From Big Data to Insights
Big Data is revolutionizing analytics, databases and business
There is much buzz around Big Data these days, especially in the technical community, marketing, and also at the political level because the collection and analysis of data is increasingly becoming a strategic sector that will attract investments and create new jobs worldwide.
In marketing, Big Data promises to usher in a new phase of communication by focusing on the ability to send targeted messages to consumers of the modern era: connected consumers.
In the technology world, it promises the creation of new technologies and tools capable of analyzing large amounts of data in a very short period of time or even in real time.
It will also enable the creation of new specialized professions in the analysis of these data. An example is already evident with the birth of a new role: “Data Scientist” that represents the evolution of the “Data Analyst”. While the vast amount of data can tell you many things, you need to understand what they are saying.
This is where the Data Scientist comes in — like a modern day knowledge prospector — and does the mundane job of collecting and preparing unruly digital data, panning for insights, until finally spotting the golden nuggets of information up for inspection and interpretation.
What is Big Data?
There is no precise and exhaustive definition but one that comes close is that Big Data represents the mass of transactional data generated by all users and devices connected to the Internet worldwide.
Thanks to various sources such as social media, these data represent the behavior of users, what they are looking for, what they think of a product, what they think of politics, their emotional state, their geographic location, what they buy and their own interests.
Big Data can be characterized using the now common three Vs: Volume, Velocity and Variety that well represent the different aspects.
Volume
There is a large amount of heterogeneous data that can be analyzed in different formats like text, images and video. Having a lot of data to be analyzed is certainly one of the greatest technological challenges, but the amount allows, thanks to new mathematical models, to obtain information with a higher degree of accuracy in some contexts and to obtain predictive models.Velocity
The volume of data and the ability to analyze them is the real challenge. The rate of growth in the volume of stored data is exponential. Conversely the waiting time to get answers from the analysis is constant. In fact today we are already talking about realtime analysis.Variety
The data that can be collected today are abundant and very different from each other. First, they come from a myriad of sources and have been increasing exponentially since the birth of social media and objects connected to the network (Internet of Things). These data are represented in different formats, such as text, video, images, audio and in different kind of repositories such as files, spreadsheets and databases.
While termed “Big” Data, the size of the data is not the most critical factor. What is more important is how the data is managed to extrapolate valuable, meaningful and actionable insights. While we have talked the 3Vs (Volume, Velocity, Variety), through the business perspective, Value and Validity of data should not be overlooked as well. Big data should be viewed as a holistic solution that integrates various tools, technologies and processes to manage existing data sets in a more time and cost effective manner.
The companies that will have access to Big Data and will be able to use the new tools available will be able to analyze not only the trends in a specific sector but will have detailed information about a single consumer.
Through a variety of sources, it is possible to aggregate data for a single consumer and have more precise information about his or her habits, orientations, interests and preferences about the products and brands. In this way, companies will be able to provide promotional messages targeted to the individuals so that it is really relevant.
This means being able to engage each individual and establish a relationship that can entice the consumer to a greater interaction with the company. This capability will dramatically change the role of marketers in the short term.
For example, by using existing technologies, such as NFC: Near Field Communication, it is possible to connect the activities carried out in the physical world like walking into a movie theater or shopping in a supermarket with the digital world by sending real-time promotions and highly customized initiatives.
These devices are in fact able to communicate without any settings when they are in a closer proximity to a transceiver.
Big Data for Healthcare Market
The term Big Data as we have seen is fashionable. The possible applications are not limited to marketing but will transform the health market.
Even now we can see many products capable of monitoring our physical activities or foods and publish this information in order to be processed and provide many new services. For example, you can compare food data with calories burned and your habits in order to suggest a specific physical activity or a personalized diet.
Through connected care, by comparing the considerable mass of data on the individual (health medical records, DNA, habits) with the data on the population in a given geographical area, this will aid in better and earlier diagnosis and treatment of the patient. For example, Pfizer leveraged a large EMR database of patient data, coupled with information from medical literature, to create a model to help clinicians identify patients that might be suffering from fibromyalgia earlier so that they can receive effective treatment.
Through real world data, pharmacos will also be able to better and quickly identify medicines that will be safe and effective and also determine which patients will benefit the most from them. In this fashion, patients will be able to enjoy a more precise care pathway that is customized based on their genetic, dietary and lifestyle factors.
Through data patterns, pharmacos will also be able to identify the most important aspects of a disease to study in clinical trials and by selecting the precise profile of patients to include in their studies, this may very well result in shorter clinical trials and faster delivery of treatment options.
These improvements will also signify more informed investments by a single health facility or the Public Administration in order to focus on truly effective treatments and thus enable a better quality of life and at the same time reduce costs.
From Data to Insights
While naysayers are claiming that Big Data was overhyped from the very beginning, and indeed, it is one of those technologies that is making its descent down the trough of disillusionment in the Gartner Hype Cycle, interest in Big Data remains undiminished. It has simply evolved beyond the peak because the market has matured with a set of approaches, technologies and tools. The earlier hype has merely shifted from managing big data to actually applying information and data to making informed business decisions. Compared to the theoretical discussions of the early years, today it is more about getting business value from big data, creating a big data strategy and exploiting analytics to optimize productivity.
This market will experience increasing growth even in the creative sector where some leading organizations are already beginning to produce contents, services and products based on the analysis of user behavior.
The technological heavyweights like Google, LinkedIn, Twitter and Facebook base most of their business on the sale of data and analysis tools. By today this sales model is open for every type of business and market, such as Telecommunications, Energy and Health, which will begin to base their business on the data and make them available to clients for a fee.
Thanks to more powerful and easier-to-use visualization tools, the analysis of Big Data will be presented in a simple and intuitive way and not only technicians will be able to read the data. This will increase the ability of companies to identify insights and allow them to come up with evidence-based strategies.
Technology behind Big Data
It is clear that the technology challenge is a key factor in the management of Big Data. Many companies are working to establish the enabling technologies and thus be leaders in this area.
Take for example two well-known technologies which are working on Big Data: Hadoop and Cassandra.
Hadoop
It is a platform for distributed computational problems on many servers. Initially developed by Yahoo and released as open source, today it implements Google technologies such as MapReduce. To store data, Hadoop uses its own distributed file system (HDFS) which always makes the data available on multiple computing nodes to be analyzed. Hadoop, coupled with highly scalable cloud computing can offer great benefits to the healthcare industry, especially in terms of predictive analytics, population health and modeling.
Cassandra
It is a distributed database based on the concept of NoSQL and it is released as open source. Born in Facebook laboratories to analyze data streams in a shortest possible time, such as the Inbox email, the most important feature of Cassandra is the ability to store large amounts of data in a distributed manner and the ability of grow to infinite scalability, capable of processing large amounts of data in a short time.
Planning for the Big Data transformation journey
How do we plan for and implement Data 2.0?
Here are some key factors that the teams tasked with this transformation journey need to consider:
Internal support and collaboration: Obtaining solely executive level support and rolling this out via a top-down approach is not going to work. There is the need to educate and showcase the different benefits of technology to the different business units. When engaging with the different teams, speak their lingo or in terminology that resonates with their functional areas and responsibilities. Only upon getting their buy in will the program be effectively utilized.
Phased approach for sustainability: Define the technology roadmap and drive its alignment with the business roadmap. Trying to do too much too soon will backfire. Start with something small, in phases, and of course measureable so that tangible benefits to business can be shown. The key is a modus operandi that delivers real benefits and is sustainable.
Strong information governance: Beyond the technology, new processes need to be created and people trained. Without strong governance, there will be the risk of getting partial, incomplete and inconsistent data sets.
External learnings: Benchmark and learn from competitors, partners and across industries in order to leverage innovation and best practices for your own journey.
Continual optimization: Establish success criteria and define KPIs from the start. Measure them along the way in order to build from mistakes and strengthen achievements. When embarking on a path of such a transformation journey, continuous improvement is key. Along the way, if the original strategy or approach is not proving to be effective or sustainable, do not be afraid to re-evaluate and re-route to ensure constant alignment with overall business objectives.
In conclusion
Data will no longer be a commodity available to a mere department or a group of managers. In 2015, a phase of democratization of data will begin in all departments of the company, which will result in more insights from different sources. It will evolve into a single centralized strategy in all activities that via the generation of different data sets will lead to new insights every day.
