Employers Want Skills. Your Certification Isn't Proof Yet.

Every few years the hiring world announces that the degree is over.

This time the announcement arrived with better numbers than usual. Seventy percent of employers now say they hire for skills rather than credentials, up from 65% a year earlier, and 53% say they have eliminated degree requirements outright, up from 30%. Those figures come from the National Association of Colleges and Employers' Job Outlook 2026 survey, reported by Forbes this month, and the argument for why now is genuinely new. AI is changing job requirements faster than a four-year degree can refresh, and more than a third of entry-level postings already ask for some AI competency, nearly triple the share from a year ago.

If you run an association with a certification, this should be the best news you have heard in a decade. Your profession's credential is suddenly the thing employers claim to want.

The last time employers said this, almost nothing happened

In 2024, Harvard Business School's Project on Managing the Future of Work and the Burning Glass Institute studied what actually changed at firms that had removed degree requirements. Removing the requirement raised the share of hires without a bachelor's degree by 3.5 percentage points. Across the whole labor market, the promised opportunity showed up in fewer than one in 700 hires.

The breakdown is the interesting part. Thirty-seven percent of the firms studied drove nearly all of the real change. Forty-five percent changed the job posting and nothing else. Twenty percent hired more non-degreed workers for a while and then reversed course.

So the announcement was real and the behavior was not. The usual reading of that finding is that employers are hypocrites, or that credentials do not really matter. Both are wrong, and the correct explanation is the one that should keep association executives up at night.

Why do employers ignore the credentials they say they want?

In March 2026, OneTen and the Burning Glass Institute analyzed more than 1,000 major U.S. employers and named the missing ingredient. Credential fluency is the ability to identify, validate, and hire based on quality credentials. Firms in the top decile of credential adoption hire non-degreed workers at a rate eleven percentage points above the bottom decile, and their skills-based hires stay 20% longer. The firms that get this right have done the work of learning what credentials actually mean.

The report puts the failure in one sentence.

Hiring managers don't trust what they don't understand.

OneTen and the Burning Glass Institute, March 2026

Then it supplies the reason they cannot understand. There were 1.1 million distinct credentials available in the United States by 2024. Most firms that ask for a credential in a job posting show no correlation with actually hiring credentialed people.

Read that again. Asking for the credential and hiring for it are statistically unrelated.

What has broken here is signal. When a hiring manager sees four letters after a name and has no way to know what they certify, how current they are, or who stands behind them, the letters carry no information, so the manager falls back on the one credential they believe they understand. The problem is not that there are too few credentials. It is that there are too many to read.

Your credential is one of 1.1 million

Association executives tend to hear that number and feel exempt from it. Ours is different, the thinking goes. Ours has a psychometrically validated exam, a job task analysis, an accredited standard, a recertification cycle, and forty years of professional consensus behind it. All of that is usually true, and none of it helps.

Rigor is what makes a credential worth trusting. Legibility is what lets somebody outside the profession recognize that in the eleven minutes they spend on a stack of applications. They are different properties, and associations have spent decades investing in the first on the reasonable assumption that the profession would vouch for the credential on their behalf.

That assumption held when hiring managers came up through the profession. It does not hold when a generalist recruiter, or an applicant tracking system, is doing the first cut.

What your certification actually sells today

Here is the honest shape of most association certification programs. A member pays a fee. They receive a body-of-knowledge outline, perhaps a self-paced course or a practice exam, a testing window, and on the other side a set of letters plus a continuing-education obligation that renews annually.

What that produces is a transaction.

I want to be careful here, because certification is often the healthiest non-dues revenue line an association has, and I am not arguing against it. The exam is real, the standard is real, and the rigor is usually the best work the organization does. But ask what the transaction produces that a hiring manager can actually see, and the answer is a line on a resume with no practice behind it.

A gym membership and a training group both cost money and both involve exercise. Only one of them has people who notice when you stop showing up. Associations have been selling the membership and calling it the program.

The consequence shows up twice. Candidates do not finish, and the ones who do carry a credential nobody can read.

What happens when candidates prepare alone

Self-paced credential prep does not work well, and the research on this is not close. A meta-analysis of 225 studies published in the Proceedings of the National Academy of Sciences found failure rates fell from 33.8% under passive instruction to 21.8% under active, structured formats. A nine-year tracking study published in Science found completion in large-scale passive online courses sitting around 3%.

A certification cohort is a small group of candidates, typically six to ten, preparing for the same exam window on a fixed meeting cadence with a facilitator who already holds the credential. Cohort formats report completion in a completely different range than self-paced study, and the mechanism is social accountability rather than better content. Eight people who agreed to sit the same exam in April behave differently than one person with a login.

It is worth noting that most cohort completion figures come from providers reporting on themselves, so treat the high end skeptically. The direction of the effect is not in doubt. The magnitude is.

What would a hiring manager need to see?

Almost every association publishes its body of knowledge in the language of the profession, written for candidates. Almost none publishes what the credential means in the language of somebody holding forty applications and eleven minutes. That document is the cheapest item on this list to produce and the closest thing to a free win available to a certifying body right now.

It answers five questions, on one page, in public:

• What can this person do unsupervised? Named tasks, not domains. "Can independently reconcile a payer contract variance" beats "Domain III: Revenue Cycle."

• What was tested, and what was not? Scope honesty builds more trust than scope inflation. Say what the exam does not cover.

• How current is it? The recertification cycle, stated as a date rather than a policy.

• Who stands behind it? The accreditation, the number of holders, and how many are in this metro area.

• How hard is it? The pass rate. Associations hide this, and it is exactly the number that separates a credential from a certificate of attendance.

Credential fluency is a supply problem, and associations are the supply. Nobody else can speak for what your credential certifies. If your profession's credential is illegible to employers, the remedy is a document your organization has not written yet.

Three shifts worth making

1. Stop selling the exam. Run the cohort.

The certification is the reason a member shows up. The cohort is the reason they finish, and the reason they are still there afterward. PMI administers the largest professional credential in the world and has been building group formats around it rather than treating the exam as the entire relationship. If the biggest certifying body in any profession has concluded that the exam alone will not carry the relationship, that conclusion probably travels down to 3,000 members too.

This is the same argument I have made about discussion boards, arriving from a different direction. The container determines the behavior.

2. Keep the cohort after the exam.

Continuing education is currently a compliance mechanism that members resent and staff administer. The same hours, delivered as a standing peer group of people who passed together, produce something a hiring manager can actually read: evidence that the credential holder is still practicing among peers rather than banking hours before a deadline.

The World Economic Forum's Future of Jobs Report 2025 projects that 39% of workers' core skills will be disrupted by 2030. A credential that certifies what somebody knew in 2024 is a depreciating asset. A credential attached to a group that meets monthly is not. That distinction is going to matter more every year, and it is the one thing a credential aggregator or an AI tutor cannot replicate, because it is made of people who know each other.

3. Change the metric.

Most certification dashboards report credentials issued, exam volume, and pass rate. Those are production metrics. They tell you about last year's marketing.

The number that predicts whether your credential still carries weight is how many credential holders engaged with the profession this month. Not logged in. Engaged, in a group, with other holders. If that number is small, the letters are drifting toward decoration no matter how rigorous the exam was, and no amount of recertification paperwork will stop it.

What to do this quarter

• Write the employer-facing one-pager. The five questions above, on one page, published publicly. No budget required, no platform required, no board approval required. Do this one first.

• Publish your pass rate. If it is high, it is proof of preparation quality. If it is low, it is proof of rigor. Either way it is a signal, and silence is not.

• Pick your next exam window and run one cohort into it. Six to ten candidates, a fixed weekly slot, one volunteer chair who already holds the credential. Compare completion against your self-paced baseline. One cohort is enough to know. Our certification prep work exists for exactly this, but a recurring calendar invite and a spreadsheet will prove the concept before you buy anything.

• Convert one continuing-education requirement into a peer group. Same hours, same standard, different container. Then measure renewal among the people who were in it against everyone else.

Employers spent the last two years announcing that they would hire for skills and then hiring exactly as before, because a credential they cannot interpret is not a signal. That is a failure of translation, not a failure of standards. And translation is work an association can do this month, without a strategic plan.

Many associations already own the standard. What it has not built is the thing that makes the standard visible: a group of practitioners, still in the room, still current, that an employer can point at.