A professional certification is worth only if the person holding it matches the performance it promises. Employers hire on it, regulators rely on it, and candidates invest years in it because passing is supposed to signal capability. That link, between clearing an exam and being ready to perform, is quietly weakening, and it is starting to show in the numbers certification leaders watch most closely. Preparation has never been better resourced. Courses, question banks, mock exams, and video libraries are far more advanced than they were a decade ago. Pass rates, though, have largely held flat. On the CPA Exam, one of the most widely held credentials, the overall pass rate has sat near 50 percent, and the AICPA states plainly that when the number moves, it reflects how well candidates are prepared, not how hard the exam is.
Harbinger’s Power Hour, From Course Completion to Certification Readiness: Rethinking Candidate Success, examined why. The session was hosted by Shrikant Pattathil, President and CTO at Harbinger Group, with Angeline Brown, Vice President, Product at Becker Professional Education; Daniel Mummert, Director, Online Education at the American Academy of Ophthalmology; and Poonam Jaypuriya, Vice President, eLearning at Harbinger Group. Their discussion provides the foundation for the questions explored in this blog: what actually improves candidate readiness, how providers can act on earlier signals, and what this shift means for candidate outcomes and the long-term value of a credential.
Why the Completion-Readiness Gap Is a Business Risk
The commercial logic of a credential rests on one promise: the people who hold it can do the work. Each time completion climbs while pass rates stall, or candidates finish every module and underperform once the material is gone, that promise loses credibility, and the differentiation the credential competes on erodes with it. The exposure looks different from each seat: a CEO feels it as credential differentiation, a product leader as candidate experience, a technology leader as the architecture needed to connect content and signals, and a learning leader as the ability to prove capability rather than completion.
Buyers are already applying pressure. Deloitte’s 2025 Global Human Capital Trends found that 66 percent of managers and executives considered most recent hires not fully prepared, and named the experience gap, rather than the skills gap, as the harder one to close. As employers scrutinize whether a credential actually predicts performance, the burden of proof shifts to the providers that issue it.
The strain shows in credential data too. First-time pass rates for a leading behavior-analyst certification fell to 51 percent in 2025, the lowest the certifying board has recorded, even as candidate volume grew. Part of the cause is self-inflicted. When a program is judged on completion, everything organizes around completion: content shortens, assessments lighten, the number rises, and readiness does not. Optimizing the metric that is easy to move can quietly degrade the one that carries the value.
That observation frames the question every certification leader now faces. If more content is not the answer, what actually moves the number?
What Actually Moves Certification Pass Rates
Two levers move readiness, and neither is content volume. Both are within reach of programs that already exist.
The first is how comprehensively candidates use what they already have. Becker Professional Education studied tens of thousands of candidates and found that those who practiced across the full range of question types, including the harder application-style items rather than only familiar multiple choice, passed at materially higher rates. Candidates who meet its exam-day-ready standard pass at roughly 65 to 75 percent, against 45 to 50 percent for those who do not. Same materials, different use, a twenty-point swing. The pattern is consistent: candidates gravitate toward the formats they find comfortable and skip the harder application work that decides the exam.
The second lever is timing. Many programs discover gaps only after a high-stakes exam confirms them, when the room to intervene has already closed. Reading signals earlier keeps the outcome changeable.
The highest return on investment is often not more content, but program design that drives comprehensive practice and surfaces gaps while they can still be closed. That is the shift from a completion view of the candidate journey to a readiness view.
| Completion view | Readiness view | |
|---|---|---|
| Core question | Did the candidate finish? | Can the candidate perform without the material in front of them? |
| Primary metric | Modules and courses completed | Evidence of applied competency |
| When gaps surface | At the final score | Early, through low-stakes signals |
| Practice emphasis | Familiar question types | The full range, including harder application items |
| Provider’s role | Deliver content | Diagnose, intervene, and confirm |
The readiness view can sound expensive, which is the objection worth answering directly.
Building Readiness Without Rebuilding Your Program
Readiness rarely requires new content or a new platform. It requires connecting what already exists and running it as a cycle. Personalization is often imagined as a unique path for every learner, expensive to build and fragile to maintain, and that assumption is what makes the readiness view look unaffordable.
“We have digitized exam preparation much faster than we have redesigned it for an online world,” said Poonam Jaypuriya. “We took a textbook and converted it into an e-learning course, added a question bank, flashcards, and a mobile app. They are all useful, but the learner is still often left to put the pieces together. Personalization isn’t about giving every learner different content. It’s about giving each learner the right intervention at the right time.”
Two decisions make that affordable. The first is to start from existing assets. In many programs, a large share of the content a readiness journey needs already sits in the catalog;
the work is to validate it, map it to the competencies the exam measures, and connect it, rather than to produce more. Harbinger’s eLearning content audit for a global leadership publisher followed that order, validating the content foundation before anything new was built on it. The second is to run readiness as a loop rather than a line. A traditional path moves from course to course to a test and stops. Harbinger’s readiness loop keeps circling.
| Step | What happens | Signal it produces |
|---|---|---|
| 1. Diagnose | Assess against the competencies the exam measures | Where the candidate is weak, and how weak |
| 2. Guide | Route to the specific intervention, not a full course | The precise content the candidate needs next |
| 3. Practice | Apply the concept in exam-representative formats | Whether the candidate can use it, not just recall it |
| 4. Measure | Reassess the same competency | Confirmation that the gap moved, or did not |
| 5. Adapt | Raise difficulty, change modality, or add human support | The right next step for this candidate |
Each cycle produces something a completion record cannot: proof that a candidate is measurably closer to ready, expressed as competency-level progress rather than a percentage on a dashboard. That evidence is also a management asset, showing leadership where the product produces outcomes and where candidates repeatedly break down, which then guides content and assessment investment. The scope can stay small. In practice, five questions separate a completion program from a readiness one, and they are worth putting to any executive team: are we measuring activity or readiness; do we know where candidates struggle before the final exam; are candidates practicing the competencies that matter most, or only the ones they prefer; are our existing assets connected around a competency model or scattered as a catalog; and what is our value proposition once candidates can generate unlimited practice on their own. That last question is now the sharpest.
Where AI Fits in Certification Prep, and Where It Backfires
Generative AI has changed the economics of preparation. Candidates can ask a general-purpose model to generate practice questions, explain concepts, and build a study plan in seconds, which forces a direct question for any provider: what makes the official preparation worth paying for? Content volume alone is no longer enough to answer it, because plausible content is now cheap to produce. High-quality, validated content still differentiates, but the defensible value is the guided readiness experience a candidate cannot assemble alone.
The evidence explains why unguided AI prep underdelivers. In a randomized study of nearly a thousand students at The Wharton School, those given unrestricted AI help during practice improved practice performance by 48 percent, then scored lower on an unassisted exam than peers who studied without it. When the same AI was rebuilt to guide with hints instead of supplying answers, the loss largely disappeared. The design, not the tool, decided the outcome.
For providers, that is both a product principle and a moat. A readiness system should move a candidate toward the answer and the reasoning behind it, because the exam and the job will supply neither, and that guided experience is what free tools cannot replicate. Building it well, with clear oversight of how AI generates and recommends content, protects both candidate outcomes and the provider’s reason to exist. Harbinger’s work operationalizing AI governance for skills intelligence and career recommendations applies that discipline: useful signals, responsibly governed.
Each of these threads leads back to the same number every certification leader is ultimately accountable for.
The Metric Worth Building Toward
Completion confirms that a candidate finished the course. Readiness confirms that a candidate can perform, and it is readiness that sustains the trust employers place in a credential and the premium the market pays for it. The providers pulling ahead are instrumenting their programs to generate evidence of the second, not merely a record of the first, and that evidence is what protects the credential’s value as scrutiny rises. Whatever the title on the org chart, the priorities converge on one question: can the organization show that its preparation experience moves candidates toward readiness?
None of this requires discarding a program that works. It usually starts with one certification track: define what readiness means, map its content and assessments to the competencies the exam measures, find where candidates drift, and use earlier signals to trigger the interventions that close the gap before a final score does. To see where a single track stands today, Harbinger helps certification and professional-education teams assess a representative sample of content and candidate data and pinpoint the readiness gaps worth closing first. Contact us to start with the track where readiness matters most.
