Can we get personal for a minute? Picture your closets at home. Do they house a (now dusty) bread maker, food processor or an espresso machine? Is your basement crammed with unused sports equipment, exercise gadgets or home improvement tools? Each item probably entered your household with the promise of impacting your life in dramatic ways. Perhaps they did ­ for a week or two. Eventually, they lost their luster and were shelved in favor of the next market fad.

Is your corporate business intelligence system at risk of following a similar life cycle? Many data warehouse/DSS projects effectively address the technical aspects (database, data model, OLAP tools, meta data, etc.) but have limited focus on the key component to success ­ building acceptance within the user community.

So what can you do to encourage your knowledge workers to adopt their business intelligence systems and tools for the long term? Let's look at some of the critical success factors.

Involve Key Users

The data warehouse/DSS project development team must have respected, dedicated representation from the user community. These subject matter experts (SMEs) must identify the objectives and priorities for the system, key success factors and the metrics to be gathered which will measure the impact on the organization. SMEs must develop the communication and roll-out plan for the entire user community. It is critical that the right people be selected for this team. The credibility of the entire project will be reflective of these representatives, so don't skimp ­ free up key knowledge workers to participate in this critical role.

Match the Tool to the User

Have you evaluated the user base and segmented it appropriately? Just as you tailor product strategy, positioning and messages for each customer segment, you also evaluate the needs and skills of your user base and categorize it for use in selecting data access tools. For example, consider factors such as need for local/remote/off-line access, frequency of querying detail data vs. aggregates, existing skill sets, etc. Don't assume one software tool will meet everyone's needs. Consider the specific applicability of OLAP/ROLAP tools, multidimensional "cubes," report generators and Web-based access options. Then put the right tools into the right hands.

Investigate executive information system (EIS) interfaces to encourage quick, easy access for senior managers. Once these key managers begin using the DSS tools to look at business results, it will drive everyone to follow.

Skill Development

For each of the user segments identified, develop a matrix of skills necessary to effectively utilize the selected business intelligence (BI) tools. Be ruthless in matching the right features/functions to the right users. Does every user really need to know sophisticated ad hoc query techniques on the first day? Some users should start with basic DSS reporting and drill downs. Provide more advanced training later as they gain confidence in their skills.

Develop an integrated business intelligence skills curriculum. Don't assume you can just train on the specific DSS software tools. Consider including training for complementary skills such as a thorough review of the data relationships and restrictions, manipulating data once exported out of the DSS tool, basic troubleshooting techniques or effectively interpreting the output from data mining applications. Correlate all training to actual business questions and use real-life exercises where possible.

Integrate Business Intelligence

Prepare a change management plan for phasing out the "old" tools and procedures. This may include a business process reengineering phase to identify the areas of your business that will be impacted by the new BI tools. Plan to update departmental procedures, job descriptions, business rules and interdepartmental communication guidelines. One common mistake is that we let the organization hang on to the "old," familiar ways, which ultimately results in your DSS system becoming a redundant resource in the company and certainly will result in increased costs rather than reduced information generation/distribution costs.

Be sure your business intelligence system is accessible and appealing. Consider using your corporate intranet site as a launch point for DSS applications. Link up a bulletin board or knowledge repository Web page to enable users to post tips and tricks, as well as share success stories of how they gleaned valuable business insight from the data warehouse.

Set up key queries to be automatically refreshed and displayed whenever data is updated in the warehouse. Prepare and broadcast "teaser reports" that entice users to investigate further. Utilize exception reporting (alerts) to focus on anomalies and outliers in the data rather than forcing continual scans through huge volumes of detail data. Making the accessibility, speed and business insight highly visible to the user community will encourage system utilization.

Redeploy IT resources (who formerly coded reports from the OLTP systems) toward performing more sophisticated analyses against the data warehouse, as well as investigating new access and data mining tools for the knowledge workers. For example, in the pharmaceutical industry, sales and marketing analysts are constantly evaluating the mix of physicians that the sales reps should call on. Through basic OLAP queries, these analysts might look at changes in prescribing habits using a 12-month trend line graph. However, through the use of data mining/neural networks, it is possible to identify just those prescribers statistically most likely to adopt newly launched products based on past behavior. That type of information clearly provides highly valuable, actionable direction which could significantly impact a new drug launch.

Assure Proper Administration

Queries that run too long on a slow, over-taxed network will turn users off. Continue to monitor how users are actually accessing the data. Many DSS administration tools enable sophisticated analysis of which tables are being scanned, how often and for what types of reports. An assessment of this data can result in improved aggregation and indexing strategies and, thus, optimized performance on the queries the knowledge workers actually use.

Providing direct access to a complete, integrated repository of your company's operational, sales, marketing and financial data can transform your company's capabilities for garnering insight into the business. However, it is necessary to provide appropriate structure and user support to increase the probability of long-term success. Building this business intelligence system should be a process not a product ­ a journey not a destination. Be sure to allot resources each year for new user training, refresher courses and advanced training. Continue with a stream of enhancements by integrating new data sources, additional historical data and new tools. By considering these tips, you can certainly buffer your investment from the possibility of falling into the technology graveyard.

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