Job search statistics, fact-checked: the DACH and pharma edition

"75% of CVs are rejected by an applicant tracking system"

The figure traces to Preptel, a vendor of CV-optimisation software that used it in its marketing around 2012. No study, sample or method has been published, and Preptel closed in 2013. A 2013 article by Dr John Sullivan on ERE.net, quoting Talent Function Group, then described 75 of every 100 completed applications being screened out by either the ATS or a Recruiter. As the number travelled, the Recruiter dropped out of the sentence.

Jobscan, a company that sells CV-scanning software, now writes that an ATS rarely "rejects" a candidate automatically and works mainly as a search tool for Recruiters. Germany has a measured figure. In Bitkom's representative survey of 602 companies, run in May and June 2026, 6% use AI for the first review of incoming applications and 12% use AI anywhere in hiring. Its March 2025 survey of 852 companies put AI pre-selection at 1%. The two samples differ, so read that as a direction of travel rather than an exact growth rate.

What does screen many candidates out at an early stage is the application form: knockout questions about work permits, years of experience or salary, set up by a Recruiter before the role opens. I took this apart in The 75% auto-rejection myth.

Sources: Bitkom, 28th September 2026; Bitkom, 10th March 2025; Jobscan; ERE.net, 20th May 2013.

"85% of jobs are filled through networking"

The number comes from a LinkedIn post by hiring consultant Lou Adler dated 29th February 2016. Over twelve months he had asked people how they found their most recent job. About 3,000 answered, most of them in staff or management roles, and they chose to take part, so it isn't a random sample. Adler then weighted the answers with separate LinkedIn estimates of how many people are actively looking and concluded that 85% of "critical jobs" are filled via networking "of some sort".

Germany has far better data. The IAB-Stellenerhebung, a representative quarterly survey of around 9,000 employers run by the research institute of the Federal Employment Agency, asks how each hire was actually made. For 2025, employers used their own staff or personal contacts in 48% of hires, and that channel decided the hire in 27%. In 2022 it decided 37%. Online channels together decided 52% of 2025 hires.

A contact still decides roughly one German hire in four. For a single channel, that's a lot. It's still a long way from 85%. The longer version is in No, 85% of jobs are not filled through networking.

Sources: Lou Adler on LinkedIn, 29th February 2016; IAB-Stellenerhebung 1/2026, 8th June 2026; IAB-Stellenerhebung 1/2023, 11th May 2023.

"70 to 80% of jobs are never advertised"

Career counsellor Bernard Haldane used the 80% figure in 1966, citing a pilot survey of employers in Rochester, New York, that the Ford Foundation had funded. In that survey, about a quarter of the hires at 27 companies had appeared in a newspaper advert. A later analysis points out that Rochester's large manufacturers had agreed through their industry council not to advertise in local papers, which pushes the unadvertised share up for that one city in that decade. Career books carried the number forward from the 1970s on, and the modern "70 to 80%" versions usually cite nothing at all.

In the IAB data for 2025, German employers used their own careers page in 70% of hires and internet job boards in 57%, and 82% of hires involved at least one digital channel. Unadvertised hiring exists. From my own recruiting work, it's most common at senior and executive level, where search firms run confidential mandates. For most German hires, though, the role was posted somewhere. More on this in There is no secret door to the hidden job market.

Sources: Learning Curves, Hidden origins of the hidden job market; IAB-Stellenerhebung 1/2026.

"Every corporate job opening attracts 250 CVs"

The earliest version I could find is the same May 2013 ERE.net article by Dr John Sullivan, which gives the figure without any method. Glassdoor repeated it in its "50 HR & Recruiting Stats" collection in January 2015, and most later citations point to Glassdoor.

Thirteen years on, one large platform lands close to it. Greenhouse's Hire Standard report from March 2026, built on more than 640 million applications at over 6,000 companies, counted 244 applications per job in North America in 2025, up from 116 in 2022. Its UK edition counted 183. German numbers look nothing like this. The IAB found 16 applications per hire in 2025, up from 9 in 2022, and employers judged about 4 of them, or 28%, suitable. On BioSpace, a US life sciences job board, the average was 4.4 applications per posting in Q2 2026.

Per job, per hire and per posting measure different things, so the figures sit far apart, and the 250 figure says little about a specialist pharma role in Germany.

Sources: Greenhouse, The Hire Standard, March 2026; Greenhouse UK edition; IAB-Stellenerhebung 2/2026, 3rd September 2026; BioSpace Q2 2026 job market report; Glassdoor, January 2015.

"Recruiters spend six seconds on a CV"

TheLadders, a US job site, ran an eye-tracking study with 30 professional Recruiters over ten weeks in 2012. It found about six seconds for the first fit or no-fit decision. A 2018 repeat found 7.4 seconds and did not publish its sample size.

So the number is real, but small and narrow. It describes the first skim of a pile, measured by a company that sells job search services. A CV that survives the skim gets read properly. The practical lesson from both rounds holds up well: put job titles, employers and dates where a reader expects them.

Sources: ERE.net, Is the 6-second resume scan a myth?; HR Dive on the 2018 study.

"One job ad in five is a ghost job"

Greenhouse reported in December 2024 that in any quarter, 18 to 22% of the jobs posted on its platform are classified as ghost jobs, meaning positions advertised with no intent to hire. Hunter Ng of Baruch College, City University of New York, ran nearly 270,000 Glassdoor interview reviews through a language model and estimated that up to 21% of job ads may be ghost jobs, with higher shares at large firms and in specialised industries. That paper is a working paper on arXiv. Both figures are estimates with their own definitions.

Germany adds a twist. In the IAB survey for Q2 2026, about 81% of vacancies were to be filled immediately and the other 19% later. A role planned for later is real. From the outside, it looks exactly like one that never starts. I covered the German side in Ghost jobs, ghost candidates and the German data.

Sources: Greenhouse 2024 State of Job Hunting report; Ng, Why is it so hard to find a job now? Enter Ghost Jobs, arXiv:2410.21771; IAB-Stellenerhebung 2/2026.

"Women apply only when they meet 100% of the criteria, men at 60%"

The claim comes from an internal Hewlett-Packard report, quoted in Lean In and in a 2014 Harvard Business Review article by Tara Sophia Mohr. The report itself has not been made public, so its method can't be checked from outside the company. Mohr ran her own survey of more than 1,000 mostly US professionals. Among the women, 78% of the reasons given came down to reading the listed qualifications as hard rules, and 15% of women against 8% of men said they were "following the guidelines about who should apply".

The better number comes from behaviour on a platform. LinkedIn's 2019 Gender Insights Report found that women applied to 20% fewer jobs than men and were 16% more likely to be hired after applying.

Sources: Mohr, Harvard Business Review, 25th August 2014; Curt Rice on the missing evidence, April 2014; LinkedIn Gender Insights Report, March 2019.

"Hiring filters screen out qualified people"

Harvard Business School and Accenture surveyed more than 2,250 executives and over 8,000 workers in the US, the UK and Germany for their 2021 report Hidden Workers: Untapped Talent. 88% of the employers agreed that qualified high-skills candidates are vetted out of the process because they don't match the exact criteria in the job description. For middle-skills candidates the figure was 94%. The report also describes filters that remove candidates with an employment gap of six months or more.

The caveat matters: these are executives describing their own processes in 2021. That's strong evidence that filters cost them good people, and it gives no rejection rate.

Sources: Harvard Business School, Hidden Workers: Untapped Talent; HR Dive summary of the report.

"Germany's skills shortage means employers will take whoever applies"

The IAB counted about 1.02 million open positions in Germany in Q2 2026, 11% fewer than in the quarter before. On average there were 291 registered unemployed people for every 100 vacancies, 339 in the east and 279 in the west. Employers received 16 applications per hire in 2025 and judged 28% of them suitable. KfW Research, working from the same IAB survey, reports that employers had difficulty filling 31% of their vacancies in 2025, about 340,000 roles.

Both things are true at once. Some profiles are genuinely hard to find, and the average applicant still faces competition and a strict test of fit.

Sources: IAB-Stellenerhebung 2/2026; KfW Research, Fokus Volkswirtschaft No. 562, 3rd November 2026.

"Biopharma layoffs keep getting worse"

BioSpace's Q3 2026 job market report counts 27 companies that made or planned cuts in the quarter, affecting 2,568 employees. That is 90% fewer affected employees than a year earlier and the lowest quarterly figure in 2025 and 2026. Job postings live on BioSpace were up 31% on Q3 2025. The tally leaves out CROs, CDMOs, tools and services firms and medical device companies.

These are US figures. I haven't found a comparable public tracker for Europe, so anyone claiming a European trend in either direction should show their source.

Source: BioSpace, 2026 Q3 job market report.

"Pharma hiring processes take months"

In the pharma interview index I compile every month from public Glassdoor figures, candidates report an average hiring process of 17 days at Worldwide Clinical Trials up to 43 days at UCB, across 39 pharma, biotech and CRO employers. GSK sits at 41 days and Bayer at 37.

These are candidate-reported averages for the interview stage, so they don't include the weeks before a role is approved or the notice period after you sign. Treat them as a signal of how long the process runs once you're in it.

Source: MoreThanCareer.de pharma interview index, November 2026 edition, based on Glassdoor interview pages.