Only 37% of employers now rate resume credentials among their most reliable signals of talent, according to Willo's Hiring Trends Report 2026. And yet 59.6% of hiring teams still lead their process with a resume review, which means the resume has become a gate the reviewer no longer respects but still uses.
That gap between what employers trust and what they screen with is the whole game for anyone applying in 2026. This piece walks through the numbers from Willo (2.5 million candidate interviews, 100+ hiring pros at employers including Toyota GB and Microsoft), the contrarian Harvard finding underneath the "skills-based hiring" narrative, and the three-layer sandwich that gets you past both filters.
Why 37% resume trust is a ceiling, not a floor
The 37% figure is the ceiling on how much employers believe the resume. The floor on how much they use it is 59.6%, because when 76.6% of hiring teams see AI-generated applications at least occasionally, the resume is still the cheapest way to cut volume. Distrust does not equal disuse.
Willo's 2026 data quantifies exactly where trust sits now:
| Signal or method | Employer trust or usage | Source |
|---|---|---|
| Employers still leading with resumes | 59.6% | Willo 2026 |
| Employers actively moving away from resume-first | 41% | Willo 2026 |
| Employers who have largely replaced resumes with skills or scenario tests | 10% | Willo 2026 |
| Employers exploring alternatives to resumes | 15% | Willo 2026 |
| Trust in behavioral interviews as top signal | 67.7% | Willo 2026 |
| Trust in credentials (resume) as top signal | 37% | Willo 2026 |
| Trust ratio: behavioral interview vs resume credential | 1.83x | Derived |
Read the last row twice. Employers now trust a well-answered behavioral question 1.83 times more than the resume that got the candidate into the room. That is the largest gap in the report, and it dictates where your prep hours should go.
Willo's 2026 data shows 67.7% of hiring teams rank behavioral interviews as most reliable, against 37% for resume credentials.
The mechanism: AI applications broke volume filtering
76.6% of hiring teams report encountering AI-generated or AI-assisted applications at least occasionally in 2026. That flood is why the resume survives as a gate: no team can behaviorally interview 4,000 applicants. So the resume gets used the way a bouncer uses a dress code. Nobody thinks the jacket makes you interesting. It just keeps the line moving.
Job seekers who read "37% trust" and stop tailoring their resume get filtered out before the trust layer even engages. The correct move is a resume that survives the gate cheaply and a proof layer that shows up the moment a human clicks through. Writing that resume from scratch for every posting is the exact work Refolk takes off you: paste the job description, get your own history rewritten into a version tuned for that specific screen.
The skills-based hiring gap Harvard actually measured
Skills-based hiring is mostly policy theater until you find the roughly 37% of employers who followed through. Harvard Business School and the Burning Glass Institute reviewed more than 11,000 US job postings across nine years and found that dropping a degree requirement raised the share of non-degreed hires by only 3.5 percentage points, or roughly 97,000 people out of 77 million hires in a year. That is fewer than 1 in 700.
But the average hides the win. Nearly all of that movement came from about 37% of the companies that made the change, and those firms increased their share of workers without degrees by close to 20%. The other two-thirds edited the job description and kept hiring the same people. Harvard and Burning Glass also found that non-degreed workers hired into formerly degree-required roles had retention 10 percentage points higher than degree-holding peers over two years, which is the quiet reason the follow-through employers keep going.
The takeaway for a job seeker without a traditional pedigree:
- Targeting companies that "removed the degree requirement" is a losing strategy on average.
- Targeting the subset that actually shifted hiring is a strong strategy.
- Public evidence of follow-through beats stated policy. Look for named non-degree hires, promotion posts, and state-level programs.
Two concrete follow-through employers to name in your search:
- Pennsylvania state government opened 92% of positions to non-degree candidates after dropping the requirement.
- Maryland state government saw 41% more non-degree hires in year one.
And the counter-stat that keeps this honest: as of November 2025, Indeed Hiring Lab still finds roughly 1 in 5 US job postings list a bachelor's degree or higher as the true minimum. Skills-based hiring is real. It is also a minority sport.
Harvard and Burning Glass reviewed 11,000+ US postings over nine years. Most "skills-based" policies changed language, not hiring.
The behavioral interview is doing the real filtering
The screening layer that actually decides who gets hired in 2026 is the behavioral interview, not the skills test. Willo shows 67.7% of hiring teams rank behavioral interviewing with specific examples as their most reliable assessment method, cultural alignment and soft skills at 61.3%, and only 10% of employers have moved primarily to skills or scenario tests.
That means the "skills-based hiring 2026" narrative is directionally right and tactically misleading. If you spend your prep hours on TestGorilla-style assessments and neglect STAR stories, you are optimizing for the wrong 10%.
The verification methods Willo's respondents use to identify AI-generated applications tell you where to invest:
- Behavioral interviewing with specific examples: 67.7%
- Hands-on skills demonstrations: 54.8%
- Video assessments (used at the top of Willo's own funnel)
- Situational questions requiring personal context
Notice what these have in common: they all reward someone who can produce a real, dated, personal story on demand. They all punish someone who wrote a generic application and cannot back it up.
Practical prep, in order of leverage:
- Write out 8 to 12 STAR stories from the last three roles, each anchored to a metric, a date, and a named person or system.
- For each story, prepare a 45-second version and a 3-minute version.
- Record yourself telling them once. Video assessments are already normal in Willo-powered funnels.
- Attach one artifact per story: a Loom, a screenshot, a GitHub link, a case study PDF.
The 10% skills test gets the headline. The 67.7% behavioral interview does the actual filtering.
The AI application flood punishes volume, rewards personalization
Willo's 2026 report names three top trends: demand for AI-proof candidate assessments, an increase in candidate quantity paired with a decrease in perceived quality, and a growing focus on employee upskilling. Translation: the more you look like the AI-spray pile, the less you get read.
76.6% of hiring teams see AI-generated applications at least occasionally. That number is the reason a generic ChatGPT cover letter now hurts you. Recruiters have learned the tells: identical structure across five candidates, buzzwords with no dated example, cover letters that describe the company's mission back to them.
What still works in 2026 is personalization with proof:
- One paragraph that names something specific about the team, the product, or a recent change (a launch, a filing, a hire).
- One paragraph that names your closest analog, with a metric and a date.
- One artifact linked inline (a repo, a Loom, a memo, a Figma file).
This is another place where tooling matters. Refolk drafts a cover letter tuned to the specific posting rather than a template with the company name swapped in, and it scores how well your history actually fits the role before you send. If the fit score comes back weak, that is a signal to spend your time on a stronger match, not to polish a bad application.
Resume vs skills assessment: pick the layer your target company actually uses
The resume vs skills assessment debate assumes an either/or. In practice, most 2026 processes stack both, with a behavioral interview in the middle doing the real work. Here is the layered reality for the 41% moving away from resume-first hiring:
| Stage | What it screens for | What to prep |
|---|---|---|
| Resume (still 59.6% of gates) | Baseline coherence, keyword match, timeline sanity | Tailored one-pager per posting |
| Behavioral or video screen (67.7% trust) | Real examples, personal context, story specificity | 8 to 12 STAR stories with artifacts |
| Skills or scenario test (10% primary, more secondary) | Hands-on demonstration | Domain reps, portfolio, dated projects |
| Human final decision (78.7% insist on this) | Judgment, culture, references | References, thoughtful questions |
78.7% of hiring professionals in the Willo sample believe final decisions must remain human-led, and not a single respondent thinks automation can effectively run all hiring stages. The human on the other end wants a story they can retell to their team.
The recruiter-side gap you can exploit
The employer side is not as ready for skills-based hiring as the headlines suggest, and that inexperience is an opening for candidates who show up with their own proof. In Refolk's index of professional profiles, roughly 89,419 US-based profiles hold a Recruiter title. Only about 74 profiles across the entire global index mention "assessment" in their headline alongside a recruiter or talent-acquisition title. That is fewer than 1 in 1,200 recruiters publicly signaling assessment-design expertise.
The mechanism this creates:
- Most "skills-based" processes on the employer side are being run by people whose entire background is resume screening.
- Those recruiters are anxious about the AI application flood and understaffed on scoring rubrics.
- The candidate who arrives with their own proof artifacts (case studies, GitHub, a Loom walkthrough of a past project, a two-page memo) is making the recruiter's job easier.
- Candidates who make the recruiter's job easier get advanced.
This is the practical version of "personalization beats polish." A tailored resume plus one artifact is worth more than five polished but generic applications.
Build the sandwich
If you take one thing from Willo 2026, make it this three-layer stack:
- Top slice, the resume: tailored to the specific posting, keyword-clean, timeline-honest, one page for most roles. Gets you past 59.6% of gates.
- The filling, proof: 8 to 12 STAR stories with artifacts attached. Wins the 67.7% behavioral trust layer.
- Bottom slice, the demonstration: a portfolio, repo, or scenario response ready when 10% of employers ask. Also useful for the other 90% as a differentiator.
The candidates who treat the resume as dead lose the gate. The candidates who treat the skills test as the main event lose the filter in the middle. The 2026 winner sits in the middle and brings receipts.
FAQ
Does the resume still matter in 2026?
Yes, but not for the reason it used to. Only 37% of employers rate credentials as a top reliability signal, but 59.6% still lead their hiring process with a resume review because it is the cheapest volume filter when 76.6% of teams face AI-generated applications. The resume is now a gate rather than a decision, so tailor it per posting to pass the gate cheaply and put your energy into the behavioral and proof layers behind it.
What does the Willo Hiring Trends Report 2026 actually say about skills-based hiring?
The report, based on 2.5 million candidate interviews and more than 100 hiring pros at employers including Toyota GB and Microsoft, found 41% of employers are actively moving away from resume-first hiring and 10% have largely replaced resumes with skills or scenario assessments. Another 15% are exploring alternatives. The trend is real, but 59.6% still lead with the resume, so most 2026 processes stack a resume gate, a behavioral interview, and sometimes a skills test rather than replacing one with another.
Resume vs skills assessment: which should I prepare for first?
Prepare for the behavioral interview first, because it is the layer employers trust most (67.7%) and it filters more candidates than either the resume gate or the skills test. Build 8 to 12 STAR stories with dated metrics and named systems, then attach one artifact per story (a Loom, repo, memo, or case study). After that, tailor your resume per posting to survive the gate, and keep a portfolio ready for the 10% of employers who run a skills test as the primary filter.
How do employers detect AI-generated applications, and how do I avoid getting flagged?
Willo's 2026 respondents use behavioral interviewing with specific examples (67.7%), hands-on skills demonstrations (54.8%), video assessments, and situational questions requiring personal context. The common thread is verifiable personal detail: dated stories, named collaborators, real metrics, and artifacts you can walk through on camera. To avoid the flag, keep your applications personalized with one specific reference to the team or product, one closest-analog story with a number, and one linked artifact, rather than sending a high-volume generic spray.