Quoin led the development of a sophisticated platform for media monitoring and analytics. Our project team was responsible for technology selection, system architecture, and initial implementation of the system, which emphasizes high-performance delivery of analyzed content and metrics. We have continued to support the development of the platform and services, including use of mobile app and mobile web clients, as well as the evolution of the core architecture.
Since 2007, Quoin has worked with PublicRelay on the architecture, design and development of their media analytics Software-as-a-Service (SaaS) platform. Quoin has played a central role in the growth and evolution of the platform's core architecture, from PublicRelay’s earliest stages as a start-up, through prototype and commercial release, to its current fully developed technology stack. The PublicRelay platform provides big data analytics and actionable intelligence to organizations including dozens of Fortune 500 companies and some of the most recognizable brands in the world.
PublicRelay’s platform performs monitoring and complex analysis of media coverage and public perception of client companies, providing competitive intelligence and insight into metrics such as brand reputation and employee satisfaction. The robust architecture Quoin designed and implemented enables the PublicRelay platform to process and analyze a daily average of 7 million distinct news articles, blog posts, and social media content via more than 35 different feeds including various news aggregators and social media platforms. For example, Quoin implemented machine learning for predictive analytics to measure sentiment and likely relevance of content. The resulting dataset is made available to customers via custom web and mobile dashboards.
Our project team built a flexible and extensible architecture for content aggregation and analytics that supports the implementation of natural language processing and machine learning techniques. While the exact PublicRelay tech stack is proprietary, this provides a basic blueprint of the application.
A critical aspect of our work was the conceptual design and implementation of content metrics for the relevance of content and sources. These algorithms use both standard statistical methods and a range of machine learning techniques to continuously improve the quality of key metrics. Since PublicRelay uses human-aided artificial intelligence, we also mapped which aspects of content analysis would be performed by the application and which aspects would be performed by users (with application guidance).
Visualization of analytical data is an integral aspect of the design and implementation of the PublicRelay platform. Quoin’s project team created the initial user-interface for this product and many of the modes for end-user interaction with complex content sets and associated metrics. These included the client and media analyst user-interface, as well as pre-defined charts and graphs for key content metrics. Many of these automated reports are displayed below.
PublicRelay clients use this interface to track media coverage and public sentiment surrounding their brands.
This visualization breaks down public sentiment by timeframe, sector, and overall tone.
Users can also view graphics of public sentiment by tone and subject matter over time.
Quoin designed PublicRelay's user interface to be fully responsive whether on desktop, tablet, or mobile devices, as illustrated in this mock-up of the site on a mobile screen.