What's the best data source to determine a person's skill set? This specific question comes our way rather often -- it's a very sensible one, too: where should you look to find an accessible summary for a person's abilities? In almost every situation, the first answer that comes to mind involves job titles. This might be entirely unsurprising, given that job titles exist only to describe the roles that people work in. However, there is one big reason why I'm not a big fan of these terms: they reduce people to a single dimension.

Hiring Efficiency

From our experience processing hundreds of millions of vacancies, job descriptions often don't rise to their name entirely: they describe the company culture, benefits and more, but increasingly miss out on providing a full overview of the content of the job at hand. Again, this is hardly a surprise: people have short attention spans, so why would you want to tell them things they already figured out from the job title? From an efficiency point of view, vacancy titles are a godsend: in just a handful of words, they set the scene entirely, to the extent that the contents of a vacancy text itself don't even matter that more. Looking at job postings for accountants, for example, it's not even that rare to run into vacancies that forget even to mention that accounting is a required skill.

That's not to say that firms forget to mention the essential skills for a job entirely: when you collect a thousand vacancies with the same title and group their mentioned skills, you end up with a surprisingly accurate description of the job at hand. At the same time, out of those 1000, only a few will reflect the same contents accurately on their own. This can be attributed to the fact that vacancies are written with their titles in mind: the writer assumes certain background information is conveyed through the title. Any gap between this base concept and the actual open position is then bridged through the description. While this is to some extent an oversimplification of the actual process, it's not that far off, and applying this same idea to the machine interpretation of documents inside and outside of HR has proven very successful.

Despite this informational efficiency, a first problem lies in the very nature of a job title, which stresses what you are rather than what you do, and importantly, to what end. Research has shown that this distinction is perceived in online recruitment, where Millennials show a preference of job action statements over job titles (White et al., 2019). For example, subjects showed a preference for "Lead high-performance teams to increase company revenue" over the more traditional "Sales Manager". It's not surprising to see this specifically in context to Millennials: the generation's view on their job shows a clear expectation of change over time, either through changes within their jobs or by switching jobs entirely. Within this view on work, it's only to be expected that the emphasis moves towards doing rather than being.

Compression Artifacts

Seeing this as a problem with Millennials and their world view would be short-sighted, however. Their perspective translates to a similar trend in work, which is not restricted to a single generation: agile. While some companies might consider agile work to be no more than putting their work on a kanban or in sprints, the trend goes deeper, breaking down the silos that have previously characterised many occupations. This break away from the traditional approach means individuals develop and apply their talents differently, and arguably in a much less predictable manner. In contrast, job titles used within companies are typically static, changing only once every couple of years and even then along the same traditional paths.

When we look at a typical hiring process, it seems that most organisations tend to agree that an in-depth knowledge of a candidate's skill set is crucial to determine whether they're fit for the job. It's somewhat ironic then that, after a hiring flow in which every aspect and skill of a candidate is studied with the utmost care, they are placed inside the organisation labelled just with their job title. After all, if that very same company were to buy and install a new laptop, every capability and piece of software on the device would be known and managed with precision. So why do we reduce people to a single phrase?

To some extent, the answer lies in convenience: at the point of onboarding, a new employee and the position they were hired for are deemed to contain the same skill set -- or at least any remaining difference is considered to soon be taken care of, and this idea somehow negates the need to keep this information on file. It's a bit like how you might tell yourself you don't need to write down a password you just came up with, only to realise you only know the gist of it anymore when you need it a couple of weeks down. However, while you as an individual might take more care the next time, large enterprises are generally terrible at iterating on their processes. Secondly, and most importantly, a substantial amount of information is lost in translation between different HR processes, if it is communicated at all. The disconnect between different phases of the employee lifecycle means that many things don't make it into a shared system of record, despite being known at some point throughout the cycle. Job titles, on the other hand, always do.

Less is More

All that said, there's a strong case against the continued use of job titles to describe people. On the other hand, there's no denying that just dropping them would leave a gaping hole in how HR approaches internal roles and recruitment. Interestingly, modern competency frameworks have tackled this issue in two distinct ways, which are each other's mirror image. On one hand, the competency standard zooms out to replace job titles with more general occupation titles -- on the other, the framework zooms in to break jobs down into high-level competencies. As such, these frameworks create a compact, shared language to describe talent: using a base occupation as well as added competencies, an accurate picture can be created of any employee's skill set, using a handful of items rather than hundreds.

The reduced complexity of these competency frameworks enables people outside of HR to describe skill sets as well, potentially giving them the power to take control of their own listed competencies. These competency profiles can then be leveraged in learning & development or talent mobility, benefiting both the employee and the company substantially. That's not to say that manually describing skill sets doesn't still create a lot of friction -- but I guess that's nothing a little bit of AI can't fix.