Remember the movie, "What Women Want"? The main character, a man, had to figure out an advertising strategy for a bunch of women's products. He developed the ability to hear a woman's thoughts. Suddenly, he understood what motivated them, what made them "tick" - and he was credited with great insight into the female psyche.

Corporations have spent billions of dollars on customer relationship management (CRM) and business intelligence (BI) technologies thinking that this expenditure would give them the same type of insight and the oh-so-elusive competitive edge over their rivals. How much success have they had to date truly understanding their customers' needs? Well, the jury is still out, but if customer satisfaction scores are any indication, there has been only a slight uptick in overall satisfaction. It appears that many companies still don't "get" what their customers want.

The same situation is true within the organization. We, in IT, struggle to understand what our "customers" (the business communities) want in and from their BI environments. And we struggle to understand what we need as well. Could we read our internal customers' minds like Mel Gibson did in the movie? Probably not, but I suggest that you are a good source of answers to these questions. After all, you are a customer too, right? You shop. You compare products. You evaluate features and functions. You examine options. What are you looking for? Use your own situation to determine what makes internal customers "tick."

Here is what I want as a customer. I look for two things in products and services - simplicity and reliability. Granted, expectations for simplicity and reliability have gone way up since my grandparents' time, but the basic needs are the same. Can I easily use the product or service, and does it perform unfailingly?

Why would we think that a BI environment should be any different? Simplicity and reliability are the two features we want in a BI environment as well - from both the view of the implementers as well as the business communities using it.

We'll start with simplicity. Whether you are implementing a BI environment or using it to make better decisions, simplicity takes on a significant role. For implementations, technologists want simplicity in:

Designs: Data models, schemas and data acquisition processes (ETL - extract, transform and load; data hygiene; and so on) should be simple to create and understand. Highly complex designs are usually more brittle and less likely to be suitable for a broader audience.

Technology: End-to-end technological solutions are gaining momentum because they do eliminate much of the complexity in interfaces and integration. Suites of BI technologies allow seamless flows of data and better documentation, which in turn leads to simpler environments.

Maintenance: Because BI environments are constantly changing due to changing business situations, the implementers must be able to "turn on a dime" - satisfying new enhancements to existing components, replacing aging functionality with newer technology or applications, or adding new functionality. To do these tasks, you must simplify the overall environment and processes, thus streamlining any maintenance needs. This also leads to reduced overall costs.

For business users who access the environment, simplicity comes in the form of:

Usage: The old saw of "easy to use" is a big part of a successful BI environment. Yes, a simple interface between the business community and the underlying technology is mandatory these days. The business communities must be able to quickly and simply address their BI problems by using the BI environment with minimal IT interference.

Understanding: A BI environment is considered simple if business users can easily decide how they can get the appropriate answers to their business questions. Are the documentation and meta data simple to access and comprehend?

Flexibility: Your BI environment will be successful if it is easy to perform a variety of analyses simply and easily. Data mining, multidimensional analysis, ad hoc queries - a business user should be able to seamlessly move from one form of analysis to another. Even within one form of these analyses, the user must be able to alter queries and run new or different analyses by simply choosing different parameters or selecting/deselecting columns or rows of data.

Now let's look at reliability from the same two viewpoints. For implementers, reliability is demonstrated in the:

Technologies used: There are many moving parts in terms of the technologies used for a BI environment - ETL, data quality, data delivery, databases, platforms, networks, storage devices, etc. A reliable environment is one with predictable performance and scalability. We have certainly come a long way from the early days of BI where we had to bolt a variety of technologies together and hope that they performed as advertised. Monitors, performance boosters and even predictive performance tools help us understand and maintain more reliable environments.

Processes created: Today, we have solid methodologies to use in implementing and maintaining BI environments. Even more useful today are the built-in capabilities in our ETL, data quality and even analysis tools. For example, ETL tools now come with packaged programs for specific ERP packages, data quality tools have algorithms for name and address hygiene and analytical tools come with embedded algorithms, KPIs (key performance indicators) and other industry- or regulatory-specific calculations and reports. These types of "canned" capabilities greatly enhance the reliability of the overall technological underpinnings of a BI environment.

Communications channels: Technolo-gists must have reliable answers and communications with their business users. We have made strides in ensuring that business users contribute their business knowledge and savvy to the IT implementers, but this area is still weak for many implementers. Shoring up the relationships between the business and IT still remains problematic and continues to be a source of unreliability in the overall implementations of BI.

From the business users' point of view, reliability is demonstrated in these three areas:

Availability: Today's BI environments are becoming more and more mission critical. They must now be accessible 24 hours a day, every day. Business communities are pressuring the IT folks into service level agreements requiring that the BI applications be completely reliable - to the same degree as the operational systems.

Consistent quality: Reliability in business terms comes in the form of the data or information that is produced, the performance or response times coming from all the technologies and, finally, in the meaning of the analytical results being presented. This last point may require a bit of clarification. What is meant here is that calculations, summarizations and derivations used in analytical results are constant and unchanging. Should any of these change due to a change in the ETL or data delivery processes, or to a change in the embedded algorithms in the tools, the business users must receive immediate notification of the change. Reliability does not mean non-changing; rather, it means that each person using the environment knows what he or she is getting at the end of a query. This directly leads into the last point on reliability.

Understanding the business user's context: Mature BI environments should be self-healing and self-annealing as the patterns of usage from the business communities change. This means that meta data - especially business meta data - must take a place of prominence in the overall architecture. To create a reliable environment of our users, we must face up to the fact that meta data is a critical component that goes a long way toward giving the business communities a reliable environment. Secondly, we must come to grips with the "right-time" data needs of the users. This does not mean making everything real-time. It does mean getting the business community the right answers by using the right data at the right time.

Now you have some answers to what your BI "customers" want. Generically, it's what we all want in our lives. The job now is to ensure that the BI environment supports simplicity and reliability for the folks implementing it and for those using it for decision making.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access