The Assessment Gap: Why Skills Shortages Are Getting Worse Even as Change Slows Down

The World Economic Forum's 2025 Future of Jobs Report contains a contradiction worth sitting with. Skills are becoming more stable. Employers are more worried about skills gaps than ever.

Between 2023 and 2025, the expected rate of worker skill disruption fell from 44% to 39%. Companies have absorbed remote work, adopted AI tools, and built continuous learning programs. The workforce is settling after years of upheaval.

In the same period, the share of employers naming skills gaps as their primary barrier to transformation rose from 60% to 63%. It now ranks as the single biggest obstacle to business success across 52 of 55 economies surveyed and 19 of 22 industry sectors.

Change is slowing while the shortage worsens. That points somewhere specific: not at the pace of change, but at our ability to assess and hire for the skills we need.

Is the workforce actually stabilizing?

Partly, and the trajectory is more interesting than the level. Based on the WEF survey of over 1,000 global employers representing 14 million workers, two-fifths of the skills workers need today will transform or become outdated by 2030. That is still real disruption.

Skill instability peaked at 57% in 2020, dropped to 44% in 2023, and now sits at 39%. Companies are getting better at anticipating which skills matter and investing accordingly. 50% of workers have now completed some form of upskilling, up from 41% two years ago.

So the puzzle stands. The skills that are changing are not changing randomly. They are shifting in specific directions that traditional hiring is badly equipped to see.

Which skills are growing, and can you screen for them?

The ten fastest-growing skills through 2030, by net increase in importance:

1. AI and big data (87%)
2. Networks and cybersecurity (70%)
3. Technological literacy (68%)
4. Creative thinking (66%)
5. Resilience, flexibility, and agility (66%)
6. Curiosity and lifelong learning (61%)
7. Leadership and social influence (58%)
8. Talent management (58%)
9. Analytical thinking (55%)
10. Environmental stewardship (53%)

Half are technical skills that did not appear on most job descriptions ten years ago. The other half are cognitive and socio-emotional capabilities that have always mattered and rarely get assessed systematically.

The decline side is just as telling. Manual dexterity, endurance, and precision show a net decline of 24%, the first time physical abilities have seen negative demand in the report's history. Reading, writing, and basic mathematics are slipping. So are dependability and attention to detail.

Which creates the specific problem. The skills employers need most, adaptive thinking and technological fluency and emotional intelligence, are precisely the ones resume screening and unstructured interviews cannot evaluate. You cannot keyword-search for resilience. You cannot verify creative thinking from a LinkedIn profile.

What happens to the eleven workers in every hundred?

They miss the training they need, and the reason is diagnostic. The WEF survey asked employers to estimate training needs across their workforce. Out of every 100 workers, by 2030:

• 41 will not need significant retraining
• 29 will be upskilled in their current roles
• 19 will be reskilled and redeployed to different functions
• 11 are unlikely to receive the training they need

That last group is not missing out because the training does not exist or because budgets ran out. Companies are investing: 85% of surveyed employers plan to prioritize workforce development, and 70% plan to hire new staff with emerging skills. They miss out because organizations cannot identify who needs what, or who could successfully move into a growing role.

Deciding who belongs in which bucket is an assessment task, and most companies do it with tenure, performance reviews, and manager intuition. None of those reliably predicts someone's capacity to develop a new skill or succeed in a different function.

Without a structured way to evaluate skills, a company cannot answer the three questions that matter: which current employees have the foundation to learn AI tools, who has the creative thinking and adaptability for strategic roles, and which candidates claiming five years of experience actually have the judgment the work requires. The WEF data shows analytical thinking treated as core by 70% of employers, resilience and flexibility by 67%, and leadership influence by 61%. None of them appears on a resume.

Why do job postings keep missing?

Because they specify proxies and then screen on them. A company posts for a data analyst and lists Python, SQL, three to five years, and a bachelor's degree. What the role actually needs, on the WEF findings, is analytical thinking, technological literacy, curiosity for continuous learning, and enough creative problem-solving to apply data to messy business questions.

The resume screen checks the boxes. Python, yes. Five years, yes. Degree, yes. It says nothing about whether the person can think analytically under ambiguity, pick up an unfamiliar tool, or explain a finding to a non-technical stakeholder.

Two failures follow. You hire someone who meets the requirements and lacks the capabilities. Or you screen out someone short on years who has exactly the cognitive and adaptive skills that predict success.

That mismatch is the likeliest explanation for why 37% of employers cite an inability to attract talent to their industry and 27% an inability to attract talent to their firm. The talented people exist. The process cannot find them.

Are technical skills easier to assess?

Easier in one way and harder in another. Technical competence is partly objective: you can test whether someone writes a working SQL query or understands network architecture.

The difficulty is that the definition of proficient keeps moving. What counted as advanced Python three years ago is now table stakes. The AI tools someone used last year may be retired before their start date.

Which is why technological literacy, meaning basic fluency with digital tools, now counts as a core skill for 51% of employers and is growing at a net 68%. That general fluency matters more than deep expertise in any single technology. The people who do well are the ones who learn new tools quickly and apply them to unfamiliar problems.

Traditional hiring misses this too. Postings name specific tools and years. They do not assess learning agility or problem-solving under uncertainty. They measure what someone has already done rather than what they can do next.

Where does demographic pressure make this worse?

In exactly the roles that are hardest to screen for. The report identifies two workforce shifts running at once: aging and declining working-age populations in high-income economies, and expanding working-age populations in lower-middle-income economies.

Those pressures drive specific demand. Aging populations raise demand for healthcare roles such as nursing professionals. Growing working-age populations fuel education roles such as higher-education teachers. Both need talent management and teaching or mentoring skills.

Both are also brutal to assess conventionally. How do you evaluate empathy? How do you verify a capacity for active listening, or a talent for mentoring? Those skills rank 7th and 16th among the core skills employers name, and most hiring processes evaluate them in unstructured interviews, where reliability is weakest.

The outlook reflects it: 42% of employers expect talent availability to decline over the next five years, a net negative outlook of -13%. Two years earlier, 39% expected improvement.

Does this look different by industry?

The assessment problem is the same everywhere; the combination it has to resolve changes.

Telecommunications needs rapid growth in design and user experience, networks and cybersecurity, and programming, all at roughly twice the global average, plus above-average resilience and adaptability. Screening for deep technical skill and the flexibility to pivot as infrastructure changes is a hard pairing.

Insurance and pensions management is starker. 83% of employers there name curiosity and lifelong learning as core, against 50% globally, and 97% expect AI and big data skills to rise in importance. The work demands constant learning, and hiring still leans on credentials and years in specific product lines, which is close to the opposite signal.

Financial services runs the same way. Creative thinking and analytical reasoning are both growing at over 80% in banking, while hiring still centers on transaction-processing experience and familiarity with legacy systems. The people needed for the next decade do not resemble the people who succeeded in the last one.

Why isn't 85% upskilling enough?

Because upskilling depends on two assessments most companies cannot make: who has the foundation to benefit from the training, and whether the training actually built the capability.

Without structured assessment, upskilling turns into a checkbox exercise. People go through courses, certificates get issued, and nobody can say who gained analytical thinking or whose adaptability improved. The budget is real and the measurement is not.

What does the guesswork cost?

The WEF projects structural transformation creating 170 million new jobs by 2030 while displacing 92 million, for net growth of 78 million positions, about 22% of today's total employment in churn.

Put a hiring process that cannot reliably evaluate the relevant skills against that much movement and the costs are predictable. Bad hires. Qualified internal candidates overlooked for reskilling. Recruiters paid to find unicorns who do not exist, because the job description lists credentials instead of capabilities.

70% of employers plan to hire staff with emerging skills. A screening process that cannot assess analytical thinking, technological adaptability, creative problem-solving, or resilience is not screening for emerging skills. It is screening for degrees, years, and titles that do not predict success. Which is how spending rises while accuracy does not, and why skills gaps remain the top barrier for 63% of employers.

What actually closes the gap?

Breaking the job into specific, measurable skills, then assessing each one directly, regardless of where the candidate learned it.

The WEF data shows 19% of employers now name removing degree requirements and running skills-based hiring as a promising practice for increasing talent availability. That is still low, ranking 12th of 16 business practices, and it points the right way. It also mirrors the wider gap between skills-first intentions and actual implementation across the research.

For a data analyst, that means dropping "three to five years experience" and naming the core skills: analytical thinking, technological literacy in SQL and Python specifically, creative problem-solving in business contexts, and the communication needed to translate technical findings. Then assessing each through targeted questions, work samples, or practical exercises.

Four things improve at once.

The talent pool widens. Removing credential requirements surfaces candidates the filter would have dropped. Someone with two years of experience and exceptional analytical thinking becomes visible. A career-changer with strong foundations becomes a candidate for an upskilling program.

Accuracy improves. The 2022 Sackett meta-analysis ranks structured interviews as the #1 predictor of job performance, more than twice as predictive as unstructured ones (0.42 against 0.19). The comparison there is between interview formats, and it is the best-evidenced upgrade available to a hiring team.

Internal mobility gets easier. A clear inventory of who holds which skills makes it possible to decide who to upskill, who to redeploy, and who needs something different. That is the direct route at the eleven per hundred who otherwise miss out.

The decision leaves a record. A skills-based assessment documents what was asked, what was said, and how it was scored against a written standard. That consistency is what makes a hiring decision explainable later, to a hiring manager, a regulator, or a rejected candidate.

What does it take to implement?

Real work upfront, which is why only 19% of employers have done it. I won't pretend otherwise. Renaming your job descriptions skills-focused does not count. The actual sequence:

1. Break the role into specific, measurable skills.
2. Weight those skills by priority and required proficiency.
3. Write structured assessment questions for each skill.
4. Train interviewers to evaluate answers against the same standard.
5. Combine the skill assessments into a decision by a fixed rule.

That is time-consuming when you are the hiring manager, and harder when you depend on recruiters without subject-matter depth in the role.

The alternative is hiring on credentials and gut feel while the gaps widen. 63% of employers already name skills gaps as their biggest problem. The open question is whether they address the assessment gap underneath it.

Traditional hiring worked reasonably well in a stable market, where job titles meant roughly the same thing across companies and years of experience tracked capability. That market is gone. The skills that matter most now, adaptability, analytical thinking, technological literacy, creative problem-solving, cannot be read off a resume.

So the shift is from proxies to evidence. Years of experience is a proxy for capability. A structured assessment that directly evaluates analytical thinking is evidence of it. Being specific is what makes that possible: not "five years of data analysis" but can this person structure an ambiguous business problem into testable hypotheses, implement a solution with current tools, and explain the result to someone non-technical.

How Ratio fits

Building a structured, skills-based assessment used to take subject-matter expertise and hours of work per role. Ratio does that preparation, working from whatever you have on the job: a job description, your notes, a hiring-manager transcript. It extracts and weights the skills the role needs for success, then builds a complete interview kit: targeted questions for each skill, planned follow-ups, and written scoring guides. Interviewers run the conversations and score the answers; Ratio never auto-scores candidates.

The assessment gap is solvable. The teams that solve it first will pull ahead.

References

• World Economic Forum. (2025). The Future of Jobs Report 2025. https://www.weforum.org/publications/the-future-of-jobs-report-2025/
• Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection. Journal of Applied Psychology, 107(11), 2040-2068. https://doi.org/10.1037/apl0000994