Posted by: Nayab Naseer
Big data, in essence, entails churning all available pieces of information, be it text, statistics, databases, (on-hand or obtainable) to derive meaningful and actionable insights.
Businesses tend to focus on the first part of the initiative. They refine the process to make data collection and integration easy. But they forget that big data is only a means to an end. Without big data throwing up “big answers”, it becomes an exercise in futility. Big data analytics should ideally provide real information that allows business executives to see the complete picture, enabling them to make better decisions, and avoid repeating the mistakes.
Following are some tips towards this end:
- When co-opting data, do not ignore any source of information, be it online or offline. Even if the information seems to be out of context or make no sense now, it may do, at a later stage, when all the information adds up.
- A big data initiative usually comes with an enterprise-level data governance framework to collect and feed data into the system. However, such a framework should not come at the expense of flexibility. Too much rigidity may simply make things too difficult, and become counterproductive. It is important to strike a balance between the two.
- Merely integrating all sources of data into a single repository is not enough. Have a system in place to ensure that the data collection and analysis is seamless. The harder or time-consuming the process, the less inclined the rank-and-file would be to partake in the initiative wholeheartedly. This leads to incomplete data, which in turn leads to wrong analysis, and ultimately faulty decision-making.
- An initiative to change the organizational culture, to promote transparency, and sharing of information has to accompany the big data initiative. Without a culture of openly sharing data, the data in the centralized repository would remain incomplete. Without an open mind, the employees would be reluctant to feed information into the centralized repository, fearing that such information may be used against them.
The correct approach to big data is to work on collecting and analyzing data objectively. While you can use Big data to gain insights or validate assumptions, it is necessary to feed in all available data, and subject it to the right analysis.