Pantea Lotfian PhD, Camrosh Ltd.
Donald Rumsfeld, when serving as secretary of defence under G.W. Bush , famously said:
“There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don’t know. But there are also unknown unknowns. There are things we don’t know we don’t know”.
In the documentary by Eroll Morris, 2013, titled the “Unknown Known”, about Donald Rumsfeld and his years in office, there is another part to this famous quote, that Rumsfeld added when interviewed during the making of the documentary. This may be the result of the years of contemplation and hindsight. He adds, when discussing his own quote that there are also:
“…Unknown Knowns, that is to say things that you think you know, that turn out you did not”.
This part of the quote struck me when thinking about technology, innovation and how businesses deal with the fierce competition they are facing while the tectonics of the competitive landscape is changing through advances in technology and the entry of agile new players to the markets they operate in.
The volume of information available to any business today is gigantic. Businesses often only see the tip of the iceberg of information and get a sense of knowing what has to be known and can be known. Some may also consider that they still don’t know enough, and venture to imagine the “Unknown Unknowns” to derive competitive advantage by being anticipatory rather than reactionary. What often remains unexplored are the “Unknown Knowns”.
The explanation Rumsfeld gives about “Unknown Knowns” in the documentary is more akin to knowledge that in hindsight we discover was a mere perception, or our insistence on seeing only one version of reality, the reality we favoured at the time (“…that is to say things that you think you know that it turns out you did not”).
With the advances of Big Data Analytics and increasing availability of live streaming data through connected devices (the Internet of Things) both at an industry and consumer level, the importance of awareness and willingness to face the “Unknown Knowns” increases dramatically.
Working with large volumes of data has its own challenges. There are conflicting requirements, such as the need for speed vs. control over quality of data, or understanding the intricacies of data and the nature of outliers vs data visualisation in an easy to understand and meaningful way. These requirements will impact on any attempt to use data analytics in any form in an organisation, both to devise competitive strategy or to optimise operations. Also the context that the data is interpreted in and domain knowledge are other dimensions that need to be considered.
Most businesses with a data strategy still mainly focus on collecting data and have not yet managed to fully capitalise on the benefits and capabilities IoT analytics technologies offer. As more businesses start using Big Data and IoT Data Analytics for cost reduction, improving product and services, safety and efficiency, and enhancing customer experience, the competitive value of data will shift in its application towards innovation and strategy design. This means, selecting the right vendor for implementing IoT Data Analytics systems in the company will increasingly become a strategic choice rather than a technical choice. This makes it imperative to select smarter rather than simply invest more.
In one of our projects (http://www.camrosh.com/iot/) we have examined the IoT Data Analytics landscape and discussed key product features and factors to consider when selecting an IoT analytics tool including:
- Data sources (data types and formats analysed by IoT data analytics)
- Data preparation processes (data quality, data profiling, Master Data Management (MDM), data virtualisation and protocols for data collection)
- Data processing and storage (key technologies, data warehousing/vertical scale, horizontal data storage and scale, data streaming processing, data latency, cloud computing and query platforms)
- Data Analysis (technology and methods, intelligence deployment, types of analytics including descriptive, diagnostic, predictive, prescriptive, geospatial analytics and others)
- Data presentation (dashboards, data virtualisation, reporting, and data alerts)
- Administration management, engagement/action features, security and reliability
- Integration and development tools and customisations.
- Scalability and flexibility
- Industry focus and use cases
Given the high importance of the ecosystem structure and positioning of a vendor in this ecosystem, understanding technical differentiators and specialisations of vendor’s offerings is not enough for selecting the right IoT data analytics provider. Considerations for the number and nature of partnerships, the history of M&As, breadth and depth of use cases and other “unknown knowns” become valuable differentiators when selecting IoT Data Analytics tools for building end-to-end solutions. Detailed analysis and mapping of the partnerships formed by IoT analytics vendors presented in the IoT Data Analytics Report shows that the majority of them are interconnected to one or more of the sample set, other IT, Data, Cloud and IoT vendors, and a list of partners from different industries.
Partner Ecosystem Map of Featured IoT Analytics Vendors (Source: IoT Data Analytics Report 2016, Camrosh Ltd. & Ideya Ltd.)
The map shows the key role partnerships play in the ecosystem enabling vendors to create bespoke services through partnering with enablers in the ecosystem. Furthermore, IoT Data Analytics service buyers will be able to use such maps to think more strategically about selecting a vendor by taking the vendor’s position in the ecosystem and their partnerships into consideration. Partnerships not only open access to enabling technologies, but also increase access to different IoT use cases and the domain knowledge that comes with it.
In a recent report titled “Broken Links: why analytics investments have yet to pay off” a report written by the Economist’s Intelligence Unit and sponsored by ZS Associates, a sales and marketing consultancy, identifies a number of issues as possible reasons for the slow realisation of benefits from implementing data driven initiatives, despite the existence of a high level of investment or expression of commitment to investing in analytics in the near future. Reasons include immaturity of the analytics infrastructure within a company and resulting inability to deal with more complex and larger volumes of data, lack of seamless communication between business and analytics leaders in organisations to enable solution design and change management, and disproportionate cost over value decisions, rather than strategic decisions that lead to a fully integrated infrastructure that enables analytics.
A clear map and strategic view of the vendor ecosystem will enable decision makers to become aware of the “Unknown Knowns” when selecting IoT Data Analytics tools and services so they can address challenges such as buy-in from the technical and operational units within the business while keeping a strategic view on outcomes they envision from creating a data driven business.
Get in touch to find out how we can help you.