Skills-based hiring means selecting candidates on what they can demonstrably do, not on proxies such as a degree, a job title or years of experience. In practice you define five to eight skills per role, test them with a work sample or a structured interview, and score every candidate against the same rubric. A degree stays a valid signal where the law or the job requires it; it just no longer decides who gets seen.
Skills-based vs. credential-based hiring
The difference shows at every stage of the funnel. The table compares how each approach handles the stage.
| Stage | Credential-based | Skills-based |
|---|---|---|
| Job ad | "Degree in X, 5+ years" | 5–8 named skills, split into must-have and nice-to-have |
| Screening | Keyword and degree filter | Evidence of each skill (projects, work history, answers to knockout questions) |
| Assessment | Unstructured conversation | Work sample or job-related test, same for everyone |
| Interview | Gut feeling, "culture fit" | Structured interview with a scored rubric |
| Main risk | Rejects capable people without the "right" CV | Poorly built tests that do not predict the job |
How to implement skills-based hiring in six steps
1. Run a job analysis
Ask the hiring manager and two people who are good in the role: which tasks decide success in the first year, and which skills do they need? Write five to eight skills as observable behaviours ("writes a stakeholder update that a non-expert understands"), not as traits. The EU's classification ESCO helps with a shared vocabulary for occupations and skills. In Germany, DIN 33430 describes the same logic: derive every selection method from a documented requirements analysis.
2. Rewrite the job ad
List the skills first and the degree last, and only if it is legally required. Every requirement you keep must be needed on day one.
3. Screen on evidence
Replace the degree filter with knockout questions tied to must-have skills (automating knockout criteria) and read the CV for proof of the skill, not for the title. This is where most teams lose time, because reading 200 CVs against eight skills by hand takes a day; CV screening against a fixed criteria list shortens that.
4. Test the skill that matters most
Use one work sample of 20–45 minutes that mirrors a real task. Keep it job-related and identical for every candidate; see pre-employment skills testing for formats and pitfalls. A test nobody validated is the most common way skills-based hiring fails.
5. Interview with a rubric
Ask every candidate the same questions, each tied to one skill, and score answers on an anchored scale (1 = no evidence, 5 = clear, repeated evidence). Two interviewers score independently before they talk.
6. Measure and adjust
Track pass-through rates per stage, time to hire and 90-day performance. Compare them with your last credential-based hires. If a test rejects one group far more often than others, fix the test, not the group (see the legal section).
What the data says
Intentions are far ahead of practice. A study by Harvard Business School and the Burning Glass Institute (February 2024) found that dropping degree requirements changed about 97,000 hires out of 77 million a year in the US, fewer than 1 in 700. Of the large firms studied, 37 % followed their pledge with real change and 45 % announced it without changing hiring. At the firms that did change, workers hired without a degree stayed about 10 percentage points more often than their degree-holding colleagues (Harvard Kennedy School summary, HR Dive, as of February 2024). The public sector is moving by law: the US Chance to Compete Act of 2024 (Public Law 118-188, signed 23 December 2024) pushes federal hiring from self-assessments to technical, skills-based assessments.
Worked example: what a degree filter costs you
These are assumptions, not measurements. A role draws 200 applications. If 35 % have no degree, the filter removes 70 people before anyone reads them. If one in five of those would have passed your skills test, you lose 14 qualified candidates. Checking all 200 CVs against your must-have skills with Sprad's CV check costs 2 credits each, 400 credits in total, which is 40 € at the list price of 0.10 € per credit (list price, as of 29 October 2026). That is the price of reading what the filter would have hidden.
Legal limits: skills tests are selection procedures
A skills test counts as a selection procedure, so it must be job-related and checked for adverse impact.
- USA: the Uniform Guidelines on Employee Selection Procedures (29 CFR Part 1607) apply to tests. A selection rate for any group below four-fifths (80 %) of the highest group's rate is generally regarded as evidence of adverse impact. Example: 40 of 100 candidates in group A pass, 24 of 100 in group B; 24 ÷ 40 = 0.6, below 0.8, so check the test.
- Germany: the AGG forbids discrimination in selection. If you show indications of discrimination, the other side must prove there was none (§ 22 AGG). A fully automated rejection falls under Art. 22 GDPR; see can AI reject candidates automatically. Introducing AI-based assessments usually involves the works council (§ 87(1) no. 6 BetrVG).
- EU AI Act: AI systems used to select applicants are high-risk; those obligations apply from 2 December 2027 (Regulation (EU) 2026/1744, as of 30 October 2026).
Where Sprad helps (and where it does not)
Sprad is a vendor itself, so weigh this accordingly. Skills-based hiring starts before the application: with AI people search you describe the role by requirements, and Atlas checks each profile individually against them instead of matching keywords. That finds people whose profile lacks the usual title or degree. In Sprad's own test (15 August 2026, three roles, 218 profiles, each system's own hits excluded), Atlas's coverage was 38 % against 9 % for the best of the four other tools. That is a vendor measurement, not an independent test; the method is on the benchmark page. After applications arrive, CV screening and voice interviews score every applicant against the same criteria: CV check 2 credits, voice interview 30 credits in 12 languages. Your talent pool keeps the people you did not hire this time. Sprad does not validate your tests; that stays with you or your occupational psychologist.
FAQ
Does skills-based hiring mean no degrees?
No. Regulated professions such as physicians or lawyers still require their qualification. Skills-based hiring removes degrees as a default filter for roles where the degree is not what makes someone good.
What is the difference to a competency-based interview?
A competency-based interview is one method inside the process. Skills-based hiring is the whole approach: job analysis, ad, screening, test and interview all follow the same skill list.
Is skills-based hiring allowed in Germany?
Yes, and job-related selection is what the AGG expects. What you must avoid are tests or criteria that disadvantage protected groups without a job-related reason, and fully automated rejections without human review.
Does it work for senior roles?
Yes, with work samples that match the level: a 90-day plan, a case from the candidate's field, or a structured reference conversation. Track record stays a skill signal, but it is scored against the skill list instead of the title.
How do I know it worked?
Compare pass-through rates, time to hire and 90-day performance with your previous hires, and check each test for adverse impact at least once a year.






