When we began working on TeamFit we had a hypothesis that the organization of skill models should emerge bottom up, from the data, rather than be imposed top-down by experts. Some of us had worked building competency models, and we were all too aware of their limitations. Competency models tend to be overly complex, reflect a theoretical view of how work gets done and they generally do not reflect the reality in the field. Most competency models are outdated by the time they are developed. We were searching for a better way.
Several people on the team had worked on social tagging system and we wanted to see if the kinds of self-organizing behaviours of social tagging, a folksonomy rather than a taxonomy, would work better. What is a folksonomy?
“A folksonomy emerges when users tag content or information, such as web pages, photos, videos, podcasts, tweets, scientific papers and others.” (See the article on Wikipedia on folksonomy.) Folksonomies are often represented as tag clouds. The skill names that appear on the left of a TeamFit Skill Map are a kind of tag cloud.
There are many criticisms of this approach.
In our previous work, on solutions like Gleanr and a number of proprietary systems, we found that the shortcomings of folksonomies could be managed through curation and the use of natural language processing and semantic networks. Together these could filter out most of the noise and handle the superficial objections, such as issues with spelling and abbreviations and people using different words to mean the same thing. Semantic technologies also provide powerful ways to organize folksonomies (the TeamFit Skill Graph is a semantic organization of our skill database and the relationships between skills).
Things get a bit more complex when it comes to using ‘the same word to mean different things.’ The reason for this is that people can do the same job in many different ways. Basic skills like ‘Negotiation’ or ‘Sales’ or … pick any complex task … can be done in many different ways. The skills I apply to something like ‘Due Diligence’ will be very different from those that a financial analyst or a technology expert brings to the table. So people using the same word to mean different things is actually a strength of the system, as long as the UX encourages people to look deeper into associated skills and skill families (we have more work to do on this at TeamFit).
Why have categories?
Given our success with folksonomies, why do we categorize skills? Once a person has more than a dozen or so skills, one needs some method to organize in the user interface. Categories are a simple way to do this. Using categories helps us get a quick feeling for a person’s focus. Do they have more Business or Technical skills? If they have a lot of Technical skills, what Tools do they use? Categories can help us better understand ourselves. I have several hundred skills in TeamFit (and even that is but a small fraction of the skills I actually have and use) and looking at them through the lens of different categories helps me reflect on the gaps in my own skills.
The category system is imperfect. There are many skills that fall into more than one category, or that belong in different categories for different people.
For example, in general Pricing would be considered a business skill, but for people who work as pricing experts, it seems more natural to think of it as a technical skill. The boundary between Technical skills and Foundational skills is fuzzy. We treat mathematics as foundational skills, but how about something like Fluid Dynamics. This is a technical skill for people designing systems that involve flow, but the math and concepts are widely applicable and some people use this as a Foundational skill.
No one category system will ever capture the richness of human skills. We already have the ability to support multiple category systems and we are exploring ways to allow users create their own category systems. We are also layering tagging systems onto the skills, so that rather than being caught in one category a skill could have multiple tags. This is probably going to be the best way forward, but I suspect that skill categories will be with us for many years.
Why these categories?
Given the need for a category system, what categories should we use? The six default categories on TeamFit are as follows:
Foundational – the skills we use to develop other skills. This includes all of the basics, reading and writing and arithmetic, basic logic, critical thinking and so on. It also includes meta-skills like learning (and could include learning styles). We also include things like Music as a foundational skill.
Social – the skills we use to build relationships with and work with other people.
Domain – wide domains of human knowledge, such as knowledge of an industry, a culture or a discipline.
Business – the many skills used to conduct business.
Technical – engineering, computer science, some of the more technical forms of design.
Design – creative work of all kinds, from architecture to graphic design, to User Interface (UI) and User Experience (UX) to more abstract forms like Service Design and Instructional Design.
Tools – everything from hammers and drills to programming languages and software packages like Photoshop. On TeamFit, we categorize TeamFit as a tool.
We treat human languages like English or Punjabi as a separate category and do not include them in the tag cloud. We also have a category of Other for the skills we do not know how to categorize and at any time there will be some skills that are Uncategorized (skills where our automated categorization has choked).
We do not use a category of ‘soft skills’ as we have found it to be meaningless. Too many competency models are based on old models of learning and work and fail to represent the relationships between skills. The bucketing of social, foundational, many design and some business skills as soft skills does not help us think clearly about our abilities or potential.
The category that surprises some people is Design. We included this as a category as the skills associated with the things like architecture or graphic design seemed meaningfully different from technical skills and one of the things our users are often looking for is people who combine technical and design skills, or teams that have a good balance of technical, business and design skills. This choice, to call out Design as its own category, reflects the growing importance of design, service design and design thinking at many organizations.
How to use these categories
The categories act as filters so that you can focus in on the skills that are of most interest for the task at hand. When I first look at a person’s skill profile I look at their foundational skills first. This gives me a sense of who they are. As you look at these skills, see if you can guess what role this person plays at TeamFit and how you would work with them.
Then I look at their social skills for clues on how to interact with them.
Finally, I see what category dominates and spend some time looking at the skills they have in their dominant category.
When looking at my own skill profile I tend to flip back and forth between two different categories to see how my skills connect across categories. Here is a picture of my own foundational and social skills.
Support for other category systems
We find our category system pretty useful, but it does not meet all needs. There are established skill models for specific industries that we need to support. We are also developing skill category systems for specific companies. We also plan to develop some additional general purpose category systems so that you will be able to see into your own, your colleagues and your organization’s skills from different perspectives. It is not any one skill that matters, but the connections between skills and the relationships that they help people to form. How do you categorize your own skills?
Understanding the skills you have and the skills you need shouldn’t be so hard.
TeamFit can quickly and precisely give you the skill insights you have always wanted.
Steven Forth is a Co-Founder of TeamFit. See his Skill Profile. Surveys of various types have been a popular tool for people in talent and performance management. So popular that many of us feel surveyed out. I confess …
Gregory Ronczewski leads the design of the TeamFit platform - view his profile on TeamFit Top image: Hotel "Pod Orlem" (The Eagle) in Bydgoszcz, around 1910, fotopolska.eu I remember well my first real job. I felt so …
Steven Forth is a Co-Founder of TeamFit. See his Skill Profile. Product management is emerging as one of the key roles in the innovation economy. Many of the fastest growing companies position themselves as 'product first' and events …
Gregory Ronczewski leads the design of the TeamFit platform - view his profile on TeamFit Top image: East and North faces of Matterhorn. Photo: Zacharie Grossen Vision is the most powerful of all our senses. Over 80 …