Developer guides

How to find which ATS a company uses (and when it switched)

The fastest tell is the hostname of the apply link: boards.greenhouse.io, jobs.lever.co, jobs.ashbyhq.com, *.myworkdayjobs.com and a dozen others identify the system without any scraping of the page body. Source: Arcadium Data, data as of 2026-10-25.

Sales teams selling to recruiters, vendors defending a base, investors sizing a market and job seekers wondering how their CV is parsed all ask the same question: which ATS does this company use? Search engines answer it with resume-advice pages that are often wrong; the real signal is in the company's own job links.

A company can also use more than one system: a parent on Workday, a subsidiary on Greenhouse, an old Lever board that was never deleted. Treat the answer as "the board that carries the open jobs today" and keep the date.

The manual and scripted routes

1. Read the apply link

Open a live job ad on the company's careers page and look at where "Apply" goes. This works for all hosted boards because the vendor serves the form:

2. Look for an embed on the company's own careers page

Many companies keep a branded careers page and embed the board: an iframe or script from the vendor host (Greenhouse: boards.greenhouse.io/embed/job_board/js?for={token}), a ?gh_jid= parameter on job URLs, or JSON-LD JobPosting blocks whose url points at the vendor. A custom domain that is a CNAME to the vendor (Teamtailor and Recruitee support it) shows up in a DNS lookup.

3. Probe the vendor API with a guessed token

If the token is likely the company name, a request to the public endpoint answers yes or no: HTTP 200 with jobs means a board exists. This is the cheapest check, and the least safe: a token proves that a company with that slug has a board, not that it is yours.

curl

# Does a Greenhouse board called "stripe" exist? (200 = yes, 404 = no)
curl -s -o /dev/null -w "%{http_code}\n" "https://boards-api.greenhouse.io/v1/boards/stripe"

# Which host does the apply link of a job ad point to?
curl -sI "https://careers.example.com/jobs/123/apply" | grep -i '^location'

Python

import re
from urllib.parse import urlparse

import requests

PATTERNS = [
    (r"(?:job-)?boards\.greenhouse\.io|boards-api\.greenhouse\.io", "greenhouse"),
    (r"jobs(?:\.eu)?\.lever\.co|api\.lever\.co", "lever"),
    (r"jobs\.ashbyhq\.com", "ashby"),
    (r"\.myworkdayjobs\.com$", "workday"),
    (r"(?:jobs|careers)\.smartrecruiters\.com", "smartrecruiters"),
    (r"apply\.workable\.com", "workable"),
    (r"\.recruitee\.com$", "recruitee"),
    (r"\.jobs\.personio\.(?:de|com)$", "personio"),
    (r"\.teamtailor\.com$", "teamtailor"),
    (r"\.oraclecloud\.com$", "oracle"),
]

def ats_of(url: str):
    host = urlparse(url).hostname or ""
    for pattern, name in PATTERNS:
        if re.search(pattern, host):
            return name
    return None

def ats_of_page(careers_url: str):
    html = requests.get(careers_url, timeout=30).text
    found = {ats_of(u) for u in re.findall(r'https?://[^\s"\'<>]+', html)}
    return sorted(x for x in found if x)

print(ats_of("https://boards.greenhouse.io/stripe/jobs/123"))

TypeScript

const PATTERNS: [RegExp, string][] = [
  [/(job-)?boards\.greenhouse\.io|boards-api\.greenhouse\.io/, 'greenhouse'],
  [/jobs(\.eu)?\.lever\.co|api\.lever\.co/, 'lever'],
  [/jobs\.ashbyhq\.com/, 'ashby'],
  [/\.myworkdayjobs\.com$/, 'workday'],
  [/(jobs|careers)\.smartrecruiters\.com/, 'smartrecruiters'],
  [/apply\.workable\.com/, 'workable'],
  [/\.recruitee\.com$/, 'recruitee'],
  [/\.jobs\.personio\.(de|com)$/, 'personio'],
  [/\.teamtailor\.com$/, 'teamtailor'],
];

export const atsOf = (url: string): string | null => {
  const host = new URL(url).hostname;
  return PATTERNS.find(([re]) => re.test(host))?.[1] ?? null;
};

4. Use archives for history

To learn which ATS a company used in the past, the Internet Archive and Common Crawl hold old career pages. Common Crawl's index lets you query a hostname prefix such as boards.greenhouse.io/* per crawl, which is also how you discover boards at scale.

A decision order for scripts

When you automate this for many companies, run the checks from cheap and strong to expensive and weak, and stop at the first confident answer: first the apply link or embed host from the careers page, then JSON-LD JobPosting blocks on a job page, then a vendor API call with a token taken from a link you saw (not guessed), and only last a guessed token with a name check. Record for each answer which step produced it, the date, and the URL you saw. Re-check on a schedule, quarterly is enough for most uses, since migrations are rare but do happen; a changed apply host is the first sign.

Traps

• Parent and brands. The board of a subsidiary does not make the group "a Workday company". Say which entity the board belongs to.
• Token collisions. The same slug exists for unrelated companies in different countries. Check the company name on the board page.
• Dead boards. A company that moved from Lever to Ashby may keep an empty Lever board. Count only boards with open jobs, and note the last capture date.
• Own career sites. Some large employers (Shopify is one) run a custom careers system with no public vendor feed. "Not found" is a valid result.
• Resume-checker sites publish "which ATS does X use" pages from guesses. A job ad's apply link is better evidence.

The fast route: the ATS and its history by domain

For companies whose board we track, GET /v1/companies/{domain} returns the primary board (hiring.ats, hiring.boardUrl, hiring.openJobs, hiring.lastSeen), how sure the match is (hiring.matchConfidence) and every board (boards[] with ATS, token and open jobs).

Calls on this page work without a key from one IP (10 requests a minute, 100 a day, short answers, point lookups). A free key lifts the limits: sign up. Routes, parameters and fields: OpenAPI.

curl

curl -s "https://api.arcadiumdata.io/v1/companies/stripe.com"

Response (abridged)

{
  "domain": "stripe.com",
  "name": "Stripe",
  "hiring": {
    "ats": "greenhouse",
    "boardUrl": "https://job-boards.greenhouse.io/stripe",
    "openJobs": 725,
    "matchConfidence": "high",
    "lastSeen": "2026-10-22T18:39:18.699Z"
  },
  "boards": [ ... ]
}

To see whether a company changed system, ask for its dated switches. The anonymous tier answers this per domain (up to five rows):

curl

curl -s "https://api.arcadiumdata.io/v1/ats/switches?domain=stripe.com"

Each row of data has fromAts, toAts, switchMonth, the last capture of the old board (fromLastSeen), the first capture of the new one (toFirstSeen), how the boards were tied to the domain (evidence) and a confidence. An empty list means no switch was detected, which is the normal case. With a key, filter all switches by fromAts, toAts, country and since (for example Lever to Ashby).

Python

import requests

def current_ats(domain: str):
    r = requests.get("https://api.arcadiumdata.io/v1/companies/" + domain, timeout=30)
    if r.status_code == 404:
        return None  # no board we know of: the company may use its own career site
    r.raise_for_status()
    hiring = r.json().get("hiring")
    return hiring and (hiring["ats"], hiring["matchConfidence"], hiring["lastSeen"])

def switches(domain: str):
    r = requests.get("https://api.arcadiumdata.io/v1/ats/switches?domain=" + domain, timeout=30)
    r.raise_for_status()
    return [(s["fromAts"], s["toAts"], s["switchMonth"], s["confidence"]) for s in r.json()["data"]]

print(current_ats("stripe.com"), switches("stripe.com"))

TypeScript

const res = await fetch('https://api.arcadiumdata.io/v1/ats/switches?domain=stripe.com');
const { data, note } = (await res.json()) as {
  data: { fromAts: string; toAts: string; switchMonth: string; confidence: string }[];
  note: string;
};
console.log(data.length ? data : 'no switch detected', note);

Point-in-time history since February 2023 (open jobs per capture of the public web archive) comes from GET /v1/companies/{domain}/history. The free which-ATS tool runs the lookup in the browser, and company pages show the answer at /company/{domain}/. To test a list, POST /v1/coverage checks up to 50 domains for a known public board.

Limits: we only know boards we have found and tied to a domain. A company on a custom career system answers not_found. The match confidence reflects how the board was tied to the domain; treat low as a lead, not a fact. Switch dates are month-level and come from archive captures, so recent switches appear after the old board has been absent for a few captures.

Comparison

Related

• Which ATS tool
• ATS vendors
• Job board APIs: Greenhouse, Lever, Ashby
• Companies using an ATS
• Workday career site JSON
• SmartRecruiters and Personio endpoints
• ATS market share

Sources

• Greenhouse Job Board API
• Lever Postings API
• Common Crawl index
• Internet Archive: Wayback Machine

Last updated 2026-10-25. This guide is ours and independent of the organisations it describes. Found a mistake? Tell us and we correct it.

Cite as: Arcadium Data, "How to find which ATS a company uses (and when it switched)", https://arcadiumdata.io/guides/which-ats-does-a-company-use/, data as of 2026-10-25.