The aggregator republishing problem: why one role shows up on several boards
The same remote role turns up on more than one board. That is not an accident on any of them, and the mechanism is documented by the boards themselves: one collects postings from other people's platforms, another sends applicants back with tracking attached, a third says it takes postings directly from employers. The reader pays for this in one specific way, which is that a listing on a board is a listing on a board. It carries the board's view of the role, not the employer's, so the page has less on it to check the employer against. This piece measures what each board publishes, quotes what each board says about its own sourcing, and shows what our own catalogue does instead. What it does not do is tell you who copied which listing, because that needs a crawl we did not run.
The boards publish their own rules about it #
This is the part that surprises people. The copying is not hidden, it is documented, and in one case it is forbidden in writing by the board doing the collecting.
Himalayas, on its own terms: "Save time by pasting a job URL from your ATS. We automatically extract job details from Greenhouse, Lever, Workday, SmartRecruiters, and other popular platforms." (https://himalayas.app/post)
Remote OK, on its own terms: "Candidates are sent to your external URL (ATS, Greenhouse, Lever, your own careers page, etc.) with UTM tracking so you can see where the traffic came from." (https://remoteok.com/faq)
Remotive, on its own terms: "Enter the URL of your public job page. Alternatively, you can provide an email address. We'll redirect applicants directly to you there." (https://remotive.com/post-a-job)
We Work Remotely, on its own terms: "We don't scrape listings from other platforms, we have the best & biggest remote community in the world, and we fill 90% of the positions posted." (https://weworkremotely.com/post-a-remote-job)
Jobgether, on its own terms: "Hidden Job Market: Companies hiring before roles go public" (https://jobgether.com/post-a-job)
Read those together. They describe three different supply chains between them. One board tells its readers that the postings were extracted from other people's platforms, which is collection by scraping. Another tells employers that applicants are sent back to them with tracking attached, which is the business model. A third says it does not scrape at all and takes postings from employers directly. None of them is hiding the arrangement, and none of them is claiming the posting is theirs. What a reader gets is the same in every case: a listing on a board, carrying the board's view of the role rather than the employer's.
What you can check for yourself #
- Whether the employer published it: open the employer own career page and look for the role there. If it is there, that page is the employer version, and the board listing is a re-listing of it.
- Which clock the date is on: compare the date the board shows with the date on the employer own posting. A board date tells you when the board last had it, which is not the same thing.
- Whether the pay survived: compare what the board shows with what the employer own posting says. Where the two disagree, the employer posting is the one that counts.
- Who the click is for: an apply link that carries the board own parameters is a tracked click, and it tells the board that your application came from them.
- Whose wording it uses: a posting that has been reformatted carries the board's phrasing. The employer own title is usually the first thing to check.
The scale, measured rather than asserted #
| Board | What it publishes | Whose postings those are | What it adds that the employer did not |
|---|---|---|---|
| Remotive | 141,931 jobs | employers' postings, listed by the board | the employer own posting date, the employer own pay, or anything a reader could check the posting against |
| Himalayas | 118,595 jobs | employers' postings, listed by the board | the employer own posting date, the employer own pay, or anything a reader could check the posting against |
| Jobgether | 98,202 offer URLs | employers' postings, listed by the board | the employer own posting date, the employer own pay, or anything a reader could check the posting against |
| Remote OK | 58,115 job URLs | employers' postings, listed by the board | the employer own posting date, the employer own pay, or anything a reader could check the posting against |
| We Work Remotely | 506 job URLs | employers' postings, listed by the board | the employer own posting date, the employer own pay, or anything a reader could check the posting against |
| Dracon Jobs | 15,664 listings | read from 1,099 employer career pages (our 2026-11-16 catalogue) | nothing: the employer date, the employer pay and the employer wording are the ones we publish |
Between them the boards above publish 417,349 listings, which is 27 times the 15,664 we hold. We are not claiming ours is better. We are claiming it is checkable, and checkability is the thing a re-listing cannot be.
What our own catalogue shows about the same problem #
| Measure | Figure | What it means |
|---|---|---|
| Listings that repeat an employer and title already in the catalogue | 0 | no posting of ours is listed twice |
| Groups where one employer used the same posting text for several roles | 1,028 | a template, not a copy: the employer wrote it once and reused it |
| Groups where the same posting text appears under two different employers | 31 | 70 listings, most likely a shared agency template |
| Listings that arrived from another job board | 0 | the sourcing rule, enforced by the build gate on every row |
The third row is the honest one. We do find the occasional posting text that appears under two employers, 70 listings out of 15,664. The most likely explanation is a shared agency template or two employers running the same recruitment ad, but we did not check the other explanation, so we are not going to call it one. It is a small number, and the catalogue it was measured in holds nothing but postings read from employers' own boards. It is published here because a piece that claimed our catalogue has no duplication at all would be claiming more than we checked.
How to tell an original from a re-listing #
- Open the employer own career page and look for the role there. If it is there, the employer version is the original and everything else is a re-listing.
- Check the date on the posting. If the date belongs to the board rather than the employer, you are reading a re-listing.
- Check whether the pay is stated. Where a board listing drops it, the employer own posting is the one to read.
- Apply on the employer page, not the board's, so nobody is tracking your application between them.
Our catalogue figures describe the 2026-11-16 refresh and the board inventories were read on 2026-11-18. Both dates are printed because both move.
Sources & provenance #
- Our catalogue: 15,664 listings, none of them sourced from another board (refresh 2026-11-16)
- Remotive, its own published inventory (read 2026-11-18)
- Remotive, its own statement about where its postings come from (read 2026-11-18)
- Himalayas, its own published inventory (read 2026-11-18)
- Himalayas, its own statement about where its postings come from (read 2026-11-18)
- Jobgether, its own published inventory (read 2026-11-18)
- Jobgether, its own statement about where its postings come from (read 2026-11-18)
- Remote OK, its own published inventory (read 2026-11-18)
- Remote OK, its own statement about where its postings come from (read 2026-11-18)
- We Work Remotely, its own published inventory (read 2026-11-18)
- We Work Remotely, its own statement about where its postings come from (read 2026-11-18)
- Himalayas, its own API terms on redistribution