You only have to have tried setting up Google Analytics for a site to begin to realise how technical it can get. I don’t think it is beyond anyone to grasp and eventually master assisted conversions, A/B testing, setting up goals and funnels and developing meaningful dashboards to understand your traffic data but it certainly is a technical and potentially overwhelming endeavour. An endeavour that is worthwhile, dare I say absolutely crucial, to pursue but one that usually requires expert knowledge.
With a properly configured analytics setup and development plan you will grow your business and be able to make informed adjustments to your marketing while empowering your organisation to refine business processes more strategically.
Website analytics is just one example of where the correct analysis of data can be used to gain strategic advantage for your business. Data science is emerging as a profession and careers are being forged for people self-titled “Data Scientist”. The role currently adds great value. Going back to the Google Analytics example, someone with knowledge can save you substantial time and fast track the implementation and gains but looking at the machine learning and other trends surfacing, I do wonder for how much longer…
According to WikiPedia, Machine learning is the subfield of computer science that “gives computers the ability to learn without being explicitly programmed”. With this definition in mind surely the role of the “data scientist”, even being a new and emerging one, will become redundant. Obviously timeframes, like in any prediction, need to factor in as a variable to provide relevance to the thesis. My feelings are that machine learning, with respect to data analytics, is going to accelerate at pace rendering the Data Scientist role redundant sooner rather than later – it is certainly not a career I would encourage my little daughter to set her eyes on.
PowerBI is an analytics tool developed by Microsoft which allows users to produce really beautiful and dynamic data visualisations. Currently it is largely still in the user’s (data scientists) control to produce the visualisations in a meaningful way through a combination of industry understanding, data “massaging”, interpretation and creative dashboard building but there are already built-in algorithms that take a first crack at interpreting any imported data for you. I have to admit that PowerBI machine’s interpretation of the data on one of my recent projects produced a chart that told a story about the data that I had not considered. I ended up folding the chart into the dashboard for the project. This is a tiny step for machine learning in analytics but I feel could signal the start of a giant step for progression where the automated and immediate interpretation of the data yields results that satisfy the needs of the majority case. This will be truly empowering to small businesses that do not have the in-house abilities or cannot afford to hire expert “Data Scientists”.
Back to, and finishing off with, Google Analytics (GA); it is an amazing test bed for innovative data analytics and coming from Google you can rest assured that there are algorithms (the foundation of machine learning) driving the analysis. For Google it all started for them with the search algorithm and again you can rest assured that immense work is going into the analytics platform which monitors the use of its original product. It is a great place to experience beta versions of their thinking and early stage machine learning.