Imagine a future as depicted in the Tom Cruise thriller Minority Report, in which police arrest future criminals before they commit the crime, based on predictive foreknowledge. That future is not so far-fetched, as companies such as Amazon work to ship packages to you before you order them, hospitals and doctors treat you based on predictions of your future health, employers assess job candidates based on future predictions,  and, yes, even law enforcement prevents crime based on forecasts of future criminal offenses. More than ever, companies are utilizing new and evolving forward-thinking technology to gain valuable insight into their business and ultimately steer their future actions and decisions. A rapidly growing trend, businesses are ever-increasingly using predictive analytics technology to anticipate future business needs, behaviors, and outcomes.
In today’s business world, predictive analytics refers to statistical tools and methods used to analyze data and make predictions about the future, based on patterns, relationships, and trends found in the data. Broken down to its basic level, predictive analytics involves a data model being created, with historical and current data gathered and fed into the model. The model goes through a learning phase where it uses programmed algorithms to search for, identify, and classify various patterns and relationships, as defined by the creator of the model. The model is often tested with different sets of data until it yields consistent and reliable results. The model is then implemented, making forecasts of future events based on predictive rules it has created from the analyzed data. As more data is added to the model and as the model becomes more adept at recognizing patterns and trends, its predictions become more accurate and useful.
Figure 1: Basic Process of Predictive Analytics
(Data Collection > Model Testing > Model Implementation > Decisions Made from Results)
Already used in various capacities in industries such as financial services, marketing, insurance, retail, healthcare, manufacturing, law enforcement, government, travel, and pharmaceuticals, the use of predictive analytics and its wide range of application is rapidly increasing, with one survey finding such use in businesses tripled in a three-year period from 2009 to 2012. This trend will only increase with time. In fact, one research firm predicted that by 2016, 70% of high-performing, profitable companies will manage their business processes using real-time predictive analytics. As predictive analytics becomes more commonplace in business, even the common consumer is increasingly experiencing its use and effects, as evidenced by targeted and individualized internet advertising based on predictions of future spending. Mick Hollison of InsideSales.com states:
Whether you love it or hate it, predictive analytics has already helped elect presidents, discover new energy sources, score consumer credit, assess health risks, detect fraud and target prospective buyers. It is here to stay, and technology advances ranging from faster hardware to software that analyzes increasingly vast quantities of data are making the use of predictive analytics more creative and efficient than ever before.
PREDICTIVE ANALYTICS AND INFORMATION GOVERNANCE
As we gain a better understanding of the uses and function of predictive analytics, the versatility and potentially enormous value it brings to not only a wide swath of industries, but within the common general corporate structure, quickly becomes clear. With this in mind, how might predictive analytics be leveraged to benefit a company’s Information Governance (IG) and data management practices? Below is a quick overview of key areas in which predictive analytics might benefit an IG program, particularly with respect to areas of risk management, data management efficiency/cost effectiveness, and program synergy:
1. RISK MANAGEMENT
The common business environment is rife with risks and complexities, ranging from internal company risks (such as employee safety and health issues, employee conduct, data management, and business operational practices), to external risks (including natural, economic, and political disasters), and strategic and business-related risks.
An effective implementation of predictive analytics within an IG framework can help mitigate risks commonly associated with a company’s data and information practices and make way for preventive, forward-looking solutions and actions. A few examples in which predictive analytics could prove valuable:
- Litigation and associated discovery risks: A data model is created, analyzing and classifying past lawsuits and legal issues, types/subject matter of litigation, records most commonly affected by/related to/used in/subpoenaed in the litigation, and legal outcomes. Using this data, the predictive model could help classify or rank records based on the level of critical legal need or importance, shedding light on data and records of future legal concern or importance. This would give management a more informed view of what data to keep (or dispose of), how long to keep it, and how to be better prepared for future potential legal issues, with respect to company information and data kept.
- Data security: Various information is analyzed, such as:
- Types of data stored or kept
- Identification of sensitive or private data and past/current management of such data
- Data access (type of data accessed, frequency of access, mode of access, parties accessing the data)
- Past data breaches (and attempts)
- Records affected by breaches
- Various data security measures taken for record types
- Departmental data management protocols and practices, etc.
A predictive model could identify current data security issues to be addressed, as well as predict future areas of concern based on its analysis. This would allow management to more effectively and efficiently address its data security, targeting specific records and IT systems predicted to be at particular risk. Regarding security, it is always better to act with foresight rather than hindsight.
- General operational risks: Information is gathered respecting different departments, records kept, varying departmental recordkeeping practices and procedures, etc. The predictive model could analyze such data and shed light on future concerns dealing with a range of issues This would allow the company to better target and address company-wide and specific departmental concerns that may currently or in the future impede its continued growth and success.
2. EFFICIENCY/COST EFFECTIVENESS
As with all things in a business, data management costs money, even up to a substantial five percent of business revenue, with poor data practices potentially costing a company 15 to 25 percent of its operating budget. With so at stake financially, it is imperative that a company revisit its IG practices and records retention policies to find ways to best streamline and increase the effectiveness of its data management, cutting costs while maximizing the value of retained data. Using predictive analytics is one way a company can proactively tackle the management of its data, helping to increase the efficiency, cost-effectiveness, access to, and value of the records and data it manages. Areas in which predictive analytics might prove useful to this end include:
- Streamlined Data Management/Cost Effectiveness: Record access data is inputted in a predictive model, which analyzes patterns of how often certain information/records is requested or accessed. Based on the patterns observed, the model predicts which types of records or classes of data will be most used and accessed, and when they will be accessed. For records not mandated to be retained to meet legal obligations, this insight can aid the company in shaping its record retention policies based on the predicted frequency of future records accessed (i.e. records never used or those that are accessed less often than others might have a shorter retention period). This ultimately increases the efficiency and relevancy of records kept, with only necessary data being kept and unnecessary or low-risk records being timely disposed of. Over time, efficient data management will result in cost savings to the company. Less data to manage means less costly data management tools and storage systems/methods to deal with.
- Efficiency of access: A predictive model looks at the frequency of records accessed and how they are accessed. Using this data, the model forecasts the needed availability of different types of records. Management can use this to more effectively organize the framework of the data management structure, allowing for more efficient search and identification of particular records, and more efficient and intuitive methods of access (as regarding to the location of the records, electronic access, etc.). This increases the value of information stored, as critical records can be accessed and utilized more quickly and directly when needed.
3. SYNERGY/EFFECTIVENESS OF AN IG PROGRAM
For an IG system to best achieve its intended objective of holistic information management and to truly benefit the company, it must operate as a strategic and fluid company-wide accountability framework, initiated and proponed at the executive level. As a company’s IG officer or committee considers how to best implement an IG framework and policies into the company culture, predictive analytics can prove useful.
- A predictive model analyzes various recordkeeping-related data of the different departments within a company (types of records kept, number of records kept, storage practices, etc.), and determines patterns of behavior, consistencies, inconsistencies, and potential shortcomings or issues within each department’s data management practices. As a first step, this gives management a snapshot of the shape of the company’s IG practices, both on a granular level and as a whole. Management uses this insight to pinpoint areas (and/or departments) of concern or needed improvement, and accordingly implements appropriate IG initiatives, training, or other programs to address and remedy identified individual and company-wide needs. As these needs are addressed at the enterprise/c-suite level, the company as a whole benefits from a more cohesive, united, and effective IG program, with more effective and purposefully crafted policies, procedures, and practices.
While the general premise of predictive analytics (analyzing existing data to predict future business needs and outcomes) has been in use for a long time, it is currently experiencing a period of unprecedented growth in usage, application, and technological advances. Leveraging predictive analytics may prove beneficial in lowering risks, cutting costs, and increasing efficiency and effectiveness of company data kept, ultimately assisting the company to shape and implement an effective IG practice. As a company assesses its IG program, it should consider how predictive analytics and other emerging business technologies might help to achieve its IG objectives. Thoughtful and strategic implementation of such technology in concert with an effectively-administered IG program can help position the company for sustained success with the management of its information and data.
 IMDB, Minority Report, http://www.imdb.com/title/tt0181689/ (last visited Dec. 4, 2015).
 Lance Ulanoff , Amazon Knows What you Want Before you Buy it (January 21, 2014), at http://mashable.com/2014/01/21/amazon-anticipatory-shipping-patent/.
 Linda A. Winters-Miner, PhD, Seven ways predictive analytics can improve healthcare (October 2014), at http://www.elsevier.com/connect/seven-ways-predictive-analytics-can-improve-healthcare (noting that “[i]n medicine, predictions can range from responses to medications to hospital readmission rates. Examples are predicting infections from methods of suturing, determining the likelihood of disease, helping a physician with a diagnosis, and even predicting future wellness.”).
 Information Management, 5 Predictive Analytics Use Cases, http://www.information-management.com/gallery/5-predictive-analytics-use-cases-10023630-1.html (last visited Dec. 4, 2015) (explaining how an analytics company helped U.S. Special Forces develop data models to assess new candidates based on a predictive framework).
 Heather Kelly, Police Embracing Tech That Predicts Crimes, CNN, May 26, 2014, at http://www.cnn.com/2012/07/09/tech/innovation/police-tech/ (describing crime-fighting data software used by the LAPD: “The program is called PredPol, and it calculates its forecasts based on times and locations of previous crimes, combined with sociological information about criminal behavior and patterns. The technology has been beta tested in the Santa Cruz, California police department for the past year, and in an L.A. police precinct for the past six months, with promising results.”).
 Gartner, Gartner Says by 2016, 70 Percent of the Most Profitable Companies Will Manage Their Business Processes Using Real-Time Predictive Analytics or Extreme Collaboration, February 26, 2013, http://www.gartner.com/newsroom/id/2349215.
 See Anasse Bari, Mohamed Chaouchi & Tommy Jung, Basics of Predictive Analytics Data-Classifications Process, at http://www.dummies.com/how-to/content/basics-of-predictive-analytics-dataclassifications.html (last visited Dec. 4, 2015) (giving a brief overview of how a predictive data model works).
 See e.g. Eric Siegel, How Predictive Analytics is Reinventing Industry (March 19. 2015), at http://data-informed.com/how-predictive-analytics-is-reinventing-industry/; Who uses predictive analytics?, FICO.com, at http://www.fico.com/en/predictive-analytics/understanding-predictive-analytics/who-uses-predictive-analytics (briefly explaining various industries that utilize predictive analytics). See also Rabe, Mark, How Predictive Analytics is Reshaping the Travel Industry, February 20, 2015, http://insights.wired.com/profiles/blogs/how-predictive-analytics-is-reshaping-the-travel-industry#axzz3YteCiZvd.
 See Accenture, Analytics in Action: Breakthroughs and Barriers on the Journey to ROI, http://www.accenture.com/SiteCollectionDocuments/PDF/Accenture-Analytics-In-Action-Survey.pdf, In a survey of 600 director-level executives and managers of “enterprise-level companies” (1,000 or more employees), the use of PA tripled from 12% to 33% by 2012.
 Gartner, commenting that “organizations with real-time analytic and decision management capabilities perform better. Improved situational awareness leads to better and faster decision making and superior customer service, revenue growth, cost reduction and risk avoidance … virtually every business operation has one or more areas where real-time analytic services or active analytics should be applied.”
 Pew Research Center, March 9, 2012, Search Engine Use 2012, (finding that 59% of internet users surveyed in 2012 had noticed targeted advertising)
 NPR, Smart Cookies Put Targeted Online Ads on the Rise, Laura Sydell, October 5, 2010, http://www.npr.org/templates/story/story.php?storyId=130349989 . (Describing various consumer data gathered by online sites to determine targeted advertising, including “[y]our gender, your age, your income, whether you have children. Psychographic information like whether or not you consider yourself to be an introvert or an extrovert or liberal or conservative.”).
 Hollison, Mick, Love or Hate It, Why Predictive Analytics Is The Next Big Thing, September 17, 2014, accessed at http://www.inc.com/mick-hollison/love-or-hate-it-why-predictive-analytics-is-the-next-big-thing.html.
 Robert S. Kaplan and Anette Mikes, Managing Risks: A New Framework, Harvard Business Review, June 2012 Issue, found at https://hbr.org/2012/06/managing-risks-a-new-framework, (differentiating between, and giving examples of common types of business-associated risks).
 Sperling, Ed, The High Cost of Managing Data, 10/11/2010, Forbes, accessed at http://www.forbes.com/2010/10/08/legal-security-requirements-technology-data-maintenance.html , (noting that “[w]hen it comes to business, time is money, but so is data. It costs big bucks to manage and store data, and the more data that inundates corporations each year, the more it costs them.”)
 Value of Data Management, http://www.usgs.gov/datamanagement/why-dm/value.php, last accessed 4/10/2015, citing TechTarget, Data Quality Trends, with Expert Larry English, http://searchdatamanagement.techtarget.com/podcast/Data-quality-trends-with-expert-Larry-English
 See McCollum, Sam, A Roadmap for Effective Information Governance, Information Management (Arma International), January/February 2013, p. 27, accessed at http://content.arma.org/IMM/online/Archives/2013.aspx (describing the basic aims and goals of an information governance program). See also Best Practices, InfoGov Basics, accessed at http://www.infogovbasics.com/best-practices/ (explaining the role executive-level personnel play in an effective information governance program).
 See generally Tadd Wodd, History of Predictive Analytics: Since 1689, August 12, 2013, canworksmart.com.