FlashPoint Newsletter

View online: tdwi.org/flashpoint


		December 2, 2010	ANNOUNCEMENTS New Checklist Report: Data Integration for Real-Time Data Warehousing and Data Virtualization CONTENTS Getting More out of BI: Why Can’t Business and IT Play Nice Together? Drilling Down the Data Mine: Why Requirements Should Drive BI Development What Is a Spreadmart? Mistake: Inadequate Knowledge Management See what's current in TDWI Education, Research, Webinars, and Marketplace
	Getting More out of BI: Why Can’t Business and IT Play Nice Together? Adrian Alleyne DecisionPath Consulting
Topic: Business Intelligence Are your business people and business intelligence (BI) teams working well together? Does one set understand the needs of the other? According to DecisionPath Consulting’s recent research [PDF download], this isn’t always the case. In this report, we explore how companies are using business analytics and business performance management applications, and we examine perspectives of both IT and business professionals (52 percent and 48 percent of respondents, respectively). Although our findings aren’t entirely new to BI professionals, they validate the different perspectives of business users and the BI team. Here are three key findings of the report: Finding #1: At a company level, there’s a widespread lack of maturity. Although BI adoption is widespread, it often becomes stuck at lower levels of BI maturity. For example, standard reports are used instead of mission-critical applications to keep executives’ fingers on the pulse of their day-to-day job functions. We asked our business respondents what terms they use to describe the analysis of business information to support better decision making and improve business results. Although, as expected, the terms “business intelligence” and “analytics” were frequent responses (64 percent for each), the term “reporting” was the most frequently selected option (82 percent). We also asked respondents to describe how they currently use BI; only 36 percent said they viewed business intelligence as a widely useable strategic or mission-critical management system. When asked about future use, more respondents envisioned BI being used strategically to monitor performance (55 percent) or for competitive advantage (64 percent). In short, although most companies recognize the potential of business analytics to be strategic or mission-critical, they haven’t quite focused or invested appropriately to implement these more sophisticated and value-adding systems. *Finding #2: At a business-function level, there’s a disconnect between BI need* and current BI use. Although BI is being used in most job functions and roles, there are important gaps in how effectively it is being used to solve day-to-day challenges. Business users want to leverage more sophisticated business analytics, but they struggle to articulate a more sophisticated solution. Here are some key areas where business users felt their business intelligence was inadequate: Finance: Using information to better connect operational performance with financial results Operations: Using information to translate company strategy into operational plans and align performance with company objectives Sales and marketing: Using information to understand the competitive landscape, acquire and retain customers, and optimize sales and distribution channels Finding #3: At an IT/BI management level, there’s a vicious cycle. The IT/BI team is deluged with requests for information, and this constant demand makes it difficult for the team to develop higher-impact applications with higher ROI. The perceived lack of responsiveness by business users causes them to overlook the BI program as a mission-critical tool. The three biggest challenges for BI teams: the business value of BI isn’t being measured (73 percent) and there is inadequate time (64 percent) and funding (55 percent) for high-impact BI programs. Given limited time, money, and management buy-in, it’s crucial that organizations formally plan their BI initiatives, identify opportunities that weigh the potential value against the risk of pursuit, and develop a portfolio of BI projects that deliver the highest return on investment. In short, both business users and BI managers must understand that utilizing information can help their companies better manage and monitor their performance, and provide them with business analytics tools that can transform their companies into analytic competitors. The problem for most companies, though, is the inability to connect their understanding of BI into a concrete program that involves the right information, technology, management buy-in, and applications. Many companies aren’t there yet, but more are moving in this direction. A Snapshot Our research provided a good snapshot of where companies stand on this issue, and we want to continue to explore it. If you’d like to weigh in on the subject, please take part in our survey. We’ll update our findings as more participants respond. Adrian Alleyne is director of market research for DecisionPath Consulting, which specializes in business intelligence, data warehousing, and performance management solutions. Contact Adrian at 301.990.1204. Drilling Down the Data Mine: Why Requirements Should Drive BI Development Mary Anne Hopper Baseline Consulting Topic:** Business Requirements Like millions of others, I watched as the first Chilean miner was pulled to safety and then followed the story as the remaining miners made their journeys back to the surface. That got me wondering what it must be like to dig deep for something and then get stuck, or what it is like to figure out a solution to help those who are trapped. I actually see these very situations quite often with data. I work with clients who have data elements that get carried from source system environments to reporting environments and nobody can account for how, why, or even if they are being used. The reason is predictable: a business person made a request and the person responsible for fulfilling it decided to chase additional data. I like to refer to this as the “data mine” approach: drilling into a source system and pulling out everything that is available without regard to its usability or value. What impact does this have on the project delivery process and to the support and maintenance beyond the end of a project? It can be significant, especially when you consider all the pieces of the data puzzle. Simply sourcing data regardless of its usefulness can waste precious time and effort at every step of the process. Requirements can be held up because data elements are being defined--elements that no one understands--all because a programmer created a vaguely named “marketing_indicator” flag. Add more time for modeling and profiling activities and your project might already be behind schedule. Now the data needs to be provisioned and stored, and don’t forget about ongoing maintenance to support questions and monitor source system changes that might impact any number of data elements. The idea of adding one data element certainly won’t make or break a project (or the bottom line), but where is the threshold? Is it two data elements? Ten? How about 465? I think the more important question is: Why are we spending time and effort on data that might not provide business value? Remember, there are business requirements and there are data requirements. Data requirements should support business requirements. If your data warehouse team is blindly sourcing data, it’s costing more money than you should be paying, so why continue? I have some theories to share, and my guess is that you’ve experienced one or more (if not all) of these situations: The business analyst took what he or she thought was the easiest and safest way to support requirements--loading all the data. The business analyst assumed the business will eventually ask for everything, so why not just save time by doing the work now? The business analyst decided the business doesn’t really know what it wants or needs. The developers made a decision that bringing in an entire table took the same effort as bringing in a few elements. The new data seems close enough in subject matter to the original request that it makes sense to let it tag along. Business users expressed requirements in the “give it all to me and I’ll figure out what I need later” mode, and developers obliged. Do these situations sound familiar? If so, take a few minutes to reflect on the impact. As a business analyst, perhaps you had to define an entirely new set of data. As a data modeler, perhaps the impact was having to fit this data into the model with little knowledge of how it might be used. As a developer, perhaps you had to build (or rebuild) the ETL process. As a report developer, the project might have required effort to build something that will never be used. As the project manager, perhaps you had to find funding for, and manage, an ever-growing project. As a business owner, you might have been delayed, waiting and wondering why it was taking so long to get your “simple” request fulfilled. The solution is as simple as following a disciplined requirements process where data doesn’t move without an articulated business need. If the business makes a request based on its need, business requirements are drafted and data is provisioned to support that need, and access to that data (in whatever format--self-service, canned report, dashboard, etc.) is developed, period. This solution may sound overly simplistic, but there is a good deal of rigor and process necessary to follow this approach. I recommend that you spend the necessary effort. Put the process in place that allows your business to define what it wants, then work to deliver that need--and only that need. You might feel a bit more pain up front, but in the end you will have delivered business value more quickly instead of getting trapped in the mine. Mary Anne Hopper is a senior consultant at Baseline Consulting and a frequent blogger on Baseline’s Web site. See www.baseline-consulting.com/blogs for more of her work. Highlight of key findings from TDWI's wide variety of research What Is a Spreadmart? In organizations all over the world, business people bypass their IT groups to get data from spreadmarts. Spreadmarts are data shadow systems in which individuals collect and massage data on an ongoing basis to support their information requirements or those of their immediate workgroup. These shadow systems, which are usually built on spreadsheets, exist outside of approved, IT-managed corporate data repositories, such as data warehouses, data marts, or ERP systems, and contain data and logic that often conflict with corporate data. Once created, these systems spread throughout an organization like pernicious vines, strangling any chance for information consistency and reliability. You’ll find them in all industries, supporting all business functions. According to TDWI Research, more than 90% of all organizations have spreadmarts. (See Figure 1.) (Click to enlarge) Source: Strategies for Managing Spreadmarts: Migrating to a Managed BI Environment (TDWI Best Practices Report, Q1 2008). Click here to access the report. FlashPoint Rx prescribes a "Mistake to Avoid" for business intelligence and data warehousing professionals. Mistake: Inadequate Knowledge Management By Karen Degner “The measurement of success is not whether you have a tough problem to deal with, but whether it’s the same problem you had last year.” --J . F. Dulles BPM is a closed-loop system of communication, alignment, and continuous improvement. While an invaluable tool, it is still the quality of people’s actions that will make or break the execution of strategy. Opportunities for improvement are seriously undermined when best practices, processes, and documentation are primarily on the laptops, desktops, and in the heads of subject matter experts. In the absence of a knowledge management (KM) program, this knowledge is available only to those who know whom to ask, rarely gets leveraged, and will eventually walk out the door. If we only knew what we know. Organizations need KM programs to harness and leverage knowledge throughout the organization. Core to this program is a quality KM repository with easy access for all. Business and IT should partner to leverage content management and collaborative enabling technologies. This will also facilitate regulatory compliance, reduce risk, and lower cost. Many organizations form professional communities around individual processes, procedures, technologies, and operations. People around the organization submit content and recommendations to the respective community teams for qualification and identification as a reuse or best practice. Content has a life, and the professional community should “evergreen” the content and the process. As with all programs, KM programs and activities need to be continuously prioritized, measured, monitored, and managed against organizational strategy and goals. Source: Ten Mistakes to Avoid Implementing Business Performance Management (Q1 2007). Click here to access the publication.