4 ways you can improve your data analysis proficiency
Four ways non-technical employees can become more conversant in data analysis.
The terms, ‘analytics’ and ‘data analytics’ are not just buzzwords in today’s business environment, Analytics is a field for which there is great demand for skilled labour, but limited supply. Moreover, that demand is projected to grow well into the future, and it is unclear whether the labour supply will ever meet demand.
It is therefore likely, that increasingly, organisations will require a large portion of their employees to possess some knowledge or skill in data analytics. Those skills would not need to be as extensive as those of a specialist, but at the very least, some familiarity would be expected.
Below we suggest a few ways to improve your data analysis competence, which do not require you to be a versed in using any specific applications, but are critical and can be applied in a variety of situations.
1. Be clear about the desired output and outcome
As obvious as this point might be, all too frequently, the desired outcome of a particular analytics exercise is not clearly defined or expressed, and consequently, the outputs may not facilitate those outcomes. Typically, analysis exercises are considered holistically, and the required steps are mapped out – with the end in mind. Further, data can be processed to address a broad range of situations or problems. It is thus crucial for any team to be clear and agree upon the desired outcomes – which may not be as easy as it seems – as everything else will flow from there.
2. Ensure the needed data is being collected
For uninitiated, and even among those who you hope would know better, there is an assumption that for any question asked, the data is readily available to provide the desired answers. That is not the case. Systems must be put in place first, to collect or generate that data.
For example, regular surveys might need to be conducted, which requires survey instruments, such as a questionnaire, to be prepared and then administered. Thereafter, there is the likely challenge of securing an adequate number of valid responses, and ensuring that the data received is in a format that allows for easy manipulation and processing. Essentially, data collection can be a tedious and time consuming process, when cannot be implemented ‘now for now’. Hence, it is always advisable to ensure that from the outset, the needed data is being collected.
3. Understand what is possible with the data collected/available
Although meticulous plans may have been made, inevitably, something comes out of left field that an already established analysis exercise might need to accommodate, such as a request from a senior executive, for which the data already collected could be used. In that regard, some thought should be given to the ways in which the data that is being collected can be used, and correspondingly, in what ways it might be inadequate for certain needs.
Further, it is important to emphasise that the results generated from data collected, and the analysis made, is not infallible. Hence, there ought to be some appreciation of the margin of error, and the extent to which the results can be relied upon. Such considerations are even more critical when when proxy or third-party data is being used to approximate your organisation’s field or situation.
4. Learn the basics of a few software applications
Finally, in order to deepen the types of discussions you can have within your organisation, along with improving your own understanding of data analytics, it would be beneficial to learn, at the very least, the rudiments of at least one software application. The application selected may one used by your organisation,, such as Tableau, Pentaho, Google Analytics Premium, and Wave Analytics, to name a few, and so directly relevant. Alternatively, you may wish to start with something like Microsoft Excel, which most of us are familiar with, and is widely used in the corporate world, but is woefully under-utilised, based on the features and capabilities it actually possesses.
In summary, the above tips are geared towards the non-technical employee, who needs to participate in data analysis discussions, understand what is going on, and ultimately, many need to be in a position to ask the right questions. Although the field can be quite intimidating. It is still possible, and even necessary, that members of the team can offer a pragmatic perspective: to ensure that the analysis produced satisfied the actual needs of the organisation, and is sufficiently authoritative to support decision making.
Image credit: NEC Corporation of America (flickr)