We all have skills we can share with others, and using others’ skills can help us deliver outcomes as well as grow relationships between our peers. Mapping people to their skills is complex however; this article seeks to outline the high-level options, and highlight issues to consider when designing an application to map skills in an organisation.
Defining skills
One of the first issues we need to agree is how we define the skills – whether they are defined from the bottom-up using a folksonomy or defined top-down using a controlled vocabulary.
Folksonomies give users control – greatly helping with adoption, and avoid the the many issues of controlled vocabularies outlined in the issues to be mindful of section below.
If we follow a folksonomy route, we should ensure the application allows users to pick from an existing list of skills sorted by the number of people currently mapped to them; alert people to spelling errors when creating new terms; and validate the term format. Having #Sharepoint, sharepoint, SharePoint, Microsoft-SharePoint, and Sharepint all appearing as separate skills won’t be helpful. We may also want to ‘seed’ the folksonomy with terms to prevent the term set becoming too free-form from the outset.
A well-governed controlled vocabulary on the other hand can provide consistency across discipline term sets, eliminate duplication, provide a hierarchy or even a poly-hierarchy (where the ‘HTML’ skill might sit under the ‘Marketing’ and ‘Software Development’ disciplines), allow for synonyms, localisation using a translation application such as Weblate, and integration with other applications that might use the same vocabulary.
If we follow a controlled vocabulary route, we might think about investing in a taxonomy solution, releasing resources to ensure that the taxonomy is regularly reviewed and updated, involve subject matter experts in its creation and maintenance, provide clear guidelines for users on how to navigate and apply the vocabulary, and implement a feedback mechanism for users to suggest new terms or update existing ones.
Defining people’s proficiency
Here we have a choice between users or their peers defining their skills; and whether we want to define levels of proficiency – say ‘Awareness / Novice / Proficient / Expert’ – against each skill or simply a Boolean ‘you have it or you don’t’ approach.
These options can be combined; so for instance people can define their skill and others can then endorse it, a familiar approach for people using LinkedIn where it’s an either you have the skill or don’t – the many issues outlined below that may arise with a more granular approach are obviated.
Speaking of LinkedIn, you may want to leverage their API to pull in the existing information from people’s profiles; not only can this relieve the need to duplicate the information and spin up a project to manage skills mapping within your organisation, but can also help to secure more work where your employees have many endorsements from their peers.
Issues to be Mindful of
Cory Doctorow provides a sharp and coherent argument against controlled vocabularies and user-defined proficiencies: Metacrap: Putting the torch to seven straw-men of the meta-utopia.
There are enormous benefits to having a taxonomy to bring skills together, and though his argument is presented in the context of the internet rather than a skills map, the arguments are valuable considerations as we seek to map organisational skills. I’ve paraphrased it to put them into the context of skills mapping:
People lie : Colin Cottonwool, prone to the odd blunder and short of work, will actively categorise and rate himself as highly as possible to appear hire-worthy. Tania Talent, like all good plumbers, will be hard to find as her overwhelming workload provides a disincentive to rating herself highly and accurately. This bias will actively work against the objectives of a skills map.
People are lazy : We might have some well-thought out vocabularies, while other discipline leaders won’t see the value in their creation which will lead to messy, if any, categorisation. To say nothing of Bobby Busy’s apathy for filling his personal one out.
Controlled vocabularies also become outdated – requiring investment to re-align them to business needs – and fail to quickly pick up on new insights or developments (it may take some time to reach agreement on where slime-mould modelling expertise should fit for instance).
People are stupid : There is the potential to add skills without checking spelling, punctuation, and so on. It’s less likely in an enterprise environment, but misspellings will be found despite our best intentions. And search will fail where this occurs.
Mission: Impossible – know thyself : Expecting users to accurately rate their own skill, and levels of it, is impossible. I’ve tried and my self-perception was significantly different to what other team members consistently perceived (my illusions of grandeur!). ‘People are lousy observers of their own behaviours.’
Schemas aren’t neutral : The Fire Leader may argue for a ‘Sprinklers/Plumbing’ hierarchy while a plumber would say ‘Plumbing/Sprinklers’. ‘The conceit that competing interests can come to easy accord on a common vocabulary totally ignores the power of organizing principles in a marketplace.’
Metrics influence results : Vocabulary owners may try to ‘up’ their words/tags over other skills areas. So the cross-network Building Physics leader might argue a more fine-grained approach to software skills is required in the Mechanical skills matrix while the Mechanical leader may think software skills are not part of the ‘Mechanical’ set at all. ‘It’s wishful thinking to believe that a group of people competing to advance their agendas will be universally pleased with any hierarchy of knowledge.’
There’s more than one way to describe something : Do we use ‘public health’, ‘water’ or ’hydraulic and fire systems’ to describe plumbers? ‘Requiring everyone to use the same vocabulary to describe their material denudes the cognitive landscape, enforces homogeneity in ideas.’
On the other hand employees tend not to be stupid, lazy, liars who won’t see value in this exercise; a well-made system can cajole most people into doing it for the interests of the firm, and there is likely more co-operation and goodwill to work on skills mapping in an organisation than on the internet in general.
Every organisation will be different – if you’re after a recommendation on what will be best for yours please get in touch. Either way, I trust the above will give you some excellent pointers on the way forward. Do let us know in the comments how you get on…