Skills based hiring: An ultimate guide
4 minutes
Ben explains what skills based hiring is and how it allows you to open up your roles to a larger, more diverse applicant pool
You've likely heard the term "skills-based hiring" (also called skills-based recruiting or skills-first hiring) in your HR and TA circles by now. But what does it even mean? "Yes, I would like my candidates to be skilled, please", have we not been hiring for skills this whole time?
Skills-based hiring is a recruitment approach where you identify the skills a role genuinely needs and assess candidates on those skills directly, rather than using degrees, job titles or years of experience as a stand-in for them.
I'm sure you've hired people who looked great on paper - strong degree, top university, relevant experience - who then didn't live up to expectations. If testing skills directly avoids that, why weren't we always doing this?
Skills-based hiring prioritises selecting candidates based on specific skills, not their experience
Most employers have been doing the opposite, using degrees and years of experience as a screening tool. A 2:1 is often a hard requirement for early-careers roles, so anyone with a 2:2, an unrelated degree or no degree at all is cut before you've learned anything about them. What's left is a smaller pool of people who tick those boxes, on the assumption that the boxes predict performance.
They don't, really. Just having experience in something doesn't necessarily make you good at it, and the research backs that up. The predictive validity of education is just 0.10, and years of experience 0.16, according to Schmidt, Oh & Shaffer's (2016) review of 100 years of selection research (where 1 is a perfect predictor of job performance and 0 is no relationship at all). Where someone studied or worked can add some context, but it's far too weak to screen on. Every university and employer has strong and weak performers, and assuming an individual's ability from where they've been is what psychologists call the "ecological fallacy".
Skills-based hiring says: instead of assuming, just find out. Let anyone interested apply, and assess the skills the role needs directly with cognitive ability tests, structured interviews and skills assessments. Your job description changes from "must have a 2:1 in a related subject and four years' experience" to "you'll need to be analytically minded, adaptable, and a strong communicator".
That said, skills-based hiring doesn't mean credentials never matter. Some roles genuinely need a specific qualification or track record. The point is to make sure every requirement on your job spec is there for a real reason, not just because it's always been there.
What are the "skills" being assessed in skills-based hiring?
Skills based hiring is not just about skills like coding or excel
- 2 minutes
Skills based hiring looks at a breadth of skills, not just hard technical skills.
Skills in hiring typically fall into three different buckets:
1. Cognitive skills are someone's natural mental aptitude
This is how they think, learn, and solve problems. Things like numerical reasoning, verbal reasoning, critical thinking, and pattern recognition. These are largely stable in adults and very hard to train.
When you combine scores across several cognitive aptitudes, you get a measure of general cognitive ability, and it's worth knowing that this is the strongest single predictor of job performance across almost every role type, with a validity of 0.65, compared with 0.10 for education and 0.16 for experience. Smart people learn faster, solve problems faster, and generally figure out the right thing to do, which feeds into everything an employee does.
2. Hard skills are specific technical knowledge or learned ability
Think coding languages, software packages, financial modelling, industry-specific expertise. These skills can be taught and developed, but they can also become outdated surprisingly quickly.
Take Excel. A few years ago, advanced spreadsheet skills were genuinely sought after. Now you can describe what you need in plain English and AI will build it in seconds, so "Proficiency in Excel" has little value as a requirement.
The same is happening across a lot of technical domains, which is why, for most roles, the ability to learn hard skills quickly is worth more than existing knowledge of them.
3. Soft skills are personality traits, behaviours, and values
These are traits like resilience, conscientiousness, attention to detail, drive, and adaptability. Unlike hard skills, these are relatively fixed in adults. You are largely who you are. That's exactly why hiring for soft skills matters: you can't train conscientiousness into someone, or coach them into being genuinely driven.
The right combination of soft skills is often what separates your superstar employees from the ones who burn out and leave early. I'm sure you've noticed it's not always the most technically capable who thrive, but the one whose personality is the right fit for how the role actually works.
Pro Tip
Of the three, cognitive and soft skills are relevant to every role. Hard skills only become essential for genuinely specialist positions. For most hiring, if you focus on cognitive and soft skills, the hard skills tend to follow quickly with time in the role.
None of these skills can be measured from a CV; you can only make assumptions about them. Skills-based hiring removes the guesswork by testing each candidate's ability directly, using targeted assessments for each skill type: aptitude tests for cognitive ability, personality assessments for soft skills, and structured interviews to dig deeper into both.
Skills-based vs experience-based hiring, a head-to-head comparison
Here's how the two approaches compare.
| Factor | Experience-based hiring | Skills-based hiring |
|---|---|---|
| Primary filter | CV, degree, years in role | Skills assessments, work samples |
| Interview approach | Probing past experience and work history | Structured questions assessing skills needed for the role |
| Bias risk | High (name, university, background) | Lower (scored objectively) |
| AI vulnerability | High (CVs easily optimised/exaggerated) | Lower (assessments harder to fake) |
| Predictive validity | Low (education 0.10, experience 0.16) | High (cognitive ability 0.65, structured interviews 0.58) |
| Fairness | Advantages privileged backgrounds | Levels the playing field |
| Scales with volume | Poorly | Highly scalable |
How skills based hiring differs from experience based hiring
- 5 minutes
Ben explains the key differences between both approaches and what makes skills based hiring the better choice
Why skills-based hiring has become so popular (and its benefits)
It predicts performance far better
Most importantly, it maximises what psychologists call "predictive validity": how well your selection process predicts real-world performance.
HR practitioners and hiring managers often forget that selection and recruitment is simply a predictions market, whereby the employer predicts which candidates will make the best employees.
Cognitive ability tests (0.65) are the strongest single predictor, and structured interviews designed to assess specific skills aren't far behind at 0.58. Use the two together and validity rises to 0.76. So, by measuring the skills that actually matter in the role, you get better information. And better information leads to better hires.
CVs can't be trusted in the age of AI
In the space of a few years, AI has moved into almost every home and pocket, and that includes your applicant pile. At least half of UK job applicants now use AI to help write their CV, and almost anyone can produce a tailored CV, cover letter or "why this role?" answer in seconds (I guess it shows good delegation skills).
That doesn't make written applications pointless. Eligibility questions still matter, and reading your shortlist's CVs gives you useful context to shape your interview questions. But written content is no longer a reliable way to screen candidates. When every application reads well, all that's left to sift on is the degree, the university and the years of experience - the same weak proxies from earlier.
It opens up a much bigger talent pool
Top universities don't have a monopoly on talent - there are high-potential graduates at every university, and plenty of capable people who never went. LinkedIn's data suggests that focusing on skills rather than credentials can grow a talent pool by up to 10 times.
AI has replaced a lot of entry-level roles, so the number of applicants you'll receive has likely increased and may keep increasing as AI grows stronger. Manual selection methods like CV-sifting are a no-go at volume.
So a bigger pool might sound like a problem: "I already have too many applicants, I don't need more!" It's only a problem if you're using manual, unscalable methods, but with scalable methods it's actually a huge benefit.
Say your requirements - a 2:1, two years' experience, a target university list - bring in 500 applicants. Dropping most of those requirements (that you were using as a proxy for ability) and using assessments designed to target the skills you need for the role might increase your applicant pool to 5000. Either way, you might take 50 candidates through to the next stage. From 500 applicants, that's your top 10% - a decent shortlist. From 5000, it's your top 1%, and statistically a far stronger one. Same size shortlist, much better candidates, just from replacing barriers that weren't predicting performance anyway with assessments that do.
It's fairer
Historically, candidates from prestigious universities, or with high degree classification, were given preferential treatment in the recruitment process. That favours people with the contacts, university access and financial freedom to build an impressive CV, and filters out those without them, regardless of their actual ability. Targeting prestigious universities disproportionately impacts minority candidates, who are often underrepresented. And with the cost of university skyrocketing, many capable people are skipping it altogether to join the workforce without the debt.
Skills-based hiring provides a level playing field for everyone: all that matters are the candidate's skills today, not the opportunities they've had.
Skills-based hiring has gone mainstream, at least on paper
Over the last few years, skills-based hiring has gone from HR buzzword to standard practice for a lot of employers, and you can see it in job adverts. Indeed's analysis of UK job postings found that 86% don't mention any education requirement at all, and only around 5% explicitly ask for a degree. Some big names led the way: in the UK, EY, PwC and Penguin Random House dropped degree or UCAS-points requirements for many roles, and in the US, IBM, Google and Accenture have done the same.
But the job advert is the easy part to change. A 2024 study by Harvard Business School and the Burning Glass Institute found that although US employers had dropped degree requirements from far more job adverts, it changed who actually got hired in fewer than 1 in 700 hires. Around 45% of the companies that removed requirements made no meaningful difference to who they hired.
The researchers' conclusion was that it takes "more than simply stripping language from job postings". If the advert changes but candidates are still sifted on their CVs, the degree is still doing the work. Skills-based hiring only works when you replace the old filters with a direct measure of skill.
Where skills-based hiring can go wrong
Skills-based hiring is only as good as the way it's put into practice. These are the most common ways it falls short:
- Removing requirements but still sifting CVs. If the CV still decides the shortlist, nothing really changes.
- Assessing skills you haven't defined. Testing for everything, or for whatever feels important, rather than the handful of skills that actually drive performance in the role.
- Using assessments that can't be trusted. Unsupervised hard-skill tests are easy to complete with AI, and assessments without validation evidence behind them don't predict much at all. The quality of your assessments determines the quality of your hires.
- Never checking the results. Without comparing assessment scores to on-the-job performance of your new hires, you can't tell whether your process is working or how to improve it.
None of these are reasons to go back to CVs. They're reasons to do skills-based hiring properly, with clearly defined skills and well-validated assessments.
How to get started with skills-based hiring
Here's the process in brief. For the full walkthrough, with worked examples and our free toolkit, see our guide to implementing skills-based hiring.
- Work out what skills the role actually needs. Speak to hiring managers and top performers to pin down the handful of cognitive and soft skills that drive success in the role.
- Rewrite your job description around those skills. Strip out any degree or experience requirement that isn't there for a genuine reason.
- Replace CV screening with assessments. Use aptitude and personality assessments as your first sift, so every applicant is measured on the same skills.
- Restructure your interviews. Ask every candidate the same skills-focused questions and score their answers against a consistent framework.
- Monitor and improve over time. Check whether assessment scores line up with on-the-job performance, and adjust your process as you learn.
Conclusion and next steps
The research making the case for skills-based hiring has been around for decades, but certain changes have forced people to catch up. AI has exposed a CV's unreliability. Degree requirements are increasingly seen as a barrier rather than a signal. And the pressure to hire fairly and reduce bias has made the old approach less viable.
Removing a degree requirement from a job advert is a start, but the real shift happens when you measure skills directly. If you're ready to build that into your process, our step-by-step implementation guide walks through exactly how to do it - including our free toolkit.
Primary author
Ben Schwencke
Chief psychologist at Test Partnership. MSc in Organisational Psychology with over 10 years' experience in psychometric testing.
Contributed to by:
