Some number of years ago, I got caught by my boss cramming on a technical topic. He asked me what I was doing, and I responded in a rather smug way that I was learning something, so I knew it better than anyone else; therefore, I was invaluable and indispensable.

He remarked that I had just screwed my career. Rather than being the only one to know something, the best way to be indispensable is to be replaceable.

His theory was that if you are the only one to know something, then the likelihood of you progressing was zero as your boss would never be able to replace you and therefore you would get stuck.

Instead of taking this track, what if one were to take the approach of documenting what you know, teaching someone else, and ensuring it’s repeatable. Once you make yourself redundant, you can finally put your hand up requesting the next challenge. These steps will show that you can adapt, progress, be utilitarian and this is how you become invaluable to a company. In turn, you are always challenged and always engaged.

I took this advice to heart and have practiced it fastidiously since the early ’90s, and I must say it has helped me more than any other nugget of career advice.

Driving into the office this morning, I couldn’t help but reflect on the rapid progression of data analytics in the last few years and the understandable but paralyzing fear of what Machine Learning and AI would do for people jobs. Added to this is the obvious, visible gap between what’s in production and what’s possible and we come to the rapid conclusion that in this technology sector, innovation is ahead of the median architectural deployments. Company analytics are not stifled by technology gaps but by people and their fear of disruption and more likely of being disrupted.

I guess this is totally understandable and similar in many industries with the onslaught of automation. But I posit that if people take the leap, become inquisitive, adapt, and push the envelope of progress, then they can ride the technology waves, become more valuable and in turn invaluable to the company.

With respect to analytics (I know it’s not perfect), but I knocked up the below chart to try and show the progressive steps of operational analytics.

The days of being valued for staring at a data feed and hitting a panic button are pretty much over. However, we do seem to be stuck in the next box or so where we have antiquated architectures based on running static, proven rules against, frankly speaking, a pile of stale data in our preferred lake and feeling good about ourselves. I believe we have the misconception that we cannot possibly process the types of complex analytics we enjoy on “at rest” data on a stream of data, but that’s simply not true anymore

There is nothing agile or progressive about the above, and nothing real-time. By making the next leap, we can learn new skills, embrace adaptive Machine Learning in the NOW and move ourselves directly out of the data flow. Let the computers do the heavy lifting. Work smarter, NOT harder.

The proposed path, therefore, goes like this: from staring at data to setting rules, to stepping out of the pipeline, to setting constructs, and finally to monitoring the adaptive organic process.

How cool is that!

As technology progresses, so should we.

Let’s not be the long pole in the tent. It’s rare a space isn’t bound by technology but by people. We are upon this very phenomena.

We should agree to strive to drive innovation to the point that the possibilities should only be bound by people’s imaginations and inspiration and NOT the technology and certainly not our own fears! I promise to do my bit… the question is, will you?