Distribution of need for Data Science skills and tools

In the foreseeable future, at least rudimentary forms of data analysis will find its way to many jobs, much as email has done.

However, I don’t think that everyone needs to be a deep-learning ninja, needs to know what Kuburnetes is or even how to do a docker-hub (😝). In fact, I think there will be relatively little need for such skills (note the italics).

I also think that data visualisation skills are and will be extremely important. It’s a huge part of the interface between data and people. However, I don’t think everyone needs to be a Tableau Master, D3JS guru or even R-ficiando. Basic graphing ability with some improvements over Excel defaults will a huge gain for many people.

Obviously the above specialised skills do have immense value, but I think the value of data skills is equally present across the spectrum of skill levels. Here’s a diagram to illustrate:

This triangle of data-skills and tooling are relative to the total “size” of available data and technology. As data continues to grow, the “size” of the triangle will grow along with it. Huge gains in efficiencies and productivity can be made when a large portion of the work needing done at the “bottom” of the pyramid are reliably and repeatedly implemented via a few simple calculations in Excel. There will be far more opportunities to realise these gains at the bottom of the pyramid than at the top. The aggregate value created by the base skills is equal to the value created at any other level. If I am ever surprised at how wrong this last statement is, it’ll probably be because the aggregate value at the bottom far outweighs that at the top.

This also leads to the idea of data science journey of an organisation. There are some good articles around about this, but suffice to say the general advice (which I 100% agree with), is that companies who want to jump into deep learning when they don’t yet have even a basic data culture in place are likely to waste a lot of time and money.

So why am I saying all of this? I think that the work being done at any level needs to take into consideration of the work at other levels and recognise their equal value. That will enable all levels to create the maximum possible value for society. It will help people feel less constrained to their own level and freely move between levels wherever their scarce and valuable skills are needed.

Anyway nothing more to say about this for now. Thanks for reading!