Beyond the catchphrase of data analytics
As a person working with data, I hear the phrases constantly: Self-Service Business Intelligence, Analytics as a service, Data Virtualisation, IOT and other buzzwords.
Tech vendors are raving about their offerings for Machine Learning, Hadoop and Cloud computing etc. With all this it is easy for someone working with data to get caught up in the hype and lose focus of what is important.
These new technologies have changed the way that data is delivered to decision makers in the modern enterprise, heck, even in SME`s and start-ups most people now have the tools and the capability to do gather, massage, visually analyse and even do some advanced analytics on their data. Data and analytics have truly been democratised, and computing power is now available as a commodity in the cloud - readily available at the click of a button.
Why is it then that a lot of people still struggle to find value in their data and analytics programs?
My opinion is that the foundation of any successful analytics strategy is a sound data management and cleansing regime, together with a very clear goal or question that you want answered, something that has never changed when talking about turning data into insight. Doing analytics on data that you don`t trust, with an unclear idea of what you hope to gain from the exercise, rarely produces anything of value.
A well-known anecdote about data scientist is that they spend most of their time gathering in cleaning data, a bit of a waste for someone who often has advanced training in mathematics and statistics. a 2016 study has shown that data scientists often spend 80% of their time in data preparation:
My point here is that having a good plan for getting data, cleaning it and storing it not only saves time and money, but significantly increases your chances of obtaining some actionable insights. Traditional data professionals, who have years of experience in just these regimes and processes are becoming increasingly more valuable, and should not be at odds with those trying to innovate with new technology, one cannot function correctly without the other.
If you are looking at choosing the latest shiny new tech, be very critical in asking what exactly it is you are trying to do with it. No BS. Ramp-up time and costs for most new analytics and data services are now so low that switching technologies or platform midway through is not the show stopper it used to be. There is some space to change your mind, you don`t have to know everything up-front.
Here is to sanity and clarity in the haze of buzzwords! So don't get lost in courses like Coursera offers...