Employees who relish the world of data-led decision-making are happier and more productive, and that alone is worth exploring to see where the magic resides.
There is something that makes someone who is part of the data-culture different. But it’s not one thing - its a series of mindsets, behaviours, skills, and attitudes. And whilst some might be born to the data life, naturally gifted with inquisitiveness, numerical skills, and problem solving, becoming a problem solver and analytical champion is definitely a career path and destiny that is open to all. In the spirit of demystifying the magic for a broader audience, here some of the key factors that make the data worker the perfect analytical hero - and what makes them different from non-analytical workers.
Mindset and attitudes
There are some attitudes, some characteristics, that enable the data worker to problem solve and leverage the variety of analytical tools and tricks open to them to make real-world changes that conform to our best understanding of the past and our predictions of the future.
This person is above all inquisitive - a problem solver. Someone who has a burning need to really KNOW what the causes and reasons of a challenge are - and then the determination to plan and enact the solution to that challenge.
They will likely be a self-improver. Someone committed to upskilling themselves because they will be comfortable quantifying their worth in addition to that of other assets. As such, it’s likely that they will, consciously or not, apply mental models to their life.
There are a wide variety of these, and well-worth looking into for personal as well as professional development. Books have been filled with these models. They range from useful ‘life hacks’ to very detailed models for use in specific and even scientific circumstances, but they share a common goal: To allow a practitioner to look at a problem in new and helpful ways that break them out of their habitual ways of thought for better, more accurate results.
Some very simple mental models that a new data worker will want to know include:
1. Second- and third-order effects
Every activity or action has consequences that may not be predictable. In turn, these consequences have others down the line in a cascade. These second-order effects occur with every change made. There are examples, like New York’s Federal rent controls from the 1940s-1950s, whereby a noble goal resulted in the opposite impact desired (landlords couldn’t raise prices and let tenements crumble as they couldn’t afford to keep them up on their rental income).
People tend to think in certain, evolved ways. We like to plan for a successful future. We don’t tend to ask questions like ‘how could this go wrong’. Shifting your thinking from only planning for the best and adding in additional planning to avoid failure is a valuable tool for the budding data worker. It’s a simple way to cover more mental ground and ensure more complete decision-making is made.
3. The five ‘whys’
A very simple diagnosis tool. Simply ask why something occurs - and with the next answer, ask why that occurs too. Drilling back a certain number, say ‘five’, should eventually result in discovering the real, prime cause of failures or successes. This can then be fed back into the system to avoid that failure, or to better build on the prime source of success.
Ideally the data worker have some key thoughts always front-of-mind that help guide them when they feel about to make a ‘gut’ decision, without careful thought, for example:
"I am easy to fool" - they will acknowledge confirmation and experimenter bias is easy to fall into
"Everyone knows what it takes to change my mind" - they encourage a culture of critical thinking and they ‘avoid congruence bias’ (they test alternative hypotheses)
"A good explanation can be proven wrong" an unfalsifiable position is not a logical one
"The strength of my belief is proportional to the strength of the evidence" - there exist degrees of plausibility, and data workers are comfortable working in these different levels
Blinding analyses -- not letting expectation of result drive model revision
Answering the question: What would it look like if your hypothesis were wrong for a hypothetical case? This can help take the sting out of looking for flaws in a treasured project, for example!
They will also be a determined person, a doer as well as a thinker. Some may be analysts, or data scientists, and some might be citizen data scientists (with the will, if not the formal skills and training to make use of data at some level) - after all, if they’ve spent the effort to learn something about data, they have shown they are no dilettante: The terminology and skill-set is learnable, but very different to regular language and business activity. It can put the less determined off, if they don’t have this desire to carry through self-improvement.
Data workers of decades past had a harder job - no question. Data analysts needed to learn scripting languages, querying languages, advanced spreadsheet skills, and statistical languages or programming.
Software was clunky and required coding skills to use, and often IT held the keys to unlocking the data a worker used in their daily life. Many data workers may want to pick up these programming languages to use certain tools - but in the twenty-first century this isn’t strictly needed: Self-service analytics platform solutions can guide data workers of any level through data discovery and understanding its context, its preparation, the crucial analysis itself, and then the sharing and deployment. Today this can be done ‘code-free’ or ‘zero-code’ - meaning that anyone, particularly those citizen data scientists, can get involved using the skills and knowledge from their life and career, without needing to learn a whole new profession.
Those softer, but no less important skills of critical thinking are now much more needed as the technology has democratised data, and it is now within the power of anyone to improve their world using data plus easy-to-use solutions to crunch the raw materials and deliver insights.
If the data workers is adept at conceptualising, analysing, synthesising, and evaluating data, they have the mental tools to manage the challenge, and to work in the technical elements required to succeed with business data projects.
Alan Gibson, SVP EMEA, Alteryx