You do not need me to preach the importance of data-driven decision making, I am sure that job has already been fulfilled by a raft of articles, seminars and management meetings. However, not enough businesses are really taking action and making full use of the means, and the data, at their disposal. We are here looking out for you businesses, and thus it is vital to first understand the hierarchy that exists within data science, namely Data > Information > Knowledge > Understanding > Wisdom or DIKUW for short. Then we can begin to properly explain the possible steps towards better decisions. Let’s get started.
Just a bunch of symbols
Everything at its simplest is symbols and characters. Imagine an Excel spreadsheet with a range of unlabelled numbers, symbols and letters on it. You could see an entry “2009”: is that a year, PIN number or a reference code? Who knows? It’s meaningless. That’s why the whole “Big Data” label is not endorsed at Satalia because companies having mountains of data is not advantageous, it’s what is done with the data that makes it valuable.
“Make sure to pick the metrics that matter for your specific business.”
Information is data with meaning and context. “2009” is a year – suddenly it has some meaning. Data is given descriptive labels and is organised appropriately in order to achieve information. Not all information may be especially useful to a certain business. Make sure to pick metrics that matter to your business objectives and are relatively simple to collect.
Knowledge is achieved through connecting, processing and summarising information in a way that describes what has happened. Imagine a young child had memorised their 12 times table so that they could answer 12 x 12 = 144 off the tip of their tongue. Then if you asked 16 x 18 they would struggle due to this not being within their memorised timetable knowledge base and them not actually understanding the process for relatively difficult mental multiplications. Overall, having knowledge is realising how something happened but not why it happened.
Interpreting why something happened based on new knowledge presented to you and previously held knowledge is crucial to making sound decisions based on past business information. As we build this understanding we can begin to make predictions about what will happen . Furthermore, machine learning and artificial intelligence systems can learn and iterate off knowledge gained from previous experience, human input and other sources.
Utilising the what, how and why
Finally, wisdom is the utilisation of understanding. It is the only forward-looking part of the pyramid. Having wisdom in decisions allows individuals and businesses to utilise available knowledge and weigh up consequences and outcomes in order to make decisions to achieve goals and objectives in the future. Ultimately, this is the capability that what we want systems to achieve and the realm of ‘prescriptive analytics‘.
Phases of analytics in businesses
Here is a very broad overview of the phases of analytics that a company must master in order to make some claim of having business intelligence.
|Data mining||The collection, sorting and understanding of public, competitor, customer and business data. Important to specify the metrics important to your business.|
|Descriptive Analytics||Presenting and reporting data, information, knowledge.|
|Predictive Analytics||Anticipating outcomes based on synthesised data.|
|Prescriptive Analytics||The use of optimisation algorithms to decide the best course of action.|