There’s something fascinating about working with data – it’s everything and nothing, all at once. It’s meaningless without context, and yet context cannot begin to give it flesh. That’s where analysts come in handy, wrangling insights from the metrics that animate the information.
The brain is incredibly complex, neurotransmitters firing their explosive blasts more frantically than an emotionally-charged battle scene. Your body is too slow to perceive this, it just carries out voluntary and involuntary responses. Your mind is stranger still, an invisible web of connectivity and memory that’s holding it all together. And this is where the data analysts play, somewhere between the brain and the mind and the actions they produce – insights, potentially, too, that can allow you into a space where you can predict with relative ease what actions your customers may perform.
Insights come from knowing what you want from data; analytics must be governed by the appropriate metrics to yield useful information. What you get out, in other words, depends on how well you frame and define the research problem.
Let’s say you are interested in gaining insights from the data produced in your contact centre around agent activities and the service they provide. You can monitor the time spent on calls, successful conclusion of calls, call response time and so on; these are a good place to start, but they don’t necessarily give you enough information about the quality of the interaction: did the customer have to call several times? Did the agent rush through the call to boost call time metrics? Is the contact centre adequately equipped with sufficient staffing to respond to calls, and does the system accurately know when agents are at their desks? For the latter, if your system is routing calls to an empty desk while the agent is having coffee, for example, your customers are the ones suffering through long hold or connection times.