Data Analytics
You can't live in today's world without being able to speak the language of 'data'. Advances in sensors, computing, storage, communications, and personal devices (e.g. cell phones) have lead to a world where the amount of digital data doubles every two years. Gathering and storing the data is one thing, but putting it to use is another. We are here to help.
Updated on March 23, 2021
capabilities data analytics
2 min READ
To us data is a means to an end. Data is only useful if you know exactly what you are going to do with it and how you are going to do it. Only then can you really foresee what data you will need and how it should be structured and analyzed. Only then too will you have some idea of what kinds of questions the data might help you answer.
We understand the “I might need this later, so I will just collect it now” mindset and we actually agree with this type of thinking. However, it is all too easy to get overwhelmed by this stage of the process or simply get overly excited & hyped in thinking that because we collect so much more data, we will therefore know so much more about our problem. Far too often, the analysis then devolves into “Well, let’s just look at the data, see what it says, and go from there” - a situation ripe for bending statistics to match pre-concieved notions.
What’s often missing up-front in the analytics phase is the specific What and How. What is the first question you are trying to answer? What are the possible outcomes? How would these outcomes change your thinking or How much value would each outcome bring you? Based on these possible outcomes, What are the anticipated next questions? What information do you need to answer these questions and How are you going to get it? And finally How are you going to use data to answer these primary and secondary questions?
We differentiate our approach to data analytics problems by first talking with stake-holders, oftentimes non-technical people, to ensure everyone has a clear understanding of the plan ahead. Sometimes we find that although troves of data exist, they aren’t sufficient to answer the problems at hand. In these cases we can support the curation of auxilliary data or lean on our simulation expertise to generate synthetic data as necessary.