# Changelog of Workday Jobs Scraper by Company Name (`deriverge/workday-jobs-scraper`) Actor

- **URL**: https://apify.com/deriverge/workday-jobs-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/deriverge/workday-jobs-scraper.md

##### 2026-10-11

*Features*

- **Companies or career sites** accepts company names such as `NVIDIA`, `Salesforce` or `Bank of America`. Each name is matched to its Workday career site in our directory of more than 4,000 sites, and the log shows the site it became.
- A company domain such as `walmart.com` is matched to its Workday tenant first, before the careers page is checked.
- The `COMPANY_NAMES` record in the run's storage lists each name, the career site it matched and the employer's other Workday sites.

*Changes*

- A name or domain that matches a career site already in the list is read once, so the same jobs are not charged twice.

*Fixes*

- `companyName` loses more of Workday's internal codes: tax numbers such as `52-2185192` in `52-2185192 Banc of America...` and codes mixing letters and digits such as `7Z4` in `7Z4 Pfizer, Inc.`. A number that belongs to the name stays, as in `3M`, `7-Eleven` or `2020 Companies, Inc.`.
- On HTTP 429 the actor pauses (as long as Retry-After says, or 5, 10 and 20 seconds) and slows down; a tenant that keeps refusing is left alone for the rest of the run and appears in `stoppedHosts` in `SUMMARY`.

##### 2026-09-30

*Fixes*

- In new-only mode, jobs beyond **Maximum jobs per company** or beyond the run's maximum cost are sent in the next run. They used to be saved as seen and never returned.
- `visaSponsorship` is `false` for ads saying the job is "not eligible for relocation or company provided sponsorship".
- `companyName` comes from the Workday job page: the legal entity that posted the job, without Workday's internal number, for example `NVIDIA USA`. It used to be empty on every job read from a career site.
- In the daily index search, company names from Workday job pages also lose internal codes that start with letters, such as `IN01` in `IN01 NVIDIA Graphics Bengaluru`.
- The output schema lists the fields that only rows from the daily index search have: `companySlug`, `careerSiteUrl`, `descriptionIncluded`, `indexDate` and `scrapedAt`.

##### 2026-09-28

*Features*

- Leave **Career sites** empty to search every open Workday job in our daily index.
- New fields: seniority, job category, work arrangement, normalized employment type, country codes, pay as numbers with a yearly figure, visa sponsorship, skills, years of experience and education level. Each of them except education level has a filter.
- `language` is detected from the description when the system does not give it, and jobs can be filtered by language.
- Inputs sent under other field names, such as `url`, `startUrls`, `domains`, `query` or `limit`, are understood.

*Changes*

- Job IDs are the requisition number, so two jobs at the same location no longer share a key.
- Without filters, a run stops paging once it has **Maximum jobs per company** jobs from a site.

*Fixes*

- Career sites are read past their first 40 jobs, up to 4,000 per run.
- HTML entities, double-encoded characters and broken UTF-8 are cleaned from titles, locations and descriptions.

##### 2026-09-27

*Changes*

- The watch field is now called `watchName`. Runs that still send `watchKey` keep working.

##### 2026-09-26

*Features*

- Jobs listed under several locations are opened, and every location is returned.

##### 2026-09-24

*Changes*

- Lower price per job on every plan.

##### 2026-09-21

*Changes*

- Higher price per job on every plan.

##### 2026-09-03

*Features*

- First public version: the open jobs of any Workday career site with title, location, posting date and job link, an optional description, and a mode that returns only jobs new since the last run.