A chance encounter with Crayon’s co-founder Srikant Sastri in Colombo in 2012, led to my introduction to Crayon’s founder Suresh Shankar in Singapore. It took just a few meetings with Suresh, for me to realize that he is a visual story teller. In strict technical terms, he represented a semiotic view of symbols and usefulness, while I represented the ontological view of correctness and completeness of the same data set. The unique blend of insight and imagination in a visual fashion makes Crayon’s messaging stand out in a crowd. My latest article, published on CIO Review explains the importance of visual grammar, thanks to the explosion of business to consumer interfaces through apps on devices. You can read the complete article below:
The Big Data landscape is defined by four Vs- Volume, Velocity, veracity and Variety. It also includes interactive data visuals that convey insights and engage the user. The classical understanding of visualisation in analytics involves presenting results of aggregated data in the form of decision support charts to executives within a corporation. This is to draw up a summary view of derived insights based on source data sets to aid easy decision making. In the context of traditional BI systems, it means that the results derived are visualised based on data models imposed on raw data. This argument can be stretched to state that the visual representation is about causality in data that explains the derived results.
The difference in big data systems, when it comes to visualization is that it’s used to show data in motion. Hence, these are visualisations’ of various data streams that convey their affinity in a visual manner. It is an attempt to show correlations among various data sets. It’s based on the principle that a human brain can uncover a visual pattern much better than a correlation among computed results in text mode. Most big data systems are not deterministic systems and they portray a visual map of multiple sources in terms of their strong and weak affinities. Domain experts interpret a causal relation, if it exists, and take action based on the context.
The rise of social networks and IoT is one of the primary drivers for the renewed interest in data visualisation. In social network analysis, the geo-spatial distance is not necessarily related to the social affinity between two persons. In the context of IoT, the sensor network data in a mesh configuration or in an ad-hoc network, and needs to be visualised to understand the usage or consumption and dispersion patterns. These are much easier in a visual domain as it shows both the right brained and left-brained thinking of the decision makers or experts.
The interest in visual grammar that covers iconography, interaction design, font and colour palette is growing due to the explosion of business to consumer interfaces through apps on devices. Traditional analytics focuses on business users at the executive level, while big data analytics is geared towards the operational layer including consumers. This is also the reason why visualisation in big data is so closely related to design thinking, due to the focus on design ethnography.
Given the exponential growth in this segment in India, it is time for us to develop a visual grammar, which is rooted in an Indian ethos that will appeal to both, rural and urban audiences.
In the current world, everyone wants evidence in terms of data but anyone will appreciate the message through a story based on an underlying data set, where visualization is the story teller.