If you would like to know what my personal the top 3 aspects to carry over from my “old” to my “new” profession are, you are reading the right article. Let’s get started …
1. Designing and realising a great solution is a team effort
To design an efficient car engine you need experts for combustion & thermodynamics, fluid dynamic, simulation, material sciences, cost engineering, customer insights, project management etc. working together effectively. For bringing advanced data analytics into real life in an industrial environment, simply having the brightest data scientist is not sufficient either. Only if a team of data scientists and data engineers, infrastructure experts, domain experts, project managers etc. come together, you can really generate value in the long run.
2. Acknowledging the importance of input quality for the final outcome
From the perspective of mechanical engineering it is obvious that if you - let’s say - want to produce a high-performance windmill gearbox, you will not succeed by using low-end constructional steel. High attention on selecting the right material and then processing it to the highest standards through precision machining and well-controlled heat treatment is a must. The same is true for any data-related task I have come across. You must aim for the highest quality of your input factor “data” from the very beginning - otherwise you have lost the game before you even started applying any analytics.
3. Building quality into the process
Continuing on point 2, data quality is really one of the defining factors in any data science activity. But as with physical goods like washing machines or cars, it is not enough to do quality checks at the end of the assembly line and select good from bad units. Quite the opposite is true, you have to produce this quality at each step in your production chain. Transferred to data, it means you have to make sure that your sources are of decent quality, but also your data pipelines need to be designed to produce the final quality, this includes correct methods, robust checks/tests, and also traceability and documentation.
All that education and experience will not directly make you a digitalisation management expert, of course. Here are my personal major 3 things, I had to learn on my journey.
If you would like to read more of my blog, check out the list of posts here!