The appliance of data science
Finance functions are being tasked with making more out of the data they have at hand… but can you go a step further? In his new book, Stylianos Kampakis discusses how to automate decisions based on setting up algorithms that analyse the data with
What is ‘data science’?
Data science is about using data to do useful stuff. Short and to the point. That’s exactly what data science is all about. The methods we use are important, of course, but the essence of this discipline is that it allows us to take data and transform it so we can use it to do useful things.
Data science involves everything from statistics and pattern recognition to business analysis and communication. It requires creative thinking as much as it requires analytical thinking.
Data science also involves discovering which data is useful as well as effective ways of managing it.
Furthermore, business acumen is also a necessity because while data science can be applied to any field of business, it is critical to know what types of answers the business needs and how to present said insights so leadership can understand them.
Being ‘data-informed’ and ‘data-driven’
Let’s start off by looking at the difference between being data-informed and data-driven. When we say that an organisation or a person is data-informed, it means that they are actually using data and the context of data as inputs into their conversation and decision-making process. So, if you’re using things such as dashboards and KPIs, for example, then you are data-informed. Nowadays, it’s very easy to be data-informed.
You can use Excel or another simple tool and it’s not that difficult to make a few charts to see what’s happening within your organisation. Being data-driven means taking it to the next level. This is when you start using more intelligent algorithms and methods to get the data.
Then you transform the decision-making process by using the algorithm to make the decision for you, or by taking the algorithm’s output into account. What’s the opposite of this approach? Well, the opposite of being data-informed or data-driven is when you use your guts to make a decision. Or tradition, which isn’t much better in terms of accuracy or effectiveness. Let’s take a look at a few more examples.
When you are data-informed you might be using a dashboard or Excel to collect data from multiple sources. You have, of course, organised the data appropriately.
The data is also accessible, and all your employees know about it and are aware of where it can be located and how they can use it. Being data-driven means taking things a step further. This is where most companies should strive to be in the future.
When you get to this point, it means that you are allowing the data and data science to help you make potentially disruptive decisions. In some cases, they might even completely replace human decision-making. So, if you trade using algorithms, for example, then you are data-driven.
If your app or website uses a recommender system, then you are data-driven.
Obviously, being data-driven does not mean that an algorithm needs to make every decision, but it does mean that you use algorithmic outputs in some parts of the organisation and in the decision-making process. So, why would you want to do this?
Being data-driven improves efficiency and can actually improve decision making. For example, a recommender system can make better decisions than humans as to what the client might want. Also, if you want to sell at a massive scale, after a certain point you need to have an algorithm that can actually replace humans when recommending items or curating content. This is just one example, but there are many others in different industries.
MONEYBALL… AND YOUR BUSINESS’ DATA CULTURE
I’d like to talk a little about the movie Moneyball. I’m not sure how many of you have seen this film, but if you haven’t, I urge you to do so… it’s all about building a data-centric culture within an organisation.
The film is based on the book with the same name and, basically, describes Billy Beane’s work at Oakland Athletics. This baseball team was pretty much one of the worst teams in the American baseball championship.
However, thanks to Billy Beane’s efforts, the team basically goes from the bottom of the pile to being one of the best teams and actually a lead contender. Bean did this by using analytics and statistics.
Essentially, he analysed players and even though their individual performance was poor, he saw that together, they would make a great team. So, he acquired them and because of their poor individual performance, other teams thought they were losers and they were cheap to acquire.
He managed to put together an amazingly good team on a budget. Throughout the team, you see the resistance Brad Pitt – who plays the role of Billy Beane – encounters because of the traditional ways of thinking practised by veterans in the club. They simply will not accept the new approach he’s trying to implement. Of course, he proves them wrong.
Dr Stylianos Kampakis, data scientist, thedatascientist.com