Short answer:every job posting starts in the company's own applicant tracking system (ATS) — Greenhouse, Lever, Ashby, Workday, and a handful of others cover most of the market. Job boards don't see that posting instantly; they either crawl the company's careers page on a delay (Indeed's own documentation says 24–72 hours for a standard crawl) or rely on the company setting up a feed at all. A meaningful number of real, current postings are sitting on a company's own site right now and simply haven't propagated anywhere else yet.
The source of truth is the ATS, not the board
When a recruiter opens a req, they create it in whatever applicant tracking system their company runs — that's where the posting is first written, edited, and eventually closed. Everything downstream of that — the company's own careers page, LinkedIn, Indeed, ZipRecruiter, every aggregator — is a copy, syndicated out from that original record, on whatever schedule the integration between the ATS and that destination runs on. The ATS posting is always the freshest version of the truth. Everywhere else is a snapshot that's at least a little stale by definition.
This matters more than it sounds like it should, because most job seekers search in the opposite direction — starting on a board and treating it as the primary source, when it's actually the last stop in a chain that starts somewhere else entirely.
It also explains a specific, common frustration: refreshing a job board every day looking for new postings from a company you want to work at, and seeing nothing change for weeks, only to later discover the company posted and filled two roles in that window without either ever showing up on the board you were watching. That's not the board failing to update — from its perspective, nothing changed, because it never received that data in the first place. The company's own careers page, sourced straight from its ATS, would have shown both roles the moment they opened.
Why syndication is slow, and sometimes doesn't happen at all
Indeed's own partner documentation is specific about this: an organic crawl of a standard public careers page typically takes 24 to 72 hours to show up. If a company sets up an XML feed for Indeed to crawl on a schedule, or hosts one on an FTP/SFTP server Indeed pulls from, that data refreshes four times a day — a real improvement, but still not instant, and still dependent on the company having bothered to set it up. Indeed's documentation is explicit that feeds and APIs exist specifically because organic crawling alone is too slow and incomplete on its own.
Not every company sets up that integration. Smaller companies, fast-moving startups, and any employer whose careers page uses a structure a given crawler doesn't parse cleanly can end up with postings that either take the full delay window to appear, or never reliably syndicate at all. None of that is a conspiracy — it's just plumbing, and plumbing has gaps.
Walk through the actual sequence and the delay stops being abstract. A recruiter opens a req in Greenhouse at 9am Monday. It's live on the company's own careers page within minutes, because that page is usually just rendering the ATS's own data directly. If the company has a live feed integration, Indeed picks it up at the next six-hour refresh — worst case, by early afternoon. If it's relying on organic crawling instead, Indeed's own stated range means it might not appear until Tuesday or Wednesday. A board that syndicates from Indeed rather than crawling the company directly adds its own refresh cycle on top of that. By the time a listing has propagated through two or three hops of this chain, a role posted Monday morning can plausibly still be invisible on some boards by Thursday — while sitting, fully current, on the company's own page the entire time.
Why the same listing sometimes looks different in two places
The flip side of slow syndication is inconsistent syndication. Because most boards pull from multiple upstream sources — direct feeds from some companies, crawled snapshots from others, and in some cases other aggregators re-syndicating what they collected — the same underlying posting can show up more than once with slightly different details: a salary range present in one copy and missing in another, a posted date that reflects when the board last touched the listing rather than when the company actually created it, or a listing that's been closed on the company's own page but is still live wherever it was copied to before anyone updated the source. This is a large part of the mechanical explanation behind the board-vs-careers-page mismatch signal covered separately — it's rarely deliberate deception on the board's part, just the predictable result of stitching together feeds that update on different schedules from sources that don't coordinate with each other.
The major ATS platforms and how to find them directly
A handful of platforms cover most of the market, though the exact share shifts depending on how you measure it. Looking at raw installations across a broad dataset of companies, Greenhouse, Lever, Workday, and iCIMS each hold somewhere in the 15–20% range. Looking instead at a set of top-rated employers specifically, Greenhouse runs hiring for roughly half, with Workday a distant second. At the largest enterprises — Fortune 500 scale — Workday leads clearly. The honest summary: there's no single dominant platform across every company size, but a short list covers the large majority of postings.
| Platform | Public URL pattern | Native search? |
|---|---|---|
| Greenhouse | boards.greenhouse.io/company-name | Yes |
| Lever | jobs.lever.co/company-name | Yes |
| Ashby | jobs.ashbyhq.com/company-name | Yes |
| Workday | company.wd#.myworkdayjobs.com | Yes |
Each of these reads the platform's public job-feed endpoint directly — no login, no API key. If you know or suspect a company uses one of these, you can go straight to its board and see the current listing in real time, rather than waiting on however long that specific board takes to pick it up.
Ashby is worth calling out specifically because it's the fastest-growing of the group — reportedly expanding from around 1,300 to over 2,700 customers in roughly a year, alongside a $50M funding round in 2025 — which matters practically because it skews toward newer, faster- growing companies that are exactly the ones worth watching if you're targeting growth-stage roles. A board that's barely a year old on a platform still gaining market share is more likely to be tightly maintained and less likely to have accumulated years of stale, unclosed postings than an older, larger installation.
Outside this short list, a long tail of smaller or vertical-specific systems exists — BambooHR, Rippling, JazzHR, and others, more common at smaller companies without a dedicated recruiting function. Those don't follow one predictable URL pattern the way the four platforms above do, which is exactly the kind of long-tail coverage problem that makes a single manual technique stop scaling once you're searching across company sizes rather than a specific target list.
The manual method, and where it breaks down
For a company you already have in mind, this is genuinely fast. A three-step version:
- Search "[company name] careers" and look at the actual URL in the result — if it redirects to a greenhouse.io, lever.co, ashbyhq.com, or myworkdayjobs.com domain, you've found the board.
- If that doesn't turn it up, try the URL pattern directly: boards.greenhouse.io/company-name and jobs.lever.co/company-name cost nothing to try and load instantly if they exist.
- Once you're on the board, use its native search rather than scanning the full list — Greenhouse and Ashby both support searching by keyword or phrase across the job title and description, which is faster than reading every open req by eye.
A more targeted variant works from a search engine directly: searching a role keyword together with site:boards.greenhouse.io or site:jobs.lever.cosurfaces postings across every company on that platform that match your term, without needing to know which specific companies use it. It's a genuinely useful technique — and also a good illustration of why it doesn't scale into a full search strategy: it only searches one ATS platform at a time, returns results with no fit signal against your specific resume, and requires you to already know or guess which platforms are worth checking.
Where it breaks down, more broadly, is scale. The URL-guessing and site-search techniques above work well for the five or ten companies you already have a specific interest in. Neither works as a general job-search strategy across an entire market or role — there's no practical way to manually check hundreds of individual company boards across a dozen ATS platforms every day, notice which ones posted something new, and compare that against your resume. That's the exact gap between "a good technique for a target-company list" and "a way to search the whole job market," and it's the reason aggregation exists as a category at all — the manual version simply doesn't scale past a handful of companies.
Where this leaves board-based search
None of this means boards are useless — most hiring still happens through postings that do get syndicated eventually, and the broader hiring data reflects real activity across the whole market, board-listed or not. It does mean two specific things worth adjusting: don't assume a board is missing nothing just because your search came back empty, and don't assume a stale-looking listingon a board is the company's current truth — the company's own ATS board, if you can find it, is a more current source than almost any aggregator will be.
A reasonable hybrid approach: use boards for breadth — casting a wide net across a role or location you haven't narrowed down yet — and switch to the direct-ATS technique once you've identified a specific shortlist of target companies. Boards are genuinely good at the first job; they're structurally worse at the second, and no amount of refreshing a search page will fix a gap that's upstream of the board entirely.
Where CareerFlint fits
This is the specific gap CareerFlint's job searchis built to close at scale: postings sourced directly from company career pages — reading the same ATS-native endpoints described above — rather than relying purely on secondary aggregation. That doesn't mean every listed job is instantaneous or that syndication delay disappears entirely; it means the manual technique in this article, which works well for five companies, is applied systematically across many more than any one person could realistically check by hand.
The other half of that gap — knowing which of the postings you do find are actually worth your time — isn't solved by sourcing alone. That's a fit question, not a freshness question, and it's why postings on CareerFlint are also ranked against your specific resume rather than presented as an undifferentiated list: finding a role that exists is only useful if you can also tell, quickly, whether it's one worth applying to.