Companies climbing out of the economic crisis will look to IT to transform their business operations, notes
Focus notes that companies are starting to shift their focus from survival to revival. Much of that focus is on business intelligence (BI) which, despite some characteristics of a mature market, remains a source of relative growth and innovation.
In projecting the future of BI, it is useful to note the extent to which its evolution has tracked and leveraged the development of the computing platform itself, as happened in the past eras of interactive, personal, and networked computing.
To confirm this prediction, HP conducted a survey at leading BI and data warehousing conferences in 2009, asking respondents to identify their top initiatives. Respondents named data quality, advanced analytics, data governance, and MDM. The results of these surveys comprise a white paper that states that more decisions will be automated, and more data and its context will be considered. Customer service decisions cannot be made based only on a customer’s history or on the current transaction, but must include all relevant data. In a world used to real-life simulation, you cannot hope to develop a few possible scenarios and expect that any one of them will suffice when there are so many variables that influence the situation. BI systems must support these changing needs, notes the firm.
Further, the market will strive to expand the use of BI and apply analytics to many more applications in a number of ways.
1. Enterprisewide data integration: A good investment
Recently, leading organizations have begun to employ a coordinated, enterprise-wide approach to data integration, enabling cross-functional analysis and enterprise-wide performance management, and improving applications such as customer and risk management.
An example is BlueCross BlueShield (BCBS) of Kansas City, which brought IT and business people together to design and implement an enterprise data warehouse as a single information resource for the business. It provides consistent use of information, a single point of accountability, and alignment with business needs. This connected data resource has enabled a business transformation resulting in major improvements in both member health and internal operations, including:
• Exposing data previously unavailable to their physician community, benefiting 4,000–5,000 offices
• Improved aggregate wellness score of members from 85.1% to 85.7% in one year
• Disease management savings of $10 million ($2.5 million in administrative and $7.5 million in medical costs) in the first program year
2. Increased data and business intelligence program governance
Survey respondents confirmed governance as a key initiative; intent to invest within 12 months is listed as follows:
Formal governance 53%
MDM 52%
BI competency center 45%
Standard taxonomy 40%
3. The promise of semantic technologies
Commercial application of this technology is becoming more widespread. Semantic technologies are being used today to:
• Automate product reclassification
• Enable accurate and consistent diagnosis and treatment across a hospital management community
• Perform dependency analysis for managing and reconfiguring software assets
Recent innovation allowing structured queries over unstructured data is providing greater precision, speed of delivery, and reduction of information overload when analyzing content, vs. using enterprise search.
A good first step toward eventually leveraging semantic technology (and a valuable part of a data governance process) is the development of a taxonomy, or definition of business terms that is standard across an organization. According to the HP BI study, 30% of organizations have already implemented a standard taxonomy for defining business terms, and another 41% plan to implement one within 12 months.
4. Expanding use of advanced analytics
The pressure to use data, not only to make real-time decisions, but also to predict relevant business events is increasing dramatically. A common approach is to extract data from an enterprise data warehouse (EDW) into analytic data marts for advanced analysis. But adding another layer to the data architecture increases complexity and potentially reduces the speed of decisions.
To accommodate these conflicting pressures, technology providers have begun to push the advanced analytics computation closer to the data (similar to relational database management systems [DBMS] using stored procedures to execute procedural logic within the database). Analysts see in-database analytics as the heart of the predictive enterprise.
In the analytics process, computation can be pushed closer to the data when:
• Preparing data for model building
• Building the analytical model
• Deploying and executing models
5. Narrowing the gap between operational systems and the data warehouse
Traditional data warehouses often do not provide up-to-the-minute data or results of analysis that is timely enough for making operational decisions, notes Focus Research. Information is provided as reports, in a format for use by humans, not applications or processes. Where more active data warehousing has been attempted, it has been labor-intensive and the results are brittle and resistant to change.
6. Data warehousing and business intelligence: A new generation drives new priorities
The EDW can no longer be an isolated repository supporting stand-alone BI applications. But the real time and high availability requirements needed for its expanded use are not well supported by most existing platforms. In fact, in a 2009 HP-sponsored research survey conducted by The Data Warehousing Institute (TDWI), 46% of respondents said that they anticipate replacing their current primary data warehouse platform by 2012.
7. Growing impact and opportunity of Complex Event Processing
In the HP BI survey, 59% of respondents indicated that they use Complex Events Processing or plan to do so in the next 12 months. According to Gartner, “The market for commercial CEP products is expected to have a 31% CAGR from 2008 to 2013.”
CEP can analyze and correlate many base events to infer complex links and find patterns in data streams from the Web, e-mail, sensors, and so on.
8. Growing importance of integrating and analyzing unstructured/semi-structured data
Most organizations have content management systems to manage and search unstructured content, but have limited capabilities to use the information for decision making. Text analytics is challenging, and difficult to scale to large volumes of data. It is further complicated when attempting to integrate the analysis of structured and unstructured data.
At the same time, it is becoming increasingly important to be able to glean information from content generated not only inside the enterprise, but outside as well, from new sources beyond.
9. Social computing and the next frontier for business intelligence
The dynamic conversation channels available through blogs, online communities, Twitter, Facebook, LinkedIn, and a host of social computing venues engage customers, prospects, partners, influencers, and employees—touching virtually every key constituent in an organization’s value chain.
Today most organizations are only beginning to analyze the information garnered from online conversations. Technologies such as social mining and social intelligence use sophisticated data mining and text analytics to understand the implicit meaning of this unstructured data, which is completely reliant on the context in which it occurs. These include social behaviors, attitudes, relationships, and knowledge, all of which carry subjective qualities not easily categorized.
10. Growing interest in cloud computing for business intelligence
As the sophistication of BI environments has increased, so has their complexity and cost of management. This is not unique to BI, and organizations have already begun to adopt alternative delivery models to reduce the cost and complexity of other IT solutions.
They range from open source tools and embedded functionality to bundled tools to development and starter licenses. Most notably, research surveys by HP and TDWI indicate fast-growing interest in exploring the use of utility-based delivery models such as Software as a Service (SaaS) and cloud computing for BI.
To access this white paper, click
