The word “analytics” means different things to different people. I sometimes wish the word didn’t exist, as it would force us all to talk more about exactly what we are meaning by using the word.
Here are the top five reasons why I think analytics is overrated.
Using analytics to do things like generate insights into our customer base is valuable. We can find out who our most profitable customers, perform micro-segmentation to understand how different segments behave and predict their future behaviour. But if we aren’t ready to act on these insights, then the analytics effort was mostly wasted (or at best premature). For example, we should incorporate the outputs of our analytical modelling into our actual business - things like directing high-value customers through to a high-touch call centre when they phone in, or by rewarding our most profitable customers with priority access to our services.
The multiple forces impacting IT departments (commoditisation, automation, centralisation and cloud-everything) mean that IT teams now have more time (and realise it is more valuable) to devote their efforts to “analytics” projects. This is fine, as no doubt technology is an important part of successful analytics - but this shouldn’t be thought of as an “implementation”. Analytics is far wider than just a technology project as it involves setting up the right teams of people, processes and ongoing development to ensure that as we mature with analytics, we stop "doing analytics” and it just turns into the way we do business every day.
In the race to the result, we often overlook important aspects of the process of analytics - the reasoning, questions, interrogating that helps with understanding the information produced. I’m not suggesting we adopt the mindset of a two year-old incessantly asking “why” to every answer, but by stopping and thinking and potentially adjusting course as we progress, we limit the risk of achieving a sub-standard result and will probably find out interesting and valuable things about our business that we previously didn’t know or hadn’t considered.
Because we all desire to see something tangible before emotionally buying-in (seeing is believing) we often rush to create things using tools, at the expense of planning and designing effective organisational structures, systems and processes. Absolutely, we should be starting small, testing and learning, failing fast and iterating along the journey - but having a strategy in place (even if we decide to change it as we progress) will ensure that we don’t sink vast time and money into one or more tools at the expense of others. I like the concept of combining an app store with a tool bag. We acquire multiple tools that can be used for different tasks at different times and we take them out of the tool bag when we need them and put them back when we don’t. In the same way that once we tire or grow out of angry birds, we delete it and move to another tool, whilst we shouldn’t like a magpie just jump from one shiny thing to the next, technology is moving so quickly and the commercial models support continuously adopting different tools for the task at hand.
Former NBA star Charles Barkley articulated what a lot of senior executives sometimes think about analytics, when he referred to the General Manager of the Houston Rockets as "one of those idiots who believes in analytics.”
By promoting the rise of the data scientist and creating mystique around sophisticated data mining and algorithmic modelling, we risk building a wall between the analytics team and the business team.
We create division between the off-court analysts and the on-court basketball players in Mr Barkley’s example.
Analytics is changing the world we live in, our personal and business lives every single day. By choosing to use this new power intelligently, by understanding all facets, planning and developing strategically we can really leverage this to our benefit and use it for competitive advantage and to enhance our lives - but we need to avoid some of the traps I’ve outline above. We need to work with teams who are not just analytics experts, but also understand the application of analytics in a broader sense.