Vendor and dataset evidence
ORIGINAL RESEARCH / NETHERLANDS / 2026
Netherlands blue-collar job-board coverage:not visible on two major job channels.
In a strict sample of 174 Dutch metal-sector employers with a demonstrably open vacancy on their own site, 79.9% were found on neither Indeed nor Nationale Vacaturebank. The study measures channel visibility under stated rules—not national completeness, total vacancies or every job board.
RESEARCH UPDATED / 09 AUG 2026
EVIDENCE LEDGERSOURCE-LINKED
01
STRICT SAMPLE174 employers
Every sampled employer had a demonstrably open vacancy on its own website.
02
ON NEITHER CHANNEL79.9%
95% interval: 74–86% under the strict Indeed / NVB matching rule.
03
ON INDEED15.5%
95% interval: 10–21%; generic result pages were excluded.
04
ON NVB6.3%
95% interval: 3–10%; consistent with a separate full-index census.
LIMITATIONS INCLUDED
Best-fit choices, not winner theatre
Snapshot and sample definitions
Claims stay inside the evidence
01 / KEY FINDING
Large job channels did not substitute for direct employer-source discovery.
Only 20.1% of the 174 sampled employers were found on Indeed or Nationale Vacaturebank under the study’s strict definition. A valid match required a company profile or an individual vacancy attributable to the employer. Generic pages that merely mentioned a company name or job title in a snippet did not count.
This is evidence of a source-discovery gap in the measured Dutch metal-employer sample. It is not evidence that 79.9% of all Dutch vacancies are absent from job boards, not a comparison against every provider, and not proof that each observed vacancy was unique. The result supports direct-employer collection as a complementary data layer and justifies matched overlap studies with individual providers.
02 / PRIMARY RESULT
Strict visibility across the measured channels.
The denominator is 174 Dutch metal-sector employers with an open role confirmed on their own website. Confidence intervals are the study’s reported 95% intervals.
| Channel outcome | Strict share | 95% interval | What counted |
|---|---|---|---|
| Employer visible on Indeed | 15.5% | 10–21% | A company profile or individual employer-attributable vacancy |
| Employer visible on Nationale Vacaturebank | 6.3% | 3–10% | A company profile or individual employer-attributable vacancy |
| Employer visible on either channel | 20.1% | 14–26% | At least one strict match on Indeed or Nationale Vacaturebank |
| Employer visible on neither channel | 79.9% | 74–86% | No strict match on either measured channel |
| Employer actively recruiting on LinkedIn | 36% | 29–43% | An employer post or genuine vacancy, not an employee profile mention |
The channels were evaluated by employer presence, not by exact vacancy-level overlap. Indeed and LinkedIn indexes can change after the observation date.
03 / WHY STRICT MATCHING MATTERS
Raw search-result presence almost doubled the apparent coverage.
A permissive count put both Indeed and Nationale Vacaturebank near 29%. Manual inspection showed that many results were aggregation pages such as broad lists of technician vacancies where a company name appeared in the snippet but no attributable vacancy existed on the result page.
For Nationale Vacaturebank, 50 raw search hits produced only 11 valid employer pages or individual vacancies under the strict rule. LinkedIn had a similar trap: about 70% of employers produced some result, but many were employee profiles whose work history happened to include the company and role terms. Only 36% showed an employer post or genuine vacancy.
- Count the employer or attributable vacancy, not a keyword mention in a snippet.
- Inspect the landing page instead of trusting the search-result title alone.
- Separate recruiting activity from employee-profile or résumé mentions.
- Preserve the matching rule so a later measurement can be compared fairly.
04 / SECOND MEASUREMENT
A full Nationale Vacaturebank metal census pointed in the same direction.
The study also collected the complete Nationale Vacaturebank metal segment exposed through its public index at the observation date: 1,222 vacancies across 459 employers. Sixty-four intermediaries represented 14% of listed employers but 52% of vacancy volume. Removing those intermediaries left 395 end employers and 586 vacancies.
Against an earlier Blue Collar Job Data metal-employer snapshot of 307 companies, 29 employers overlapped: 9.4% of the Blue Collar Job Data companies and 7.3% of the end-employer set in the Nationale Vacaturebank census. This separate method is directionally consistent with the 6.3% strict sample result, but the denominators and filters differ and should not be collapsed into one figure.
1,222 vacancies
Publicly indexed vacancies attributed to 459 employer or intermediary organizations.
395 companies
Organizations remaining after the study’s intermediary classification rule.
52% of volume
Sixty-four intermediaries accounted for more than half of listed metal vacancies.
29 companies
9.4% of the earlier BCJD metal set and 7.3% of the NVB end-employer set.
05 / METHOD
How the 174-employer benchmark was assembled.
The raw query outputs and classifications are retained internally. The public page publishes the decision rule, denominator, observation date and limitations so the result can be challenged and repeated. A future release should add a privacy-reviewed downloadable aggregate table; no raw vacancy descriptions or personal information are published here.
- Population condition: a Dutch metal-sector employer with a demonstrably open vacancy on its own website.
- Sample size: 174 employers in the strict benchmark.
- Observation method: controlled channel-visibility research on 08 August 2026 using fixed inclusion and classification rules.
- Indeed / NVB strict match: company profile or individual vacancy clearly attributable to the employer.
- LinkedIn strict match: employer recruiting post or genuine vacancy; employee-profile mentions excluded.
- Intervals: reported 95% intervals around the observed employer shares.
06 / LIMITATIONS
What this study cannot tell you.
The appropriate next step is repetition, not overclaiming. Rerun the same strict benchmark quarterly, preserve the old classification decisions and add pre-registered samples in other blue-collar categories.
- It does not estimate the share of all Dutch vacancies missing from all job boards.
- It measures metal-sector employers, not every blue-collar sector or company size.
- It measures employer visibility, not exact vacancy-level duplication across channels.
- It does not compare Blue Collar Job Data with JobDigger, Jobfeed, TheirStack or another commercial provider.
- Search indexes and channel inventories change; this is a dated snapshot, not an evergreen coverage guarantee.
- Intermediary classification uses an operational rule and can misclassify organizations with unusual geographic hiring patterns.
07 / IMPLICATIONS
What data buyers and recruiters should do differently.
Do not use one board as the whole market
Direct employer sources can identify active companies that a major channel does not surface under strict rules.
Ask for a matched overlap study
A provider’s global count is less informative than valid unique employers in the target market slice.
Document source mix
Job-board and intermediary-heavy data can describe a different market layer from employer-career sources.
Keep provenance inspectable
Source routes and observation rules make it possible to audit false positives and channel differences.
DECISION GUIDE
Treat direct-employer data as a complementary layer.
The research supports diversification of sources. It does not support replacing every job board or established labor-market provider with one specialist dataset.
The study supports…
- Testing direct employer sources beside major job channels.
- Using strict result validation instead of snippet counts.
- Separating intermediaries from end employers where the workflow needs it.
- Running provider-specific overlap tests in a defined market slice.
The study does not support…
- A claim of complete Dutch blue-collar coverage.
- A claim that 79.9% of all vacancies are off job boards.
- A superiority claim against an unmeasured vendor.
- A real-time or exact vacancy-status guarantee.
FAQ
Questions buyers ask before they choose.
The short answers are visible in the page; the evidence is listed below.
01Does this mean 79.9% of Dutch blue-collar jobs are not on job boards?
No. It means 79.9% of 174 sampled Dutch metal-sector employers with an own-site opening were found on neither Indeed nor Nationale Vacaturebank under the strict employer-presence rule.
02Why were generic result pages excluded?
A generic page can mention a company or role in a snippet without providing an attributable vacancy. Counting those pages inflated apparent coverage and did not answer whether the employer was genuinely represented.
03Was JobDigger or Jobfeed included in the study?
No. The strict benchmark measured Indeed and Nationale Vacaturebank, with a separate LinkedIn recruiting-activity check. No overlap claim is made for an unmeasured provider.
04What does open-status mean in the Blue Collar Job Data snapshot?
It is the dataset state attached to the latest successful observation rule. It should not be read as a guaranteed opening at the moment a reader visits this report.
05Will this research be repeated?
The recommended method is a quarterly repeat using the same strict rules, with additional pre-defined blue-collar sectors. Future releases should show trend changes only after comparable observations exist.
SOURCE LEDGER
What this page is based on.
Product facts were checked against the linked first-party sources. Blue Collar Job Data findings use the 08 August 2026 internal snapshot and the methodology stated on this site.
- Blue Collar Job Data methodology
Public explanation of source observation, retained context, classification and disclosed boundaries.
- Blue Collar Job Data coverage ledger
Dated 08 August 2026 dataset context for the wider product; the report sample has its own denominator.
- Indeed Netherlands
One of the two channels measured through strict employer-result classification.
- Nationale Vacaturebank
The second measured channel and source of the separate full metal-index census.
- LinkedIn Jobs
Used for a separate strict check of genuine employer recruiting activity.
- Report research note
Blue Collar Job Data internal study, conducted 08 August 2026; n=174; methodology and limitations reproduced on this page.
NEXT STEP
Run the same strict test on your target slice.
Tell us the trade, geography and provider or channel you want to evaluate. We will confirm whether the current data and source evidence can support a transparent overlap study.