The Need for Easier and Lower Cost Business Intelligence
por Colin White
Originally published septiembre 19, 2007
Exploiting the benefits of business intelligence (BI) software has been one of the top priorities for organizations for many years, and growth projections for this area of IT indicate this is unlikely to change any time soon. Used correctly, business intelligence can bring significant business benefits to companies in terms of reduced costs, improved business efficiency, increased revenues, and better customer satisfaction. These benefits are the reason why business intelligence continues to enjoy success. On the other hand, it is quite apparent that companies are still not getting the most out of their BI investment – in most cases, BI is being used only by a fraction of the staff that could gain benefit from its use. One of the main reasons for the lack of penetration of business intelligence in organizations is the complexity and cost of deploying and using BI solutions. Solving these cost and complexity issues is likely to be a key focus area for both BI vendors and implementers over the next few years.
Constant Change is Not Always Good
Software vendors and developers are always adding new functionality and shipping new releases of their products. Keeping up with this constant change is becoming a major headache for IT organizations. In addition to providing customer requested enhancements, vendors are also motivated to keep upgrading their products to gain competitive advantage, and to maintain revenue through maintenance and upgrade fees. Even if an organization doesn’t need the capabilities offered in a new release, it is often forced to install it because maintenance for the old release is dropped.
To be fair to vendors, our industry seems to crave new features, and this constant need to add new product capabilities is driven by this craving. This, of course, is not unique to software. Technology and reference books published more than a few years ago are deemed out of date, even if the field in question hasn’t changed a great deal in that time. A good example here is birding field guides. There are always several new guides published each year, even though the number of bird species remains the same. Birders are always looking for the perfect guide!
Another example of the constant drive to add new function is the latest releases of Microsoft Windows and Office. These new releases dramatically increase the hardware resources required to run them and have significant learning curves. Are the upgrades really worth it in terms of money and time? In my case, I decided to try out the Apple Macintosh for first time, and I was really surprised that it was intuitive and easy to use the machine.
Many of my PC tools don’t work on the Mac (I can use VMware Fusion virtualization software to overcome this problem when needed, however), but, in many cases, equivalent features are included with the product or are available as open source software. Perhaps Mac enthusiasts are not as bigoted as I thought they were! It may also explain why many more Mac notebooks seem to be in use as I pass through airports. This is why Microsoft needs to develop a new and leaner version of Windows from the ground up, and drop support for legacy hardware and software, if it is to successfully compete against companies such as Google.
Disruptive BI Solutions Gain Traction
The reason for my Microsoft example is that at some point the upgrade and cost pains are so great that people begin to look for alternatives, even though those alternatives in the short term are disruptive. I believe we reaching that point with business intelligence, and existing vendors need to develop cheaper and leaner alternatives, or face market share erosion by new and more agile BI vendors with disruptive and open source solutions.
Disillusionment with existing BI software is already beginning, and some companies are starting to rebel against the constant drive toward new product features, and are instead looking for easier and cheaper approaches. These companies typically have less IT resources and skills to use for BI projects and may be struggling to implement even basic BI capabilities. Vendors often categorize such companies as small and medium businesses (SMBs), but in reality many are quite large, or may be individual business units of global 1000 companies.
Making BI Easier to Use
BI solutions are primarily used by the business analyst community and experienced data-centric business users who are using business intelligence for strategic and tactical business decision making. BI vendors realize, however, that the future growth lies in operational BI and deploying BI to a wider of set of less experienced users.
Supporting less experienced BI users requires simpler interfaces and more automation. This is why BI vendors are beginning to deliver web-based tools with improved visualization capabilities. Support for interfaces to Microsoft Office and enterprise portals is also an indication of the move toward improved ease of use.
The problem for existing BI vendors is that they still have monolithic and bloated architectures, and recent product acquisitions by many of these vendors makes this situation even worse. Exploiting light-weight Web 2.0 development approaches and technologies often forces vendors to develop new products from the ground up, and this is why existing vendors could be threatened by new BI software companies entering the market with modern product architectures and no legacy features to worry about.
Another way to add usability is to improve the way users interact and access data. To date, most BI tools access data using GUI-driven database queries. The move toward the use of search interfaces to BI, and the entry of search vendors into the BI space, is leading to a paradigm shift in the way data is accessed and delivered. Search interfaces will become a common component in all BI applications.
Another Option: Software-as-a-Service
Another direction of the BI industry is toward the software-as-a-service (SaaS) model. This area is gaining ground rapidly with many business intelligence and data integration vendors offering and developing SaaS options. Again, the question here is can existing vendors fit their existing architecture into such a model, or are they threatened by new SaaS vendors such as Host Analytics, LucidEra, Oco, and SeaTab? Initially, SaaS will be targeted at outsourced and SaaS operational applications such as SalesForce.com; but as the SaaS approach gains ground, these solutions will become more generalized.
Other developments that are likely to improve business intelligence ease of use are event-driven approaches such as embedded BI and automated decision engines, and the move toward the use of software appliances. It is crazy that customers should have to deal with installing, integrating and maintaining a wide range of system and BI software before they can develop applications. The appliances model today is focused mainly on offering packages of bundled hardware and software for improved price/performance, but the future of the appliance marketplace lies with application appliances that bundle all the required software components into a single solution that can run on a variety of hardware platforms.
The current BI vendor pricing model is broken. It simply doesn’t scale to large deployments and web-based service-driven approaches. Multimillion dollar deployments may be acceptable to very large corporations, but for an increasing number of companies, it is unacceptable.
Microsoft is already eroding the market share of the big BI players by bundling BI features into its core offerings. Okay, the Microsoft products may not have as much functionality as some of the other BI products; but for many customers, it is more than adequate.
Watch out also for the open source BI vendors to gain traction as customers struggle to reduce costs. Open source usage at present is dominated by system software, but there is increasing growth in the use of open source search and content management products, and the BI field will be next. Companies such as JasperSoft, Pentaho, and Talend already have offerings that meet the needs of many customers. Some of these “open source” solutions have commercial components, but often they have large developer communities that enable vendors to rapidly develop new features and provide good support to customers.
As already mentioned, the appliance approach is providing easier to use solutions at a higher price/performance. This part of the BI market is growing as companies begin to use the appliance approach for deploying BI applications that were simply too expensive to develop before because of the large amount of processing and data involved. As history shows, however, basing your marketing strategy solely on price/performance is a loser over the long term, and look for the appliance vendors to become more application and software focused.
Where Are We Heading?
There is no question that customers are looking for easier and less costly BI approaches, and this will open up the marketplace to new BI vendors, and to new and disruptive BI technologies. The issue for IT staff is that this could lead to fragmented applications, information silos, and integration headaches. This, of course, is nothing new since few companies have a single integrated BI architecture, and the new developments outlined in this article mean it’s business as usual – but hopefully at a lower cost!
Recent articles by Colin White
Copyright 2004 — 2020. Powell Media, LLC. All rights reserved.
BeyeNETWORK™ is a trademark of Powell Media, LLC