What is a competency model?
A competency model describes what good performance means in a role and in your organisation. It can connect knowledge, skills, abilities and other personal characteristics with specific expectations for behaviour. Behavioural anchors illustrate what a competency looks like in everyday work. The model provides a shared framework for clarifying expectations, giving feedback and planning development.
In the literature and in interviews with 37 subject matter experts, a task force of the US-based Society for Industrial and Organizational Psychology (SIOP) found no uniform definition of the term competency. The definitions ranged from characteristics of a person, such as knowledge, skills and abilities, to observable behaviour (Schippmann et al., 2000). We distinguish between these personal characteristics and observable behaviour. In the model, we connect the two: we describe a competency through behaviour; behind it lie knowledge, skills, abilities and other personal characteristics, which we understand as prerequisites for this behaviour. Whether it actually shows also depends on the situation and the working environment. The behaviour is not the whole competency, but it is the part that can be observed and discussed in everyday work. The figure shows an example of the path from the role to the behavioural anchor.
Tree diagram from top to bottom: three role families; for the role family Customer service, four role levels (role family and role level together make a role); for the role level "Team member, working independently", ten competencies; for the competency "Coordinating with adjacent areas", its definition and three behaviours; for the first behaviour, a scale from 1 to 5 with levels 1, 3 and 5 described.
Level 1, role families
- Customer service
+ 2 more role families
- Production
- Inside sales
Level 2, role levels
- Team member, working independently
+ 3 more role levels
- Team member, working under guidance
- Leading people
- Senior management
Role family and role level together make a role. You are looking at the role level "Team member, working independently". On other role levels, the same competency may be described differently, and not every competency has to appear on every role level.
Level 3, competencies at this role level
- Coordinating with adjacent areas
+ 9 more competencies
- Clarifying customers' concerns
- Explaining solutions clearly
- Documenting cases in full
- Handling complaints in a matter-of-fact way
- Setting priorities in day-to-day work
- Coping with peaks in workload
- Sharing knowledge within the team
- Using systems and tools proficiently
- Initiating improvements
Level 4, definition
- Coordinates decisions and information with the areas affected by them.
Level 5, observable behaviour
- Involves affected areas before a decision that concerns them is made.
- Passes on information that others need for their work on their own initiative and in good time.
- Raises conflicting goals between areas openly and looks for a joint solution.
Level 6, behaviourally anchored rating scale
- 5Asks affected areas for their view while the decision is still open, and takes their input into account in the decision.
- 3Informs affected areas only once the decision is essentially settled, and considers their input only on points of detail.
- 1Decides without involving or informing affected areas beforehand.
Levels 1, 3 and 5 are described; levels 2 and 4 lie in between.
The content of this figure is entirely fictitious. It is for illustration only and not intended to be adopted. For a model in your organisation, roles, levels, competencies and descriptions of behaviour are developed from your material and from conversations with those involved, or checked against them.
Details as a table
| Level 1, role families |
|
|---|---|
| Level 2, role levels |
Role family and role level together make a role. You are looking at the role level "Team member, working independently". On other role levels, the same competency may be described differently, and not every competency has to appear on every role level. |
| Level 3, competencies at this role level |
|
| Level 4, definition |
|
| Level 5, observable behaviour |
|
| Level 6, behaviourally anchored rating scale |
Levels 1, 3 and 5 are described; levels 2 and 4 lie in between. |
What is a competency model for?
Organisations use competency models for different purposes, for example:
- Development: aligning programmes, coaching and learning paths with the expectations described
- Feedback: basing conversations about behaviour on shared standards
- Orientation: describing roles and career paths, including across areas
- Personnel decisions: selection, promotion and succession
Anyone who wants to use a model for personnel selection, promotion or succession needs additional evidence that it is suitable for exactly this purpose.
A competency model states explicitly which behaviour counts in a role. Research on role ambiguity, that is, uncertainty about what is expected in one's own role, offers an indication of why clear expectations can matter. In two meta-analyses, role ambiguity was associated with lower job performance (Tubre & Collins, 2000; Gilboa et al., 2008). The studies are correlational and do not establish causation. They did not examine whether a competency model reduces role ambiguity.
How is a competency model developed? Six steps
1. Clarify the purpose
What do you want to use the model for? Should it support development, feedback and programmes? Or should it also feed into decisions on selection and promotion? The purpose determines how thoroughly you should check the model and whom you should involve. Also clarify who backs the project at management level.
2. Gather material
The starting point is your organisation's goals, tasks and specific work situations. Examples include strategy and objectives documents, job and role descriptions, person specifications and mission statements. These are complemented by conversations or workshops with job holders, leaders and other stakeholders. We recommend working with specific work situations in these conversations and workshops: they allow participants to describe which behaviour, in their experience, helped to accomplish a task and which made it harder. Ask not only what counts today, but also what your strategy and foreseeable changes will require of the roles in future.
3. Start from the tasks
The competencies required for a role are derived from its tasks: what needs to be done, and what behaviour is needed? In studies by Lievens, Sanchez and De Corte (2004), judgements of which competencies a role requires were somewhat more consistent and distinguished better between roles when the raters had first received information about the role's tasks. Raters with little experience performed poorly on both counts.
4. Translate into observable behaviour
Each competency has a short description and specific behaviours that make it recognisable in everyday work. Behavioural anchors illustrate how different levels of a competency are expressed. General terms alone are not enough for this. They can also distort judgements of what a job demands: in a field experiment, employees mostly rated generally worded competencies as more important for their job and more frequently needed than the same content broken down into concrete tasks and abilities. The researchers see this as overestimation: worded in general terms, a job's requirements thus seem bigger (Morgeson et al., 2004).
5. Have it reviewed by the people who know the role
Have the draft reviewed by job holders, leaders and other subject matter experts, so that they can add what is missing and remove what does not fit. How the raters are prepared can also play a part: in a quasi-experimental study, judgements of which competencies a role requires were more consistent and more accurate after training on a shared frame of reference (frame-of-reference training) than without this training (Lievens & Sanchez, 2007). The study was carried out with HR consultants at a consulting firm; whether the same applies to job holders and leaders was not examined there. We carry this finding over to our work: we prepare participants for the review and build on what they already bring.
6. Introduce the model and keep it up to date
The model needs a place in everyday work: in conversations, in feedback and in development programmes. How the model is set up may matter here: in a survey and a scenario experiment from the Netherlands, employees were more likely to say that they used or intended to use competency management when it was geared towards development and participation rather than control (Heinsman et al., 2008).
We recommend trialling the model in one area first: this lets you check whether the descriptions and behavioural anchors are understandable and usable in everyday work. Prepare leaders to assess behaviour using the model, for example for feedback and development conversations. Research on rater training offers an indication: in controlled studies, mostly with students, training on a shared frame of reference made ratings noticeably more accurate immediately afterwards (Roch et al., 2012). These studies do not show to what extent the results carry over to working with competency models in everyday practice. We use them as a reference point for preparing leaders.
If your strategy or roles change, you should review the model as well.
More from the research
Three further studies concern HR practice in general, not competency models specifically; we carry their findings over to competency models. A meta-analysis of 27 studies found that the more employees were involved in their performance appraisal, the more satisfying, fair and useful they found the appraisal (Cawley et al., 1998). In a field study, the HR practice actually implemented differed from the HR practice intended; where implementation came closer to the intention, employees were more satisfied with HR practice (Khilji & Wang, 2006). In another field study, high-performance work practices were associated with employee outcomes, such as performance and turnover intention, only indirectly, through how line managers perceived their implementation (Sikora et al., 2015). These studies are correlational and do not establish causation.
Ready-made model or your own model?
You can start with an existing competency model or a competency library, or develop a model for your organisation. With a ready-made model, the structure, terms and descriptions are already in place. This can make it easier to get started. Whichever route you take, you should check whether the competencies fit your tasks and whether the descriptions make clear which behaviour counts in your roles. The language should also be easy to understand for the people in your organisation.
A ready-made model can serve as a starting point: you check it against your tasks and your material and adapt it where it does not fit (steps 2 and 3). Which route makes sense for you depends, for example, on the purpose, the time and budget available, and how well existing models fit your roles.
Which mistakes can be avoided?
An exploratory study from German-speaking Europe looked at how competency models fare in practice: in telephone interviews with HR managers from 20 organisations in Germany, Austria and Switzerland, only one organisation was rated as high quality for its model and approach; for most of them, the researchers saw room for improvement (Liebenow et al., 2020). When developing and using a model, we recommend avoiding the following mistakes:
General terms with no behaviour described. "Teamwork" or "customer focus" alone does not say what someone should actually do.
Consequence: People may interpret these terms differently. This makes it difficult to base feedback and assessments on specific behaviour, and the requirements can appear greater.
How to avoid it: Describe each competency with observable behaviours and behavioural anchors (step 4).
Adopting a ready-made model without checking it.
Consequence: Terms and examples may then not fit your tasks and your language, and the model risks remaining a document that sees little use in everyday work.
How to avoid it: Use existing models as a guide and compare them with your tasks and your material (steps 2 and 3).
Competencies with no link to the tasks.
Consequence: In studies, judgements of which competencies a role requires were less consistent without task information and distinguished less well between roles.
How to avoid it: Start from the tasks of the role and give this information to everyone involved in advance (step 3).
Leaving out the people who know the role.
Consequence: A model developed only at a desk can easily miss what counts in everyday work, and may find little support among those who are meant to work with it.
How to avoid it: Have the draft reviewed by job holders and leaders, and prepare them for this review (step 5).
Control instead of development.
Consequence: In one study, employees were less willing to use competency management that was geared mainly towards control.
How to avoid it: Design the model around development and participation, and use it in conversations, feedback and development programmes (step 6).
Created once, never maintained.
Consequence: Roles and strategy change. If the model is not adapted, it loses its connection to everyday work.
How to avoid it: Review the model when your strategy or roles change (step 6).
Too many competencies.
Consequence: An extensive model can be harder to apply. In feedback and development conversations, it may remain unclear which competencies should be the focus.
How to avoid it: For each role, check which competencies are needed for the intended purpose, where content overlaps and what can be combined or left out (steps 1 and 3 to 5).
Do you already have a competency model and want to know where it stands? We review and revise existing models. In a free introductory call, we clarify whether and how we can support you.
Book a callHow we work
The starting point is your organisation's goals, tasks and specific work situations. Depending on the assignment, we develop an organisation-specific model or review and revise an existing one.
Frequently asked questions
What is the difference between a competency model and a job analysis?
A job analysis mainly describes the tasks of a job and what they require. A competency model sets out which behaviour makes for good performance across tasks, where applicable for several roles at once. The two complement each other; a paper with practical recommendations on competency modelling describes how they differ and how they complement each other (Campion et al., 2011).
Can we adopt a ready-made competency model?
Existing models and scientific approaches provide guidance. Whether a model fits your organisation becomes clear when you compare it with your tasks and your material; where it does not fit, it needs adapting, or you need a model of your own.
Whom should we involve on our side in developing the model?
A responsible person on the client side, HR or learning and development, relevant leaders and employees from the areas concerned. We involve further stakeholders, such as the works council, early on, depending on the topic and the intended use.
What research shows
- Campion, Fink, Ruggeberg, Carr, Phillips & Odman (2011). Doing competencies well: Best practices in competency modeling. Personnel Psychology, 64(1), 225–262. https://doi.org/10.1111/j.1744-6570.2010.01207.x Practice paper; co-authors from companies and a consultancy.
- Cawley, Keeping & Levy (1998). Participation in the performance appraisal process and employee reactions: A meta-analytic review of field investigations. Journal of Applied Psychology. https://doi.org/10.1037/0021-9010.83.4.615
- Gilboa, Shirom, Fried & Cooper (2008). A meta-analysis of work demand stressors and job performance: Examining main and moderating effects. Personnel Psychology. https://doi.org/10.1111/j.1744-6570.2008.00113.x
- Heinsman, de Hoogh, Koopman & van Muijen (2008). Commitment, control, and the use of competency management. Personnel Review. https://doi.org/10.1108/00483480810906865
- Khilji & Wang (2006). ‘Intended’ and ‘implemented’ HRM: The missing linchpin in strategic human resource management research. The International Journal of Human Resource Management. https://doi.org/10.1080/09585190600756384
- Liebenow, Uedelhoven, Stricker & Nachtwei (2020). Qualität im Kompetenzmanagement: Eine explorative Studie zur Entwicklung eines Benchmarks für Kompetenzmodellierung und Kompetenzmodell (BeKom) [Quality in competency management: An exploratory study on developing a benchmark for competency modelling and competency models]. Zeitschrift für Arbeits- und Organisationspsychologie, 64(2), 120–148. https://doi.org/10.1026/0932-4089/a000324 In German; English title translated by us.
- Lievens & Sanchez (2007). Can training improve the quality of inferences made by raters in competency modeling? A quasi-experiment. Journal of Applied Psychology. https://doi.org/10.1037/0021-9010.92.3.812
- Lievens, Sanchez & De Corte (2004). Easing the inferential leap in competency modelling: The effects of task-related information and subject matter expertise. Personnel Psychology. https://doi.org/10.1111/j.1744-6570.2004.00009.x
- Morgeson, Delaney-Klinger, Mayfield, Ferrara & Campion (2004). Self-presentation processes in job analysis: A field experiment investigating inflation in abilities, tasks, and competencies. Journal of Applied Psychology. https://doi.org/10.1037/0021-9010.89.4.674
- Roch, Woehr, Mishra & Kieszczynska (2012). Rater training revisited: An updated meta-analytic review of frame-of-reference training. Journal of Occupational and Organizational Psychology. https://doi.org/10.1111/j.2044-8325.2011.02045.x
- Schippmann, Ash, Battista, Carr, Eyde, Hesketh, Kehoe, Pearlman, Prien & Sanchez (2000). The practice of competency modeling. Personnel Psychology, 53(3), 703–740. https://doi.org/10.1111/j.1744-6570.2000.tb00220.x Task force report; literature search and interviews with subject matter experts.
- Sikora, Ferris & Van Iddekinge (2015). Line manager implementation perceptions as a mediator of relations between high-performance work practices and employee outcomes. Journal of Applied Psychology. https://doi.org/10.1037/apl0000024
- Tubre & Collins (2000). Jackson and Schuler (1985) revisited: A meta-analysis of the relationships between role ambiguity, role conflict, and job performance. Journal of Management. https://doi.org/10.1177/014920630002600104
About the author
Nathan W. P. Schäfer founded UNLEASHING POTENTIAL in 2025. He is an organisational psychologist (M.Sc., Birkbeck, University of London) and a certified systemic coach (ECA, QSA). Previously he advised on leadership development, assessment and succession at an international consultancy.
This article was prepared with the support of AI tools. Nathan Schäfer is responsible for its content, source checking and approval.
Published: Last reviewed:
