Skip to content

The Covid-19 pandemic has shaken this world in literal sense. Many things that used to be the norm just a few months back have become obsolete. One such casualty is big data, which was an approach used by many organizations for analytics.

It was traditionally believed that the larger the size of your data set for analytics, the more useful and accurate insights will you be able to draw from it. However, all these presumptions completely changed as a result of the pandemic.

This global health crisis has created such a pronounced deviation from past trends that Data and Analytics (D&A) experts are completely baffled. As a result, the accuracy of many Artificial Intelligence (AI) and Machine Learning (ML) systems has drastically reduced.

According to research giant Gartner, this scenario has paved the way for a paradigm shift in how data is perceived and used now. We are now entering the era of two new analytical techniques called “small data” and “wide data”.

Both these terms may sound conflicting to you at first glance but fact of the matter is, both these D&A methods complement each other. Small data tends to deliver useful insights by leveraging less data via time series analysis techniques and a few others.

Wide data D&A technique revolves around analyzing and synergizing both structured and un-structured data, regardless of whether the data sets are directly correlated or not. Wide data aims to extract or identify links between heterogeneous data sets.

The small and wide data techniques are expected to yield positive results is retail demand forecasting, real time behavioral intelligence, hyper personalization and improvement of the overall customer experience.

Gartner estimates that by the year 2025, nearly 70% of organizations will shift their focus from big data towards small and wide data. The use of these techniques is likely to make AI less dependent on vast amounts of data.

Although the exact outcome of small and wide data techniques is yet to come, the concept is certainly promising as these techniques are capable of delivering actionable insights in areas where past or historical trends are no more relevant.

Share This Story, Choose Your Platform!