Organizations are closely watching emerging technology trends to discover the next great competitive advantage in the use of information. One trend is easy to identify: more information. Data volumes are growing across the board, with organizations seeking to tap new sources generated by social media and online customer behavior. This trend is spurring tremendous interest in better access and analysis of the variety of information available in unstructured or semi-structured content sources.
Data Discovery Accelerates Self-Service BI and Analytics
From a macro perspective, it’s easy to identify the biggest long-term trend in business intelligence: providing nontechnical users with the tools and capabilities to access, analyze, and share data on their own. However, the road to this destination has not been easy. With IT driving application development and deployment, standard approaches to extending enterprise BI and data analysis capabilities have been difficult and slow. Getting the requirements right for the data, reports, visualization, and drill-down analysis capabilities is difficult and never fully satisfactory. By the time requirements have been gathered and turned into application features, users will have identified different requirements.
2. Unified Access and Analysis of All Types of Information Improves User Productivity
As the implementation of BI and analytics tools spreads to more users within organizations, a question inevitably arises: What about all the information in text and document formats, which accounts for the vast majority of what users encounter? Difficulty in finding information, whether structured or unstructured, is a productivity cost to organizations. If one of the measures of BI’s value is improved productivity, then BI should help users access and analyze unstructured as well as structured information.
Historically, BI systems have developed in technology ecosystems limited to structured, alphanumeric data, leaving unstructured content to document and content management systems, search engines, and a lot of manual paperwork. With the majority of content increasingly being stored and generated in digital form, users are demanding better integration between content access and analysis and the structured realm of BI. Integrated views of all types of information can help managers and frontline workers see the context surrounding the numbers in structured systems. This enables them to uncover business opportunities and find the root causes of problems more quickly.
3. Big Data Generated by Social Media Drives Innovation in Customer Analytics
Customer data intelligence has long been a major driver behind growth in the implementation of sophisticated analytics for prediction and pattern recognition as well as advanced data warehousing. In the brick-and-mortar days, organizations wanted to slice, dice, and mine transaction data and interpret it against demographic information. Advanced organizations sought to mine the data to uncover buying patterns and product affinities. As e-commerce and call centers proliferated, organizations needed to expand customer analysis to include interaction information recorded in all channels, bringing more terabytes into their data warehouses.
Now, with Twitter, Facebook, and other sites, we have hit the social media age: customers are using social networks to influence others and express their shopping interests and experiences. Organizations are hungry to capture and analyze activity by current and potential customers in social networks and comment fields across the Internet marketplace.
4. Text Analytics Enables Organizations to Interpret Social Media Sentiment Trends and Commentary
Rising interest in social media analysis is putting the spotlight on text analytics, which is the critical technology for understanding “sentiment” in social media, as well as customer reviews and other content sources. Like data mining, the text mining and analytics category stretches to include a range of techniques and software, such as natural language processing, relationship extraction, visualization, and predictive analysis.
Text analytics falls within the realm of interpretation rather than exact science, which makes it a nice complement to BI and structured data analytics. Sentiment analysis, for example, employs statistical and linguistic text analysis methods to understand positive and negative comments. While this analysis can provide an early sense of the reception of a new product or service, the interpretation cannot replace the more exacting analysis of the numbers done with BI or structured analytics tools. Sentiment analysis, however, can help organizations become more proactive in taking steps to address negative reactions to products and services before they lead to the poor sales that BI and data warehouse users detect later in the reporting and analysis of sales transaction figures.
5. Decision Management Enables Organizations to be Predictive and Proactive in Real Time
Trailblazing organizations in many industries are applying automated information technology to dramatically reduce, if not eliminate, delays in how they respond to customer interactions, adjust to changes in supply chains, prevent fraudulent activity, and more. The goal is to operate in as close to real time as possible. Along with automation, organizations are striving to use information analysis to become predictive and proactive. The objective is to develop predictive models and forecast behavior patterns so that organizations can anticipate certain events; then, they can orchestrate processes so that they can be proactive and fully prepared when predicted events or patterns occur.
When limited to a reactive posture, organizations face delays and confusion in how to respond to events, which can lead to increased costs and missed opportunities. Reactive organizations lack a well-orchestrated plan and can only respond to events on a case-by-case basis. With speed and complexity rising in many industries, a reactive posture isn’t good enough. Organizations need business intelligence and analytics applications and services that will help them shift from a reactive to a proactive and predictive posture. Traditional BI systems are not enough for organizations to make this shift.
Decision management is the term industry experts and vendors use to describe the integration of analytics with business rules and process management systems to achieve a predictive and proactive posture in a real-time world. Decision management requires several technologies. Business rules, or conditional statements for guiding decision processes, are common in application code and logic; the challenge is to implement business rules systems that can guide decisions across applications and processes, not just within one system. Business process management systems help organizations optimize processes that cross applications and use analytics as part of the continuous improvement of those processes.
Along with business rules and business process management, a third technology important to decision management is complex (or business) event processing. Events are happening everywhere; they are recorded or “sensed” from online behavior, RFID tags, manufacturing systems, surveillance, financial services trading, and so on. Integrated with analytics and data visualization, event processing systems can enable organizations to pick out meaningful events from a stream or “cloud” of noise that is not important.
Organizations can use decision management technologies to automate decisions where speed and complexity overwhelm human-centered decision processes, and where there are competitive advantages to having decisions executed in real time and driven by predictive models. Decision management is an emerging technology area currently focused on specialized systems, but as demand for greater execution speed and efficiency grows, more organizations will evaluate its potential for mainstream requirements.
The Shape of Things to Come
Picking just five trends is not easy, given that we are in an exciting phase of innovation—particularly regarding the access and analysis of big data, including social media content and data for new forms of investigation, such as geospatial analysis. In addition, the trends are unfolding as the infrastructure of computing is changing dramatically to include cloud platforms and the vast, worldwide adoption of mobile devices. So, while I did not identify either cloud computing or mobile adoption among the trends in this article, these platform shifts should be kept in mind as context for how the trends are likely to play out.