Limited functionality available
Over 300,000 ‘things’ are being connected every hour, each of which is producing even more data – the Internet of Things is rapidly gaining momentum. The number of networked devices globally is almost double that of our global population and that number is expected to increase to over 50 billion by 2020.
As a result of the increase in the number of connected devices and the increase in bandwidth available, it was projected that by 2016 the amount of data generated that year would be more than all previous internet years combined – over 1 zettabyte. Organisations have the unique ability to generate, capture, access and analyse more data than ever. The following article will briefly outline the concept of big data, the shift away from causation to correlation and the top three implications to business models.
The term ‘big data’ has arisen from the exponential growth in the amount of data that now exists beyond that of traditional data storage. It has become a priority consideration for almost all businesses operating today with significant investments already underway in many organisations. The relevance of big data for business is not the data per se but the value that can be derived from the data. The key questions that should be asked about the value created should include:
For businesses, data analytics can enable faster, more intelligent decisions in turn leading to significant improvements in financial performance, strategic management and operational efficiency.
Big data represents a shift away from causation to correlation
The definition of big data usually includes the following themes:
Predictions and insights are at the heart of the value that big data can provide. Big data represents a shift away from causation to correlation – not why something is happening, but rather the fact that it is happening, the what. The key concepts within big data include:
1. It’s big – taking as many data samples as possible. We are no longer constrained by information processing and can now cost-effectively harness the full power of large data sets to improve the accuracy of predictions
2. It’s messy – using enormous volumes of data comes at a cost. Large data sets often contain erroneous figures and data inaccuracies. However, the ‘Ensemble Principle’ addresses this. Briefly, it states that ‘predictions can be improved by averaging the predictions of many’, thereby cancelling out the negative effects of messy data sets. More data trumps the quality or complexity of algorithms in creating prediction
3. It’s about correlation – correlation is one of the most important elements of big data theory. Correlations allow us to analyse phenomenon not by providing the answer itself, but determining the statistical probability that the answer selected is correct. With sufficient data, correlations surface more easily. A simple example of this is Walmart, who studied what customers bought, the time of day of the shop and the current weather. They noticed that just before hurricanes struck the sales of flashlights and pop-tarts increased significantly thereby demonstrating a correlation (not causation) between hurricanes and the purchase of these two products. As a result they were able to increase their inventory at these times and increase sales
Top 3 Changes to Business Models Enabled by Big Data
1. Differentiation through Data
Big data can provide new opportunities for differentiation. This has become possible due to the wide variety of new service offerings now available thanks to insights and prediction that big data can provide. For example, Google’s AdSense enables advertising to be delivered to users based on their search. Amazon automatically suggests similar books that may be of interest when searching for a book. Visa will target personalised location based offers based on point of sale.
2. Brokering Data
Access to the vast amounts of data presents new opportunities for new forms of information brokering. There are several options for brokering data, including selling it raw, providing insights and analysis, repackaging and/or anonymising the data. Indeed the substantial market capitalisation of many online companies such as Facebook, Google, LinkedIn, is derived predominantly from the value of the data they have collected. Sharing data though does not have to be all about commercial gain. Transport planning in some countries is optimised through data provided by Telecommunications operators’ mobile phone networks showing anonymised customer movements during peak hour.
3. Human Augmentation and Automation
Big data analysis is increasingly being used in business operations to enable quicker and smarter business decisions. For example, by placing sensors in farms that can accurately monitor soil and air moisture levels, farmers no longer need to “guess” when and where to irrigate. Retailers, no longer need to “guess” how many attendants to place at checkouts. They can now rely on real-time data of people in the store, correlate this with historical queue traffic and weekly cycles and in doing so reduce wait times for customers and more effectively utilise their workforce. Telecoms and Financial Services are becoming increasingly sophisticated with ‘next best conversation’ decision making using cognitive analytics to increase the relevancy of customer interactions.
However, without the right level of business engagement, big data is purely academic
Particular attention should always be made around the level of business engagement. This is an area that is often overlooked in the race to set-up new big data environments. New data insights may be interesting but they only add value when they are used to optimise existing business processes or through the creation of new revenue streams. Determining the high value business use cases for a new big data capability is a good place to start before deploying any new technology.
The relevance of big data lies in the insights and the foresight that the data can provide through correlation and analysis. Its value lies in its ability to augment human decision making and in some cases, automate manual processes or create whole new business models. By addressing some of the considerations outlined above and understanding how big data can turn your strategic aspirations into a reality, you can use this powerful capability to your competitive advantage.
“If we have data, let’s look at data. If all we have are opinions, let’s go with mine”
Jim Barksdale, CEO, Netscape
 Cisco IoT, http://blogs.cisco.com/news/50-billion-things-coming-to-a-cloud-near-you
 Cisco VNI Forecast and Methodology, 2014-2019
Simon is a Partner in Deloitte’s Data Risk and Compliance Analytics team and specialist data and automation leader. Simon has over 20 years’ experience both in operational and consulting data and analytics roles. Simon has an innovative style and is passionate about leveraging data insights and analytics to help businesses and risk functions to drive maximum from their data and to enable effective decision-making. Focus areas include; data and analytics strategy, digital risk, big data, robotic and cognitive process automation (RPA), cognitive engagement, CoE design and operating model development, information management, data governance and business intelligence.