The Skills Employers Say They Want vs. the Skills That Actually Get Hired

Employers say they want communication, teamwork, and problem-solving, and surveys confirm it. Candidates get hired, however, when they prove those skills with specific evidence: internships, work samples, measurable outcomes, and references.

The stated skill list opens the conversation; demonstrated proof, and a hiring manager’s confidence in it, closes the deal.

Every hiring season follows the same script. Employers publish wish lists, career centers translate them into advice, and candidates fill résumés with the approved vocabulary. The outcomes then diverge from the script in ways few participants discuss openly.

The divergence is not a conspiracy or a lie. It is a structural feature of how hiring works, and understanding it changes what a candidate should do, what a hiring team should measure, and which claims in the career-advice industry deserve skepticism.

The Wish List Is Real, and It Is Long

The most recent spring update to the National Association of Colleges and Employers (NACE) Job Outlook survey found that employers want entry-level candidates who can work in a team, solve problems, and communicate effectively.

However, the wish list is extensive, with more than half of respondents citing ten or more skills. The same research shows that technical skills, work ethic, and analytical or quantitative ability also appear among the attributes employers value.

Two details in that data matter more than the headline ranking. First, a list of ten or more required skills is not a priority list. It describes a screening document, not a decision. Hiring ultimately forces a ranking, and the ranking happens in a conversation the candidate never sees. Second, the survey’s authors stress that listing skills is not enough. Shawn VanDerziel, NACE’s chief executive, emphasizes that employers want to see examples.

The tiebreaker is revealing. NACE research consistently finds that when two candidates are equally qualified, employers overwhelmingly choose the one with internship experience. An internship is not a skill. It is a compressed record of skills observed by someone other than the candidate, which is exactly the kind of evidence a wish list cannot supply.

Where the Gap Actually Opens

The disconnect between stated and rewarded skills is often framed as employers asking for the wrong things. The data points elsewhere: employers and candidates largely agree on what matters, and disagree on whether candidates have it.

NACE’s paired surveys of students and employers show agreement on the importance of communication, critical thinking, teamwork, and professionalism. The disagreement is over proficiency. In the pairing of a 2024 student survey and a 2025 employer survey, the proficiency perception gap for leadership and professionalism approaches or exceeds 30 percentage points, and for critical thinking and communication it is roughly 25 points.

Earlier waves showed the same pattern at a larger scale. In one, roughly 89 percent of students rated themselves proficient in professionalism and work ethic, while about 42 percent of employers rated recent graduates the same way.

Technology is the notable exception. Employers rate new graduates as more proficient in technology than the graduates rate themselves, which suggests the digital-native assumption is holding up while the interpersonal skills candidates feel most confident about are the ones employers doubt.

The practical implication is uncomfortable. A résumé line reading “strong communicator” does not merely fail to help. It invites a skepticism that the data suggests is already present. Every candidate claims the skill, and a large share of employers do not observe it. Claims that cost nothing to make carry little weight in a market where the buyer has learned to discount them.

Skills-Based Hiring: Announced More Than Practiced

The most important counterweight to the wish-list narrative is the growth of skills-based hiring, at least on paper. Among employers in NACE’s most recent Job Outlook survey, 70 percent report using skills-based hiring, up from 65 percent the year before.

A joint study by the Burning Glass Institute and Harvard Business School complicates that figure. Examining job postings and actual hires, the researchers found that companies increased the share of workers without a bachelor’s degree by only about 3.5 percentage points, translating to opportunities for roughly 97,000 workers out of 77 million annual hires, or fewer than 1 in 700 hires in 2023. Postings told a different story: between 2014 and 2023, the number of jobs from which employers dropped degree requirements increased fourfold.

The gap between posting and practice is concentrated. Roughly 37 percent of the companies studied, labelled skills-based hiring leaders, made changes and followed through on their commitments. A trade analysis of the same research reported that about 45 percent of employers that championed skills-based hiring were classified as doing so in name only, with no meaningful change in hiring behaviour.

Matt Sigelman, president of the Burning Glass Institute, summarized the problem in plain terms: the shift is not working out quite the way people planned, partly because the in-house work to support the practice lags behind executive endorsement. Changing a job posting takes an afternoon. Changing the interview rubric, the recruiter training, and the hiring manager’s instincts takes years.

The reward for doing it properly is measurable. At the leading firms, non-degreed workers hired into roles that formerly asked for degrees had retention rates 10 percentage points higher than their degree-holding colleagues.

Separate Boston Consulting Group research, as reported by Business Insider, found that skills-based hires are promoted at about the same pace as traditional hires but stay with an employer 9 percent longer.

What Hiring Managers Actually Screen For

Between the wish list and the offer sits a screening process with its own logic. Three mechanisms explain most of the distance between what employers say and what gets rewarded.

The first is job-specific competence. NACE’s chief executive has noted that in addition to job-specific skills, employers seek critical thinking, communication, and teamwork.

The order in that sentence is easy to miss. Transferable skills operate as a multiplier on a baseline of role-relevant capability, not a substitute for it. A candidate with excellent teamwork and no evidence of doing the work rarely advances.

The second is vocabulary matching. NACE research found that more than 80 percent of employers highlight the key skills needed for the position when writing job descriptions.

Applicant tracking systems and recruiters scanning at speed tend to reward candidates whose résumés mirror that language. This is the legitimate half of résumé optimization: reading the posting as a specification and using its terms accurately. Résumé checker tools can help with the mechanics, but they measure overlap, not persuasion.

The third is risk reduction. A hiring manager choosing among finalists is rarely maximizing upside. The goal is avoiding a bad hire, and evidence that someone has already done a similar job under similar conditions is the strongest available hedge.

This explains the internship tiebreaker, the persistence of degree preferences even where requirements are dropped, and the weight given to referrals and prior employers. David Deming, a political economy professor at Harvard’s Kennedy School, has made the related point that hiring managers may still care whether an applicant has a degree even when it is no longer formally required.

AI Skills: A New Line Item With an Old Problem

Artificial intelligence has become the fastest-moving entry on the wish list. NACE found that 31 percent of employers are seeking candidates with AI skills when reviewing résumés, and its spring update describes AI as increasingly an expectation for early-career talent, shaping both the job market and entry-level work.

The proof problem applies here with extra force. Listing familiarity with a popular AI tool is the 2020s equivalent of listing proficiency in word processing: it signals nothing because everyone can claim it. Evidence looks different.

A candidate who can describe a workflow they redesigned, how they checked the output for errors, and what changed in speed or quality has supplied the kind of artifact an interviewer can probe. Employers are also cautious about their own use of AI in recruiting, as a NACE quick poll on the subject found, which means candidates cannot assume automated screening will surface nuance that a clear, specific résumé would not.

The Proof Ladder: A Framework for Converting Claims Into Hires

The pattern across the research reduces to a simple structure, which can be called the Proof Ladder. Each rung makes a skill claim more credible, and most candidates stop on the first.

The first rung is the named skill, such as “problem-solving” in a skills section. It costs nothing and carries almost no weight on its own.

The second rung is the dated example. The claim attaches to a specific situation: a project, a semester, a client, a quarter. This is the level NACE’s advice points to when it asks candidates for examples, and it is the minimum needed to survive a behaviour-based interview question.

The third rung is the measured result. The example includes a number, a before-and-after comparison, or a concrete change in how something worked. A candidate who reorganized a customer-support queue and can state the effect on response times has moved from describing activity to describing impact. Figures should be exact or honestly approximate; invented precision collapses under the first follow-up question.

The fourth rung is third-party verification: a reference who can confirm the story, a portfolio the interviewer can open, a certification from a recognized body, a supervisor’s written feedback, or a work sample completed under observation. This is the rung where hiring risk actually falls, and it is the rung most candidates never deliberately build.

The framework also works in reverse for employers. A hiring process that never asks for rungs two through four is effectively screening on rung one, which is the least informative signal available.

Common Mistakes on Both Sides

Candidates commonly list soft skills as adjectives instead of demonstrating them through outcomes. They also treat keyword matching as the whole strategy, producing résumés that pass automated filters and then stall in human review because nothing in them is checkable.

Credentials are another trap. Certifications can legitimately substitute for a degree in some roles, and Savvas Learning Company chief executive Bethlam Forsa has suggested that employers could look to certifications an applicant has earned as a way to hire for skills rather than a degree.

Yet one higher-education analysis argued that short-term certificates and skills-based hiring have underperformed relative to the hype surrounding them. A certificate is rung four evidence only when the employer recognizes the issuer and the candidate can show applied work behind it. Collecting low-cost online credentials in bulk adds length to a résumé without adding proof.

Employers make a parallel set of errors. Posting ten or more required skills for an entry-level role produces a screening document no real candidate satisfies. Removing a degree requirement without retraining interviewers produces the pattern the Harvard and Burning Glass research documented: a policy announcement with no change in who gets hired.

What the Data Cannot Say

Several limits deserve plain acknowledgment. NACE’s surveys are self-reported and focused on entry-level, college-graduate hiring, and the spring update drew on responses from 185 employers, a respectable sample but not a census.

The Burning Glass and Harvard analysis covered large firms and a period ending in 2023, so it may understate recent change. Mid-career and senior hiring operates differently, with track record, network, and referrals carrying more weight than any skills list.

There is also a fair counterargument to the tone of this analysis. A degree is not merely a proxy for laziness in hiring; it carries real information about persistence and baseline knowledge, and some employers reasonably continue to value it. The case for skills-based hiring rests on the view that the proxy excludes capable people, not on the view that credentials mean nothing. Both positions can hold at once.

The Practical Takeaway

The stated and rewarded skill lists are the same list read at different levels of resolution. Employers say they want communication, problem-solving, and teamwork because they do. They reward the candidates who make those skills observable, specific, and verifiable, because observation is the only thing that reduces the risk of a wrong hire.

For candidates, the work is to climb the ladder before the interview: pick two or three skills the target posting emphasizes, attach a dated example and a measured result to each, and line up someone who can confirm it. For employers, the work is less glamorous and involves rewriting rubrics, training interviewers, and shortening wish lists until they describe an actual decision.

The organizations that do this are the ones the research identifies as getting retention and promotion benefits from skills-based hiring. The ones that only edit the job posting are largely getting the old hiring process under a new name.

What People Ask