For big data, 2012 was either the year of recognition as the future of computing OR the year of overhyping. Companies such as Salesforce.com, Adobe and Oracle all make major investments and announcements regarding their “marketing cloud” and how their solutions (are they new?) use big data. Given the broad disparity of opinions over the “ascent” of big data, what can we expect in 2013?
Prediction: There is going to be even more data to interpret – publicly available data will continue to grow rapidly during 2013 creating a tsunami of information for businesses to interpret – 50% of the businesses will make use of external data (crawling web, mining social networks) to make business decisions and drive marketing programs.
The rate of data creation is accelerating as it becomes simpler for businesses and consumers to create, publish, curate and share. In 2013, we expect that the amount of data on the Internet by December 2013 will be close to twice as much as it is in January 2013. This is going to create unprecedented stress on businesses to invest in people, technology and training to capture, analyze and act on this volume of data across devices and platforms – most of which is noise.
Mary Meeker expects that the amount of content and data on the Internet will double by 2014.
The [BDO] survey also revealed that the vast majority of managers are swimming in data, with 93% saying they find it a challenge to integrate and manage all their data, and 40% describing it as “very challenging.” Winners will be those who identify ways to efficiently take advantage of big data, likely using a new class of solutions – big data applications. Losers will be those who don’t harness this data-driven opportunity.
Prediction: The initial winners in mobile ecommerce drive their success with insights they continuously extract from big data.
In 2013, more people will access ecommerce businesses via their smartphone than via their desktop (Mary Meeker, KPCB Internet Trends). Their tolerance for a slow, mismatched experience is near zero and their ability to go elsewhere is near 100%. Businesses who believe they “own” the customer will be displaced by businesses who deliver what the customer wants – either to complete a task or to fill-time. As Forrester’s Thomas Husson correctly points out – mobile will be more disruptive than the web. Only through continuous and rapid analysis, can mobile sites predict and adapt to the intentions of their visitors. Data is what you have from your own analytics systems; big data is what you realize by harnessing the collective intelligence of the Internet.
Prediction: Enterprises will aggressively adapt big data applications because building their own applications to collect, harness and act on the data is not practical.
Build versus buy is an age-old debate. In the realm of big data in 2013, the velocity of change combined with the volume of data will force enterprises to assess their core competencies around building highly specialized applications or commissioning custom specialized applications. Big data applications learn from the experiences and behaviors of consumers and businesses inside and outside the domain of the enterprise. To discern the appropriate signal from a growing cacophony of noise requires expertise in machine learning, language processing, web-scale systems and data science. All four of these specialties have significant talent shortages – a shortage that will not be filled for years. It takes 6-8 years to educate a qualified data scientist who can build applications to interpret, adapt and act based on big data.
Prediction – IT will redefine its processes to reflect more scrum models in order to stay involved with the rapidly changing online business requirements.
Amazon.com tests many different features and products every month – reputed to be over 200. How many businesses are so aligned with their technologists that they can experiment this quickly? Big data provides insights – insights that must be tested continuously to determine what’s impactful. Smart businesses know that many of the tests will not succeed – so they cycle through tests quickly. The old IT model of scheduling large scale projects, building and moving on is incompatible with this new world. Instead, IT will embrace leveraged implementations that allow business people to conduct many tests without requiring continuous IT resources. This started years ago with the advent of Content Management Systems and eCommerce Platforms. The rate of change has accelerated, the data available to drive insights has exploded and the next generation of rapid adaptability is imminent for the technology that supports online and mobile business.
Hyped or anointed, big data is a permanent fixture of the digital future. Simply talking or fretting about it won’t impact a business’s future – but determining what insights can drive better business performance and finding continuously adapting, purpose-built big data applications that deliver those insights at scale will create powerful competitive advantages.
About the author
Ashutosh is the Chief Technology Officer (CTO) of BloomReach and a true guru of all things search, with 10 years of information retrieval, machine learning and search experience. Previously, he was a Staff Scientist at Google for 4+ years, which spanned 8 product launches. Prior to that, he was at IBM research. He is also a prolific publisher/inventor, with a book on machine learning, 30+ papers, and 50+ patents. Ashutosh holds a BTech from IIT-Delhi and a PhD from U of Illinois UC. Ashutosh has numerous awards, including best thesis award at IIT Delhi, IBM Fellowship and outstanding researcher award at UIUC.