Designing competency frameworks

I’m sceptical of the core competency frameworks in general.  They often seem to me to be over-engineered lists of a mix of skills, behaviours, and other attributes.  Frequently there is little obvious link to the delivery of the organisation’s strategy or values.  Note I am less sceptical about specific technical competencies to indicate skill in a field (e.g. architecture, nursing, or UX design)

Take the OECD’s (2014) set which divides competencies into technical competencies: specific to a discipline or field of practice and core competences.  Technical competencies are the ‘requirements to successfully perform a given job’ and in their case ‘are defined in job vacancy announcements.’

Their core competencies, on the other hand – those everyone should have – are described in a booklet.  OECD lists and describes fifteen core competencies  grouped into three clusters: delivery-related, interpersonal and strategic, and 5 levels (related to type of role).  Level 1 is roles including ‘assistant’ and ‘operator’, level 5 includes Heads of Function and Directors, giving a total of 75 statements.  This form of competency framework is common.   I’ll take the OECD one as an example of why I am sceptical:

The OECD competency ‘Analytical Thinking’ at level 1 lists:

  • Distinguishes between critical and irrelevant pieces of information.
  • Gathers information from a variety of sources to reach a conclusion.

And at Level 5 lists:

  • Is sought out by others for advice and solutions on how to best interpret and use information.
  • Discerns the level of pressure or influence to apply in each aspect of the analysis in relation to the broader context.

My scepticism on this sort of thing is based on my view that the items on such lists are:

  • Subjective e.g. a Director – Level 5 –  may not be able to distinguish between critical and irrelevant information(a Level 1 competency) and who is judging what is critical or irrelevant?
  • Not relatable to role or level e.g. an assistant, Level 1,  may be sought out by others for ‘advice and solutions on how best to interpret and use information’. (A Level 5 competence)
  • Not indicators of job performance as the context will influence the ability to deploy (or not) the competence.
  • Not conducive to being ‘levelled’ by role. Any role may require different levels of competence so an assistant my require some of the competence listed at Director level.  For example, what assistant does not have to handle ‘difficult on-the-spot questions (e.g. from senior executives) listed in this framework as a level 5 competence?

But these frameworks have lots of defenders.  Take a look, for example, at the SHL Universal Competency Framework or the UK’s CIPD Competency FrameworkFactsheet.

(I notice that the SHL (2011) info says firmly that we need to distinguish between the words ‘competence’ and ‘competencies’, because ‘it is unfortunate that two very similar words have been used to describe two very different constructs. It is essential that there is a clear distinction between these two terms.’   The CIPD (2020) explains that ‘In the past, HR professionals have tended to draw a clear distinction between ‘competences’ and ‘competencies’. … More recently however, there’s been growing awareness that job performance requires a mix of behaviour, attitude and skill, and the terms are now more often used interchangeably.’)  In this sort of distinction you start to see the difference between core and technical competences.  In some cases frameworks mesh these.  See, for example, the Actuarial Competency Framework.

One person who does not defend core competency frameworks is Marcus Buckingham, who says:

  • ‘Competencies can’t be measured. So, your scores (or the scores you give your team) and all the data around how much of a certain competency a person possesses are completely made up.
  • No single person possesses all competencies. When you study people who excel at a certain job, although as a group they may have all of the competencies that are supposedly required, no one person has all of them.
  • There is no data that shows that people who acquire the competencies they supposedly lack outperform the people who don’t. So even if we could accurately determine that you are lacking a specific competency, having you take a learning and development course to plug that gap will have no effect on your performance. Well-roundedness does not predict higher performance, and it’s better to be sharp in one or two key areas instead of well-rounded.’

The topic of competency frameworks came up this week as an organisation asked me for advice on them.  They had questions related to links between the framework and delivery of strategy and values, whether they needed core as well as technical competences, how to communicate the competences to the workforce in a simple and easy to use way.

What I’ve found is that organisational values are a very good basis against which to judge employee behaviour, attitude and contribution – assuming that you have chosen values that support delivery of your business strategy.  And last week I listened to Yancy Strickler saying much the same thing.  He is the founder of Kickstarter, and he was talking about ‘the values the company created, which helps guide the way Kickstarter attracts and hires talent and constructs and operates its business’.

Marcus Buckingham is also of the view that core ‘competencies are simply values. They should be written on a wall, not attempted to be measured and learned. If you want your team to be goal-oriented and customer service-focused; express them as values, create stories around them, celebrate the heroes who demonstrate them – bring these values to life.’

The organisation who I was discussing competency frameworks with have five values on which to judge an employee’s contribution.  Many organisations are now ‘values based’ – Ben and Jerry’s is a classic example as is Patagonia

I suggested that those in organisation I was talking with re-think their core competences, instead focusing on the values – not as a measurement tool in the traditional sense but to gauge whether people are going to be, in Patagonia’s terms, not a culture fit, but a ‘culture add’.  Patagonia’s values-based approach ‘to evaluating potential hires [is one] that arises from the company’s unwavering and ironclad commitment to its mission. And it’s a reminder to every organization that they are hiring human beings, not skill sets or even experience.’

For other aspects of workforce management – career development, technical progression, management/leadership development – I suggested they introduce technical competences by job family.   For an excellent example of a technical competency framework for designers look at Jason Mesut’s approach.  (Note that it also includes some core competencies).

To recap – I don’t think most core competency frameworks i.e. items listed in progressive order by level achieve their intended outcome of supporting individual or organisational performance management or enabling, in Mesut’s words. ‘a clear way of objectively promoting or compensating people fairly …  or providing clarity of what a long-term career in the organisation might look like or giving scarce and fickle talent a reason to stay.’   A better approach is to develop technical competency frameworks based on job families and for core competencies do not have a framework by lists and levels. Use only the organisation’s values and give clear and engaging messages that employees are expected to live the values in their daily work.

What’s your view on a traditional core competency framework?  Let me know.

Image:  Global competencies

2 thoughts on “Designing competency frameworks”

  1. Shout-out to you from India. I am really glad to see you take up this topic. Some of us, practitioners and academicians, in India have been thinking about the effectiveness of competency frameworks and your article echos our thoughts.

  2. Someone sent me this email in response to this bog. ‘I totally agree. The issue is one of measurement. The question is, what can be measured. There is a critical distinction between objective measurements and subjective measurements. Subjective measurements are not actually measurements, but opinions. Opinions only matter when you collect a statistically significant number of them. An opinion poll with one respondent would be a very poor predictor of an election result. Basically random.

    We have been told that you can only manage what you can measure, and consequently armies of enthusiastic consultants and academics have set out to measure what they want to manage. But if something is not objectively measurable, then dreaming up some metric for it is simply delusional.

    The essence of objective measurement is repeatability. Opinions about another person’s competency vary wildly according to who you ask (friend, colleague, spouse, boss, minion). Averaging them out does not make them any more objective. So asking somebody to score somebody else on a form is a waste of time. Actually worse than a waste of time, because if the evaluator actually wasted their time instead, for example by idly scrolling through Tik-Tok posts, the result would cause less damage to the organization than that of somebody receiving a negative evaluation, bearing in mind that we have just established that such an evaluation is invalid. Not to mention the organizational damage caused by all the positive evaluations that affable morons receive.

    There are some things that can be measured objectively about a person’s performance. For example a call center can measure how long an agent spends talking to each caller, and they can collect customer satisfaction scores. Although these scores fall into the category of non-objective opinions, an agent can collect a sufficient number of them (hundreds) to be statistically meaningful. Call center surveys best ask questions like “was your issue resolved” and “was the agent polite.” Cases amenable to statistical measurement involve highly repetitious behavior, and hundreds (or thousands) of measurements of that behavior for each employee. Agents who perform excellently on these metrics are in danger of being reassigned to roles for which they may be totally unsuited, like supervisor.

    Managerial performance is not measurable. The best predictors of positive evaluations are gender, race, height, gravitas, charisma and sucking up. Arguably objective metrics for the first three of these exist. I am confident you could find a consultant to develop metrics for the others.’

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s