Getting Intelligent About Business Intelligence
There are, however, key components that BI platforms share and there are consensual definitions in the industry as to what constitutes BI. More than just a buzzword or passing technology fad, BI is an essential part of a business' overall strategy for success. BI is often associated with bloated, complicated and expensive enterprise technology platforms, yet there are plenty of BI vendors out there offering solutions targeted at the SMB market.
Here's a look at some general definitions of BI, the controversies surrounding it, the future of BI and, finally, the leading vendors in this oft-misunderstood technology space.
Business Intelligence: Sometimes Hard To Define
Trying to pin down which features exactly make up a BI solution can be as enigmatic as trying to define which factors make for a true cloud-based solution.
The phrase "business intelligence" is thought to be a term coined by IBM researcher Hans Peter Luhn, back in the late 1950s. Luhn described BI as the ability to comprehend the interrelationships of presented facts in such a way as to guide action toward a desired goal. In 1989, a Gartner researcher, Howard Drenser, defined BI as a set of concepts and methods to improve business decision-making by using fact-based support systems.
A more broad definition of BI: the analysis of a company's collection of data in order to achieve some goal or greater operational insight, such as an increase in revenue, comparison to a competitor or to achieve cost-cutting.
Companies can collect data using a wide range of software products--from Excel to expansive database applications that are designed for vertical markets like health care. Because collecting data can come from a large variety of sources, BI software helps to gather data into a single repository and sort it for analysis.
Currently the technology industry has pretty much agreed on the basic factors about what makes for a BI solution:
• The ability to extract data from disparate sources such as CRM, ERP systems and other applications.
• The ability to centralize, organize and standardize information in repositories.
• The inclusion of analytical tools that allow businesses to perform reporting and queries to uncover business patterns, trends and issues.
• The ability to visualize that data in some end-result format such as charts, graphs, or in other forms.
One thing is clear: Business intelligence is a critical component of an organization's day-to-day operations.
BI Controversy
BI is often thought of as a vague concept or as an umbrella term that is used more as a marketing ploy than a description of a specific technology. There is most certainly no shortage of vendors that attempt to peddle their software offerings as "business intelligence solutions" when, in actuality, their products often consist of no more than basic data-collecting, numbers-crunching and some reporting features. One differentiator between BI and your run-of-the-mill business analysis software, according to some BI tech analysts, is the ability of BI to not only do data analysis but to also do data visualization. There is a really interesting debate about this very subject on Stephen Few's blog. Few is founder and principal of Perceptual Edge, a "consultancy that was established to help organizations like yours to design simple information displays for effective analysis and communication."
The meat of the debate on Few's blog post, "Data Visualization and Analysis--BI's Blind Spot," centers on whether the visual representation of data; i.e., in the form of charts and other eye-grabbing graphic forms, is a key feature of BI solutions.
Here is Few's blistering review of one software company's business intelligence product:
"'l'll begin by admitting that this new business intelligence vendor's name isn't really "Lazysoft", but it's close. I took the liberty of transposing two of the letters in the name "Lyzasoft" to create a name that describes a fundamental problem with its software--it's the product of laziness. Some software engineers no doubt had lots of fun developing this product, but Lyzasoft approached the task in the typically lazy manner of many software companies--they didn't bother engaging the services of designers who actually understand data visualization or data analysis."
He lowers the hammer with this statement:
"I will say, however, that [Lyzasoft] fails miserably in its ability to help people explore, visualize and then present data."
To be fair, reviewers here have not had the opportunity to review Lyzasoft's product, and are not agreeing or disagreeing with Few's assessment. However, Few does make strong points about the debate surrounding what BI is and what it is not.
Few apparently takes issue with what many critics of BI software take issue with--if the product doesn't present end data in a useful manner, and ultimately provide solid data visualization, then it shouldn't be called business intelligence.
Few was challenged by a colleague, Colin White, founder and president of BI Research. White wrote to Few asking him to outline the differences between data analysis and data visualization. From White's perspective, if data is presented in any way that aids a business in decision-making, then that is indeed data visualization. Few does not agree:
"I certainly agree that the goal is to achieve understanding and support decision making, but not every way of doing this is data visualization, and not everything that would like to call itself data visualization deserves the name. Information can be presented in various ways, just as it can be verbally communicated in various languages; each medium of data presentation (the spoken word, the written word, and visual representations of various types) has its strengths and weaknesses, its appropriate applications, and its rules for effective use. Saying that every presentation that aids decision making is data visualization is not a useful definition."
It's easy to get bogged down into semantic minutiae of what data visualization is or isn't, ergo, what business intelligence is or is not. The fact is, though, that Few makes a wonderful point in his blog about how opportunistic vendors will claim a product as BI when in fact their solutions are centered on either collecting, cleansing, aggregating, and/or storing data. All are necessary and important technologies in their own right, yet each of these areas are more about the processing and warehousing of data and less about the analysis of the data and the presentation of that analysis. BI 2.0
The recession has forced companies to re-evaluate, reorganize and do some heavy-duty insight into operations and financials. These are all contributing factors to what has been described in the industry as a renewed interest in business intelligence. In a 2009 study by IBM titled "The New Voice of the CIO," BI was top of mind for many of the more than 2,500 CIOs surveyed worldwide. When asked, "What kinds of visionary plans do you have for enhanced competitiveness?" 83 percent of CIOs listed business intelligence and analytics.
One of most pervasive myths surrounding BI is that it is super-complex, super-expensive and really only for large enterprises. However, BI continues to see adoption at the SMB level. This is related to the current economic climate and the trend toward services moving into the cloud. VARs that focus on the SMB market now have a very solid value proposition with BI as a SaaS solution. Instead of having to implement complex, BI platforms on-premise for customers, BI in the cloud gives an affordable, remotely manageable way of deploying BI at the SMB level.
Advances in virtualization also provide an affordable way for the SMB to deploy what may otherwise be a daunting investment in the equipment needed for a business intelligence platform.
Another factor increasing BI at the SMB level is open source. There are a number of open-source BI platforms out there, another value prop for solution providers. VARs can deploy open-source BI solutions, taking advantage of the licensing structure of open source and providing post-install service, support and customization for customers. Two of the leading open-source BI vendors are JasperSoft and Pentaho.
The Vendors
Business intelligence is a vast arena with a lot of players in the field. There are many who are quick to call their solutions BI, yet not all of them deliver what the industry has recognized to be true, soup-to-nuts business intelligence software and tools. Here's a look at two vendors leading the pack in the BI market:
JasperSoft
San Francisco-based JasperSoft is one of the most successful open-source business intelligence vendors in the game today. The all-inclusive JasperSoft Business Intelligence Suite provides several components that make it top-of-line for BI. The product incorporates many facets of Web 2.0 including AJAX and DHTML.
JasperServer is a stand-alone and embeddable reporting server. It is based on open-source technology that includes XML, JDBC, Hibernate, Java and Web Services. A key feature is the ability of end users to create ad hoc, powerful reports without needing the skills and time of a developer--a potential cost-savings for a company that may have limited IT staff or skill sets. Users can build reports and queries all through a Web browser via dragging and dropping business view items into the query builder, adding filters and parameter prompts--and the query is complete. Reports can be saved, shared, scheduled and distributed via JasperServer's repository manager application. Using the same methods, users can build and deploy dashboards that monitor key performance indicators and metrics in near-realtime. Also, any URL-addressable content can be included in dashboards, enabling mash-ups across internal and external business applications.
JasperServer also provides rich visualization. Customizable chart themes give users granular control over the way in which charts are presented.
JasperAnalysis provides robust, Web-based online analytical processing (OLAP). With JasperAnalysis, users can do all of the housekeeping associated with BI: spot trends and patterns, identify anomalies and give visual representation to correlations. Microsoft's Excel can be used as a front-end application to JasperAnalysis.
JasperETL is a tool that pushes the JasperSoft Business Intelligence Suite a bit further. It can aggregate massive amounts of data from a variety of sources, scale BI to include datamarts and warehouses and can aid in performance by offloading query and analysis from operational systems.
Being open source, JasperSoft's BI suite has a number of add-ons to extend its feature set. JasperSoft ODBO Connect is one of the add-ons. It connects Microsoft Excel 2003 and 2007 Pivot Tables to JasperAnalysis and other OLAP servers. Users can analyze data from a central store with the familiarity of the Excel User Interface. The add-on is included with JasperAnalysis Professional Edition. It is priced at $250 per user for Standard Support and $375 per user for Premium Support.
Another useful add-on is Column Manager. This is an add-on for JasperReports. This allows developers to show or hide columns or groups of columns based on a user's input, something that is really handy when looking at a large amount of data on-screen. Column Manager is available for $99.
JasperReports has another add-on, the HTML Template Exporter. This add-on allows for richer formatting of reports that are exported as HTML files. It preserves free-form page layout designs and adds support for report interactivity such as sort, hide/show sections (also called folding), and display pop-ups. The price for this add-on is $199.
JasperSoft also integrates with the Amazon EC2 cloud infrastructure to deliver cloud-based BI. All that is required to use this service is to sign up for Amazon EC2 and RightScale, which is cloud management accounts. Once signed in, users can select the JasperServer template from RightScale, deploying business intelligence reports to the cloud via JasperServer.
JasperSoft has a partner network that includes ISVs and technology partners.
Cognos
IBM Cognos is one of the most, if not the most, renowned business intelligence platform. More than 23,000 companies and organizations worldwide use Cognos for BI needs.
IBM Cognos 8 Business Intelligence delivers the core components of BI: reporting, analysis, dashboards and scorecards on a single platform. It is large-scale, feature-rich BI solution targeted at the enterprise. However, IBM also offers IBM Cognos Express, which is integrated BI and planning for midsize companies. It delivers all of the essentials of a BI solution at an affordable cost.
Cognos Express has another cost-saving benefit as it integrates into a midsize business' infrastructure and is easy to deploy. A single, centralized Web-based management console gives administrators control over the BI environment in an interface that is easy to navigate through, intuitive to work with and pleasing to the eye.
Cognos Express Reporter meets a business' reporting needs by providing self-service reporting and ad hoc querying. It features intuitive drag-and-drop capabilities for users to easily create reports. Users can perform these tasks all from a Web browser. Reports can be automatically published to the Web as PDFs, as Excel spreadsheets, to e-mail, or to the IBM Cognos Express Web portal.
Collaboration is a big feature of Express Reporter. Users can maintain and upgrade a report using a shared interface that is accessible companywide.
Cognos Express Advisor is the component of Cognos express that does the data analysis and visualization heavy-lifting. In-memory analytics process realtime calculations. Another feature includes freeform analysis, the ability to take any combination of data and build models and custom cubes and dimensions for deeper data analysis. Express Advisor also gives business' powerful "what-if" scenario capabilities, enabling the planning of business strategy based on future actions.
Data visualization options are rich and graphics-intense. Visualizations include bar, line, area, pie, radar, bubble, scatter and range charts and graphs, plus geo maps and heat/performance maps.
With Cognos Express' Xcelerator, businesses can extend the functionality of Microsoft Excel.
Some of the features of Xcelerator include centralized management of data, business hierarchies, rules and calculations to eliminate data silos and inconsistencies throughout the organization, streamlined and continuous planning, aggregation and analysis and write-back capabilities that link "what-if" modeling results with business optimization and planning.
Xcelerator also caters to nonfinance users. The Web-based interface allows these users to contribute to and view current plans, analysis and reports. There are also in-the-box blueprints for key planning tasks such as expense and capital planning, workforce management and profitability analysis.
For organizations that may have to adhere to compliancy regulations, Xcelerator features role-based security, structured workflows and clear audit trails to improve accountability and accuracy.
IBM's PartnerWorld offers solution providers a program that is centered around Cognos products called the IBM Software ValueNet. IBM partners must be approved for this program and some of the requirements include membership in IBM PartnerWorld, an approved business plan, certified technical resources, existing customer references, available sales demonstration, proven sales lead generation capability and trained resources for selling and deploying solutions. Although the requirement may seem stringent, partners are offered a host of benefits upon entry into the program. Benefits include protecting partner investments through controlled distribution of selected products in the IBM Passport Advantage portfolio, fixed price discounting and working with experienced software value-add distributors.