HCI Design Principles and Visual Analytics for Media Analytics Platform

Media industries and advertisers are increasingly turning to big data analytics to better understand audience media consumption patterns, as evidenced by Canada’s Globe and Mail’s applications Sophi and TasteGraph [1, 2]. Data analytics and interface design provide complementary perspectives for large datasets. HCI design principles have been applied to Business Intelligence (BI) platforms, including techniques which filter and summarize large data sets, and are equally relevant to media informatics platforms for advertisers, buyers, sellers, and planners [1, 7‐9]. This analysis becomes more challenging when managing highly scalable and multi-dimensional audience survey data [3‐6]. According to Dewdney and Ride visualization tools are essential for effective decision making in the communications industry as these ease cognitive load and decision-making [15]. Kirk proves that a visualization system becomes a successful tool when it builds on the user’s extant domain knowledge, providing enhanced insights [13]. The research aims to leverage visualization design principles as defined by Tulp and Meirelles [6, 7] and in order to improve the UI/UX and visual analytic capabilities of a leading media analytics platform providing planners, advertisers, and media buyers with an interface to better understand their audience. We have analyzed and assessed the different application report parameters that explore television and radio survey datasets from a leading analytics firm. We propose design prototypes which are comprised of enhanced symbolic icons [9] through badges and glyphs, consistent colours [10], and layouts which maintain a visual hierarchy and filtration techniques [10, 11, 14] in order to minimize information clutter and cognitive overload. We propose a variety of interface designs that address user needs using HCI, heuristic design principles and novel visualization techniques [6, 7, 12, 15]. Next steps include validating our design prototypes through rigorous user testing and building high fidelity prototypes.