Workday Public Careers Monitor: New Openings on Large Employers' Workday Sites

Large employers publish their openings on Workday-hosted career sites, and those sites list thousands of roles. This actor reads the JSON that the Workday site itself loads, remembers the requisition ids it has already returned, and gives you only the openings that are new since the last run.

Who it's for

A competitive-intelligence analyst at a market-research firm. Your client sells into five large manufacturers and wants a monthly note on where each one is adding headcount. Counting the whole board each time is useless at that scale; the new requisitions per employer per month are the story.

A university career-services office. Local employers on Workday post graduate-scheme and internship roles throughout the year. Students want a short list of what appeared this week, not a dump of everything the site holds.

A staffing-agency account manager. One enterprise client is the account. A new requisition in a function you can fill is a reason to call before the client's own recruiters have finished their first screen.

All three track a handful of big sites over time and care about arrivals. If you only need a one-off count of a single site, the lookup actor in the related list is a cheaper fit.

Why this one

• No browser. It posts directly to the site's wday/cxs JSON endpoint and pages through the results 20 at a time, so a run is fast and light.
• A stable id per opening. The requisition id Workday shows on each posting, such as JR2027031, is the key the stored record uses.
• Only arrivals are billed. Billing is per posting returned. A site with nothing new returns nothing.
• Failed sites cost nothing. A bad address becomes a visible found: false row, never a charge.

What you get

Field	Meaning
job_id	The requisition id Workday shows in the posting's bullet fields.
title	The job title.
location	Workday's location text, for example "Israel, Yokneam".
posted_on	Workday's relative wording, such as "Posted Today" or "Posted 2 Days Ago". It is a label, not a date.
external_path	The path of the posting on the site.
url	The full link to the posting.
company_board_id	tenant/site for the row.
first_seen_timestamp	When this monitor first recorded the opening. This is the date to sort and alert on.

Because posted_on is relative, use first_seen_timestamp for anything that needs a real date.

Finding the right Workday address

Workday addresses have three parts that matter, and getting one wrong is the most common reason a site returns nothing.

• Tenant. The first label of the host. In nvidia.wd5.myworkdayjobs.com the tenant is nvidia.
• Data centre. The wd1, wd3, wd5 part. It differs between employers and is part of the real address.
• Site. The first path segment that is not a language code. For .../en-US/NVIDIAExternalCareerSite the site is NVIDIAExternalCareerSite.

Paste the full careers URL. With the whole address, the actor takes the host as given and every part is right. On 2026-10-08 https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite and the same address with /en-US/ in it both answered with a site size of 2,000 postings.

The tenant/site shorthand has a limit. It always assumes the wd1 data centre. We tested nvidia/NVIDIAExternalCareerSite on 2026-10-08 and Workday answered HTTP 422, because NVIDIA is on wd5. The row comes back as ERROR and is not billed. Use the shorthand only for employers you know are on wd1; otherwise paste the URL.

Pricing

• Job posting returned: $1 per 1,000 job postings

Plus a $0.001 start fee per run. Each event above is billed independently, only when it actually returns data — misses (found:false) are never charged.

Size matters here more than on smaller platforms, so a big site is never read in full unless you ask for it. Each site is read newest first, up to maxPostingsPerBoard postings: 200 by default, 2,000 at most. At the listed price, a first run on a large site costs at most about $0.20 with the default, and about $2.00 if you raise the cap to 2,000. Seed with skipFirstRun and the backlog costs nothing.

How the actor learns a site's size. Workday states the number of postings only in the answer to its very first request and sends total: 0 with every page after that. On 2026-10-08 we saw exactly this on the NVIDIA, Salesforce, Adobe and Pfizer sites. The actor therefore keeps the figure from that first answer and asks for the next 20 postings, one request at a time with a 150 ms pause, until it holds maxPostingsPerBoard postings or the site has none left. On the NVIDIA site the pages ran from "Posted Today" at the start down to "Posted 29 Days Ago" by offset 1000, so the postings that make the cut are the most recent ones. Workday itself answers an offset of 2,000 with the first page again, which is why the cap stops there. If Workday replies 429 or a 5xx error after the first page, the run stops and delivers what it has read so far; only delivered postings are recorded, so the rest are still new next time.

Keywords and the cap. includeKeywords and excludeKeywords are applied after the read, to the postings that were fetched. With includeKeywords set you get the matches among the newest maxPostingsPerBoard postings of each site, not among every posting it holds, and an older matching role is not found. Raise the cap if the filter has to reach further back.

How to use

1. In the Apify Console. Open the actor page and click Start — the company_board_id field is already pre-filled with a working example. Results land in the run's dataset as soon as each item is found.
2. Via the API. Call it directly with a POST request — no Console needed once you have an API token:

   curl "https://api.apify.com/v2/acts/accountable_eel~workday-public-careers-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
     -X POST \
     -H "Content-Type: application/json" \
     -d '{"company_board_id":["https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite"]}'
   

3. On a schedule. Save this actor as an Apify Task with the input you want, then add a Schedule (hourly, daily, weekly) so it runs on its own — no server of your own required.

Input

{
  "company_board_id": [
    "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite"
  ]
}

Full Workday career site URL (e.g. "https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite"). The short tenant/site form only works for sites on the wd1 data centre. Charged per job posting returned. Boards with no result are free. Accepted formats: https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite, nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite.

Other settings: maxPostingsPerBoard (how many of a site's newest postings to read, 200 by default), includeKeywords and excludeKeywords (matched case-insensitively against title and location), maxResults, testRun (first five sites), onlyFound, deltaMode (on by default), deltaName and skipFirstRun.

Sample output

query	found	status	job_id	title	location	posted_on	external_path	url	company_board_id	first_seen_timestamp	scrapedAt
https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite	true	OK			<location>						1970-01-01T00:00:00.000Z

A site whose endpoint returns 404 is reported as NOT_FOUND; any other failure, including the 422 above, as ERROR. Neither is billed.

Use it from Clay, n8n, Make, or an AI agent

This actor runs synchronously over plain HTTP — call it directly from a script, a workflow tool, or an AI agent, no Apify Console needed once you have an API token.

curl "https://api.apify.com/v2/acts/accountable_eel~workday-public-careers-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{"company_board_id":["https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite"]}'

n8n. Add an HTTP Request node: Method POST, URL https://api.apify.com/v2/acts/accountable_eel~workday-public-careers-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>, Body Content Type JSON, JSON Body {"company_board_id":["https://nvidia.wd5.myworkdayjobs.com/NVIDIAExternalCareerSite"]} (swap in an expression from an earlier node for a real value).

Clay. Add an "HTTP API" column: Method POST, URL https://api.apify.com/v2/acts/accountable_eel~workday-public-careers-monitor/run-sync-get-dataset-items?token=<YOUR_TOKEN>, Body {"company_board_id":["{{job posting}}"]}, mapping the row's job posting into the company_board_id array.

MCP. In Claude, Cursor, or any MCP client with the Apify MCP server, ask for "Workday Public Careers Monitor | Apify" — the agent will find and run this actor.

Tips

• Seed first, then monitor. For Workday sites, skipFirstRun is almost always right. The first run records every requisition that passes your keyword filter among the newest maxPostingsPerBoard postings and bills none.
• Narrow with keywords. A filter such as engineer keeps the dataset readable. The filter runs before the stored record is touched, so a posting it hides stays unrecorded and shows up as new if you loosen the filter later. Requisitions cut by maxResults, or not delivered because the run hit its spending limit, are likewise left out of the record and arrive on the next run. A site that posts more than maxPostingsPerBoard roles between two runs pushes its oldest arrivals out of the window, so schedule more often or raise the cap for very busy sites.
• Reset one site, not all. The record sits in the key-value store workday-delta-store, or in the store you name in deltaName. The key is baseline-<tenant>-<site>; delete only that key to start one site over.
• Keep sites in separate stores when audiences differ. Different deltaName values keep their baselines apart.
• Expect no description text. The list response carries title, location and a path. If you need the full job description, follow url.

vs. alternatives

	What it costs	What you get	Trade-off
This actor (workday-public-careers-monitor)	$0.001 per posting returned, $0.001 per run, nothing when no posting is new	New openings only, keyed on requisition id, with a first-seen date	Workday only, a relative posted label, and a first run that bills up to maxPostingsPerBoard postings (200 by default) unless seeded
Workday Jobs Lookup (workday-jobs-lookup)	Billed per company, listed at $0.004 per site found	A snapshot of the site with remote-only and keyword filters and its own new-jobs mode	Cheaper for one-off pulls and for sites that add four or more postings between runs. Choose it if you want a remote-only filter.
Workday's own JSON endpoint	Free	The same list	You write the paging, the seen-list and the scheduling

The sibling's price is the figure on its page at the time of writing; check it before relying on the comparison.

FAQ

Why do I get nothing for an employer I know uses Workday?

Most often the address is wrong. Open the careers page in a browser, copy the full URL, and paste that instead of a shorthand.

What will a first run on a huge site cost?

At most about $0.20 with the default of 200 postings, because only the newest 200 are read. Setting maxPostingsPerBoard to its ceiling of 2,000 can bring that to about $2.00. Seed with skipFirstRun to pay nothing for the backlog.

Why did a keyword search miss a role I can see on the site?

The keyword filter only looks at the newest maxPostingsPerBoard postings. A matching role older than that window is outside the read; raise the cap to reach it.

What does "Posted 30+ Days Ago" mean?

It is Workday's own wording for older postings. This monitor passes it through unchanged and does not convert it to a date.

Can I watch one location only?

Add the place name to includeKeywords; it is matched against the location text. Postings the filter hides are not recorded, so removing the place name later brings them back as new.

Related actors

• Workday Jobs Lookup is the per-site snapshot with remote-only filtering.
• ATS Jobs Unified Lookup shows which applicant-tracking system a company uses.