Skills vs. Titles: Why HR Teams are Switching to Skills-based Hiring Worldwide

In 2024, researchers at the Burning Glass Institute and Harvard Business School did something nobody selling skills-based hiring had bothered to do. They checked whether it was happening.

Skills-based hiring means judging candidates on what they can actually do, measured through tests, work samples and structured interviews, rather than on the degree or job title on their resume. The idea is sound and the evidence behind it is real. The execution, across most of the companies that announced it, was not.

Across 11,300 roles at large firms where a degree requirement was dropped, hiring of workers without a bachelor's degree rose by about 3.5 percentage points. Spread across the whole labor market, the authors of the joint Burning Glass Institute and Harvard Business School report put the result plainly: the promised opportunity showed up in not even 1 in 700 hires.

That is the honest starting point, and it is better news than it sounds. The same study found a group of firms where the change was real, and it recorded what they got for it. This article is about what those firms did differently.

TL;DR

• Deleting a degree line from a job post changes almost nothing on its own. Measured across 11,300 roles, it moved non-degree hiring by about 3.5 percentage points.
• Firms split three ways after dropping requirements: 37 percent made real change, about 45 percent changed in name only, and about a fifth went backwards.
• Where it worked, it paid. Non-degreed hires at those firms retained 10 percentage points better than degree-holding colleagues, and workers moving into formerly degree-gated roles saw pay rise about 25 percent.
• What separates the leaders from the rest is replacement, not removal. A degree screen filters people; if you take it out without putting a measured signal in its place, the hiring manager just falls back on the resume.
• The moment a test replaces a degree screen, anti-discrimination law attaches to the test. Job-relatedness stops being a nice idea and becomes the standard you would have to defend.

What is skills-based hiring?

Skills-based hiring is a selection approach that evaluates candidates on demonstrated, job-relevant ability rather than on credentials used as a proxy for it. Instead of screening for a degree, a job title or years served, you define the competencies the role needs and gather direct evidence of each one through assessments, work samples, structured interviews and reference checks.

The word doing the work in that definition is "demonstrated". A resume is a claim. An assessment result is evidence. Everything else about skills-based hiring follows from preferring the second to the first.

It helps to be precise about what is being replaced. A degree is not useless information. It is a cheap, weak signal that correlates loosely with a lot of things and measures none of them directly. Hiring teams reached for it because it was free and fast, not because anyone tested it against performance. Why skills-based hiring matters covers the fundamentals in more depth.

Does skills-based hiring actually work?

Yes, at the firms that implemented it, and that qualifier carries most of the weight. The Burning Glass Institute and Harvard Business School study sorted employers who removed degree requirements into three groups by what their hiring actually did afterwards. Nearly all the measured change came from 37 percent of them.

Read the middle row again, because it is the most likely outcome for any team that treats this as a policy change. About 45 percent of firms rewrote their postings and kept hiring exactly the same people. Nothing in that group's behavior was dishonest. They removed a filter and never replaced the question it had been answering, however poorly: how do we know this person can do the job?

One detail about the evidence is worth stating, because it is unusual on this topic: the 2024 joint report behind these figures was produced by a research institute and a business school, not by a vendor selling assessments. Most writing on this topic, including this page, comes from someone with a commercial interest in the answer.

Skills-based hiring news: what changed since 2024

The useful shift since that audit is not more announcements. It is that the measurement problem got cheaper to solve. Role-based assessment libraries, work-sample tasks and structured interview scoring are now standard tooling rather than a custom project, which removes the practical excuse the "In Name Only" group had.

What has not changed is the failure mode. A team that drops a requirement in Q1 and does not change its screening step by Q3 has joined the middle row.

Rethinking hiring fit: evidence over proxies

Hiring fit used to mean pattern-matching: the right degree, the right logo, the right title progression. The trouble with proxies is not that they are offensive. It is that they are imprecise, and the imprecision runs in a predictable direction.

Four things go wrong when a title or a credential stands in for ability:

• Titles are not standardized. A "Marketing Manager" at a 40-person company and at a 4,000-person company share a job title and almost no overlap in daily work. The title tells you about the employer's naming habits, not the candidate's skills.
• Skills decay faster than credentials expire. A degree is a permanent record of a temporary state. Someone certified five years ago in a field that has since rebuilt its tooling holds a credential that describes a person who no longer exists.
• The filter excludes people who can do the work. Capable candidates who learned on the job, through apprenticeships, bootcamps, military service or self-teaching fail a screen that was never about ability.
• It imports someone else's hiring decision. Screening on a prestigious former employer means inheriting that company's judgment, including its mistakes, and treating it as your own evidence.

Replacing a proxy is harder than deleting it, which is the whole lesson of the three groups above. You need to know what the role requires, and you need a way to see it. Our guide to mapping skills across roles and the practice of profiling a job properly are where this starts.

What are the real benefits of skills-based hiring?

The benefits that survive contact with evidence are narrower than the usual list, and more persuasive for it. At the firms where the change was real, the study measured two outcomes worth quoting.

Retention improved. Non-degreed workers hired into roles that formerly asked for a degree stayed at a rate 10 percentage points higher than their degree-holding colleagues in the same roles. That is the opposite of the quality drop most hiring committees fear. The pay finding below probably explains it: these were real steps up, won on evidence, for people a credential screen had been holding back.

Pay rose for the people hired. Workers who moved into formerly degree-gated roles saw their pay increase about 25 percent on average. For an employer, that is the clearest sign the skill was genuinely there and previously mispriced.

Two more benefits are real but less measurable. The candidate pool gets bigger, which matters most in roles where you are already struggling to fill a req. And the hiring conversation improves, because a scored assessment gives a panel something specific to disagree about instead of competing impressions.

Here is the tradeoff nobody advertises. Skills-based hiring front-loads effort. You spend more time before the first interview defining what you are measuring and building the screen, and less time afterwards arguing about shortlists. If your team cannot absorb that upfront cost, you will end up in the middle row of the table, and you would honestly be better off keeping the old filter than removing it and replacing it with nothing.

How do you build a skills-based hiring process?

Start with the role, not with a test. That sequence is the core of the Testlify Competency-to-Evidence Matrix, which maps every role to the competencies that matter, then connects each competency to measurable evidence through assessments, simulations, interviews, references and structured reviewer feedback. The output is explicit: every competency you claim to care about has to produce evidence, or it is not a hiring criterion, it is a preference.

Applied to a single role, it looks like this.

A note on that last row, because it is where teams overreach. Personality and cultural assessments are qualitative and do not produce a total score. Treating one as a pass mark is a misuse of the instrument, and it is the fastest way to rebuild the old proxy problem with a new tool.

Skill-based recruitment: a five-step rollout

1. Run a job analysis before anything else. List what the person will actually do in the first 90 days, then name the competencies each task needs. The US Office of Personnel Management builds its assessment guidance on job analysis and warns that not all assessment tools are appropriate for every job and organizational setting.
2. Rewrite the posting around outcomes. Replace "bachelor's degree required" with the work itself and the level of ability you need to see. Our job description generator is a quick way to draft that version.
3. Put the measured screen first. This is the step the 45 percent skipped. The assessment goes before resume review, not after it, or the resume quietly keeps deciding who gets assessed.
4. Set the cutoff in advance and write it down. Pick the benchmark before you see the candidates. Weighted scoring, section-level breakdowns and percentile benchmarks are only useful if the threshold was set while nobody had a favorite.
5. Keep the decision human and reviewed. Scores order a shortlist. People make the hire, with more than one reviewer on anything senior. Pair the assessment with a structured interview and keep the questions identical across candidates.

Pro Tip: Run your new screen against your last cohort of hires before you run it on strangers. Score five people you already employ, three who are thriving and two who struggled, without telling the scorer which is which. If the assessment cannot separate them, it will not separate candidates either, and you have found that out for free.

Skills tests for employment: what to measure

Measure the work, as closely as you can get to it. A skills test for employment should produce output that resembles something the person would actually produce on the job: a query that returns the right rows, a support reply a customer would accept, a short plan a manager could act on.

Keep the instrument matched to the claim. Cognitive and aptitude tests say something about general reasoning and nothing about whether a candidate can do this specific job. Coding tests measure applied programming skill. Work simulations get closest to the real thing and cost the most to build. How accurately a score forecasts performance on the job is the property that matters most, which is why a borrowed test for a role you have not analyzed is close to worthless.

Soft skills in recruitment: can they be measured?

Partly, and the honest answer matters more than a confident one. Communication, collaboration and judgment can be observed in a structured setting, which is not the same as scored on a scale. A structured interview measures job-related competencies by asking systematically about behavior in past experiences or proposed behavior in hypothetical situations, and its value comes from every candidate facing the same questions so they are assessed accurately and consistently.

What does not work is asking a panel to rate "culture fit" after a loose chat. That is the proxy problem again, wearing a friendlier label. Define the behavior you need, ask every candidate the same question about it, and score against a written standard. Soft skills and how they differ from hard skills are worth getting straight before you try to test either.

Skills-based hiring testing: picking the method

Match the method to the role's risk and volume. High-volume roles justify an automated first screen because the time saved is real and the stakes per decision are low. Senior roles deserve a work sample and more than one reviewer, because a bad hire is expensive and a single score is thin evidence. Assessment procedures narrow large applicant pools cost-effectively, and as OPM puts it, quantity of applicants does not guarantee quality.

Where candidates sit the assessment unsupervised, decide what integrity controls you need before you launch, not after a result looks odd. And if you already run an applicant tracking system, keep it as the system of record and let the assessment layer feed it; Testlify integrates with the ATS you already use rather than asking you to move.

How do you test skills fairly and legally?

This is the section most articles on skills-based hiring leave out, and it is the one that can cost you money. When you replace a degree screen with a test, the test becomes the selection procedure, and employment discrimination law applies to it directly.

Under Title VII, a neutral test or selection procedure that disproportionately excludes people by race, color, religion, sex or national origin is unlawful unless it is job-related and consistent with business necessity. That is disparate impact, and it does not require anyone to have intended harm. The Uniform Guidelines set out three recognized routes for showing job-relatedness, together called test validation, which is the formal version of the job analysis in step one above.

Three practical duties follow:

• Keep the records. Track selection rates by group for every stage of your process. You cannot notice an adverse-impact problem you are not measuring, and you cannot defend a test on validity grounds without the data.
• Validate against the job, not against intuition. The defense is evidence that the test predicts performance in this role. A test chosen because it felt rigorous is the weakest possible position.
• Watch the job ad as carefully as the test. It is illegal to publish an advertisement that discourages applicants on protected grounds, and the EEOC's own example is pointed for anyone modernizing job posts: a help-wanted ad seeking "recent college graduates" may discourage people over 40 from applying and may violate the law.

That last one catches teams by surprise. A company proud of dropping its degree requirement can still be running an age-exposed posting because somebody wrote "new grad" in the title.

None of this is a reason to keep screening on credentials. A degree requirement with no validity evidence behind it is not a safer legal position than a validated test, it is just an older one that nobody has asked you to justify yet.

Hire on evidence, not on job titles

If there is one thing to take from the audit, it is that the announcement is not the work. The firms that got the retention gain and the wider pool rebuilt their screening step. The ones that just edited job postings got nothing, and a fifth of them went backwards.

Testlify is where the measurement part happens: role-based and custom assessments, coding tests, work simulations and situational judgment questions, with weighted scoring, section-level breakdowns and percentile benchmarks so a cutoff means something. Reviewers score as a panel and can override the AI. It also runs conversational AI interviews across chat, voice and video, and it feeds the ATS you already run.

Pick one role, run a job analysis, build the screen, and set the cutoff before you meet anyone. Book a walkthrough if you want to see how the scoring and benchmarking work on a real role before you commit a req to it.

FAQs