Back to Insights
Talent Management

How to Build a Competency Framework from Your Job Descriptions

November 22, 20266 min read

Competency frameworks get built in one of two ways. The first involves offsite workshops, consulting models, and months of debate about whether strategic thinking belongs in the leadership layer or the management layer. The second starts with the job descriptions you already have and extracts the competencies that are actually predictive of success in each role.

The second path is faster and more grounded. Your job descriptions, performance reviews, and interview scorecards already contain signals about what matters. If you hire for problem-solving in one role and stakeholder management in another, those are competencies. The work is to structure them, make them consistent, and tie them to progression.

Start with an audit of what you already have

Pull your job descriptions, interview guides, and performance review templates for a representative set of roles: entry-level, mid-level, senior, and leadership. Read through them looking for skills, behaviors, and attributes that show up repeatedly. You are not looking for job duties or technical requirements. You are looking for the underlying capabilities that predict success.

Common patterns include problem-solving, communication, collaboration, decision-making under ambiguity, and adaptability. If multiple job descriptions mention the ability to work across teams or manage competing priorities, those are competencies. If they describe the need for analytical thinking or the ability to influence without authority, same thing.

List every capability that appears in more than one role. Do not filter yet. The goal at this stage is to surface the implicit competency model your organization is already using, even if it is not formalized or consistently applied.

Cluster and define the core set

Take the long list of capabilities and group them into higher-order competencies. 'Communicates clearly in writing,' 'presents to senior leadership,' and 'explains technical concepts to non-technical stakeholders' all ladder up to communication. 'Identifies root causes,' 'evaluates trade-offs,' and 'makes decisions with incomplete information' are all flavors of problem-solving.

Aim for 8 to 12 core competencies that span the organization. Fewer than that and you lose specificity. More than that and the framework becomes too detailed to operationalize. Each competency should have a clear definition, written in plain language, and a set of observable behaviors that indicate proficiency.

For example, collaboration might be defined as 'works effectively across teams and functions to achieve shared goals,' with observable behaviors like 'seeks input from stakeholders before finalizing decisions,' 'resolves conflicts constructively,' and 'shares credit and accountability.'

Map competencies to levels and roles

Not every competency applies equally to every role. An entry-level analyst needs problem-solving, but the complexity and scope will differ from what a director-level leader needs. Build a matrix that shows which competencies matter for each role family and level, and what proficiency looks like at each stage.

A simple three-point scale works for most organizations: developing, proficient, and advanced. Developing means the person demonstrates the competency with guidance and support. Proficient means they can operate independently. Advanced means they model the competency for others and coach teammates.

Use this matrix to rewrite job descriptions, structure interview scorecards, and set expectations for promotions. If you say problem-solving at the manager level requires the ability to identify solutions independently, your interview process should test for that, and your performance reviews should measure it.

Automate extraction where possible

If you have hundreds of job descriptions across multiple geographies and business units, manual extraction is not scalable. AI tools can parse job postings, interview notes, and performance feedback to surface competencies, cluster them, and suggest a draft framework in hours instead of months.

Nuvis's product Najaba does exactly this: it ingests your existing HR content, extracts the competencies your organization already values, and builds a framework you can refine and deploy. It is not a replacement for judgment, but it removes the most tedious part of the work. Learn more at https://najaba.ai/competency-assessment.

Roll out and refine iteratively

A competency framework is not built once and locked in a document library. It needs to evolve as your business changes. Launch the framework in one function or region first, gather feedback from managers and employees, and adjust the definitions and proficiency levels based on what you learn.

Track adoption. Are hiring managers using the framework in interviews? Are employees referencing it in development conversations? If not, the framework is too abstract or too divorced from how work actually gets done. The best frameworks feel like a formalization of what people already knew mattered, not an external model imposed from above.

What comes next

A well-designed competency framework gives you a shared language for hiring, development, and progression. It makes performance conversations clearer, succession planning more objective, and skills gap analysis possible. But it only works if it reflects the real capabilities your organization needs to succeed, not the generic model from a consulting deck.

If you are building or refining a competency framework and need a structured approach that starts with what your organization actually requires, we can help. Book a consultation at /contact.

Let's talk

Ready to build a workforce that's ready for what's next?

Tell us where the friction is. We'll show you what an AI-powered, evidence-led HR function could look like inside your organization.