apify issue with the board token and the relevant RUN_SUMMARY error. Keep credentials and private applicant information out of issues. Source formats and availability can change; use the results as timestamped observations.