By Bradford Donohue, President & CEO, IHRDC
Walk into any HR conference this year and one word dominates the conversation: skills. Workday Skills Cloud, Lightcast, Eightfold and a wave of advisory firms are urging companies to rebuild their talent systems around skills rather than jobs. Lighthouse Research & Advisory reports that 84 percent of employers are now embracing or exploring a skills-first strategy.
The momentum is real.
I think the skills movement gets one important thing right. I also think it is missing something fundamental for the industries we work in, where the cost of getting it wrong is measured in lives, barrels, and billions. After more than thirty years of building competency management systems for the energy sector, I want to make the case for what gets lost when a skill label is treated as enough, and why competency management is becoming more important, not less, as the skills wave builds.
The Skills Movement Got Something Right
The instinct behind the movement is sound. Static job descriptions cannot keep up with how work actually changes. Job titles are a poor lens for internal mobility. AI systems infer skills from resumes, postings, and learning records, and that data has genuine value.
For recruiting, internal mobility, and broad talent planning, a common skills vocabulary helps. Lightcast’s open taxonomy alone now defines roughly 33,000 skill labels, and that is a useful translation layer between candidates, employers, and AI tools.
Concede that, and the next question becomes the important one.
What a Skill Label Cannot Tell You
Take a capability that matters in our industry: well log interpretation. As a skill record in a modern HR platform, it is a single label with a short definition, a category, and a few aliases. As a competency, it is something quite different. There are observable behaviors at four proficiency levels. Awareness of the main log types and what each measures. Basic application under supervision. Skilled, independent execution with environmental and borehole corrections, and reconciliation against core data. Mastery, including integrating log data with seismic, designing logging programs for new wells, and adjudicating across the asset team.
That difference is not academic. A skill record tells you a name. A competency tells you what a person
can be observed doing, at what level, and with what evidence.
The same Lighthouse research that found widespread skills-first adoption also found that 63 percent of employers do not believe their workforce is genuinely adaptable, and only 16 percent of employees feel their skills are being developed for future success. The vocabulary is racing ahead of the substance underneath.
Six Things Competency-Level Detail Gives You
These are the things a skill label, on its own, does not provide.
Behavioral specificity. “Knows well log interpretation” is not actionable. “Independently corrects for borehole and environmental effects, and reconciles log data with core” tells a supervisor exactly who can be put on a well.
A real proficiency ladder. Four levels from Awareness to Mastery give an employee a career path, a manager a development plan, and HR a credible succession bench. A skill is essentially binary, either present or not.
Evidence and assurance. Assessor guides, evidence requirements, and internal verifier sign-off are what regulators, auditors, joint-venture partners, and national oil companies actually ask for. A self-claimed or AI-inferred skill tag is not a substitute.
Training tied to the gap. The behavioral statement is the learning objective. From a competency gap, an organization can build curriculum directly. From a skill label, it cannot.
Decisions that can be defended. Promotions, critical-project assignments, and mentor selection are stronger with assessed evidence than with an inferred tag. The day something goes wrong, the question is not whether the system said someone had the skill. It is what evidence proved they were competent.
A career ladder employees can see. “Here is exactly what you need to demonstrate to move from Skilled Application to Mastery” beats “develop your skill.” It changes a vague aspiration into a concrete plan, with a visible next step.
The Decision-Defense Test
A useful test for any talent decision: would you bet the well, the platform, or the audit on this?
In low-stakes contexts, a skills graph is enough. A marketing copywriter, an inside sales rep, a junior analyst. The downside of a mismatch is recoverable. In high-stakes contexts, the test fails for skills alone. Drilling supervisors. Refinery operators. Process safety engineers. Subsea systems leads.
These are roles where the question is not whether someone has a skill, but whether they can be relied upon to demonstrate it under pressure. Competency-level detail is the only foundation that holds up to that question, and to the audit that inevitably follows when something does go wrong.
Mercer’s most recent data finds that only 14 percent of companies have actually instituted a skills-based talent strategy. The talk is ahead of the implementation, because real implementation runs into exactly this problem.
Build the Skill Layer on Top, Not Instead Of
The right answer is not to choose between the two. It is to layer them.
Treat skills as the vocabulary. They are how the rest of the world, including recruiters, AI systems, employees themselves, and the platforms a client is already running, talks about capability. You need that vocabulary in the conversation.
Treat competencies as the substance. They are how an organization actually verifies that a person can perform the work, defends the decision, plans the training, and grows the career.
A single skill maps to one or more competency units. A job title declares which skills it requires, and at what proficiency level. The data model is additive. Nothing in a strong competency library has to be thrown away. The skill layer sits on top as a translation surface to the broader market, while the competency library underneath does the real work.
At IHRDC, this is the direction we are taking CMS Online. We are keeping the competency core that energy clients have built their assurance frameworks on, and adding a skills layer above it that speaks the market’s language. We think this is how skills-based talent management actually works in industries where the cost of being wrong is real.
Where to Start
For HR and L&D leaders weighing how to respond to the skills wave, three suggestions.
First, do not throw out a competency library you already have. It is the most valuable asset you bring to this conversation. If you do not have one in the areas that matter most, build the foundations before adding a skills layer above them.
Second, ask the skills vendors a single question: how does your system handle proficiency, evidence, and assessment? If the answer is “we do not, but you can attach documents,” you have found the gap that will need to be filled by something else.
Third, decide where on your business the answer matters. The skills-first model is fine for roles where mistakes are cheap. For the roles where they are not, competency management is not optional. It is the foundation.
FAQs
What is the difference between a skill and a competency?
Skills-first platforms are useful for recruiting and internal mobility, but in high-stakes roles such as drilling supervisors, refinery operators, and subsea systems leads, a skill label cannot show whether someone can be relied upon to perform under pressure. Competency management adds the proficiency levels, evidence, and assessor sign-off that these roles require.
Yes. The recommended model treats skills as the vocabulary that connects to recruiting platforms and AI tools, while competencies remain the substance underneath that verifies performance. A single skill can map to one or more competency units, so a job title can declare which skills it needs while the competency library does the underlying assurance work.
A common structure uses four levels: Awareness of core concepts, Basic application under supervision, Skilled independent execution, and Mastery, which includes integrating the capability across broader systems and adjudicating decisions. This ladder gives employees a career path and gives HR a defensible succession bench that a binary skill tag cannot provide.
A skills graph is generally sufficient in lower-stakes roles, such as marketing or inside sales, where a capability mismatch is recoverable. For technical and safety-critical roles in energy and other hazardous industries, competency-level detail with assessed evidence is the standard that regulators and auditors expect.