9 Ways to Sharpen Your Business Intelligence
Nothing succeeds like success. In the beginning, the best way to justify investments and improvements in a BI system is to demonstrate its impact with clear business improvements. Look for areas of inefficiency that would benefit, quickly, from data-driven decision making. Show how the company can improve planning, tighten schedules and budgets, and boost productivity and profits.
In a time of economic turmoil, business intelligence (BI) initiatives stand out for their potential to improve corporate performance through better use of data gathering by disparate systems. Compared with many enterprise IT projects, BI requires relatively modest investments. One of the most important trends in information technology has been the emergence of a market for general-purpose business intelligence tools that can be aimed at any sort of business problem.
Investments in BI pay off the most for organizations that are serious about doing it right. They take the time to fight the political and technical battles needed to ensure the information they analyze is complete. They root out errors to make sure the analysis the tools produce is trustworthy. Most of all, they create custom views of the data for key constituencies, so that leaders from the CEO and CFO on down have immediate access to information on the trends, opportunities and threats they must address for better performance.
BI is as much about management discipline as technology. Following are nine ways companies can maximize the value of their investments in BI, according to the Gleansight benchmark report Business Intelligence (which can be downloaded for free).
1. Customize user interfaces/dashboards for specific roles. The CEO, the director of human resources and a call center manager should have different, at-a-glance summaries and graphs of the key performance indicators (KPIs) that are most relevant to their responsibilities. Top Performers are keenly interested in targeting BI applications to multiple constituencies within their organizations and identifying the metrics that will help them do their jobs better.
2. Integrate data across departments and applications. Particularly in large organizations, it's common to have several different BI tools used by different departments. Attempts to reconcile these disparities are often frustrated when the enterprise acquires another company that has adopted yet another BI toolset. In such cases, to expect absolute homogeneity may be unrealistic. Still, the more consistent the BI platform, the better. The more that data extracted from different systems can be reconciled and rolled up into common data warehouses and data marts, the easier it is to do cross-functional and cross-departmental analysis. The more that users can obtain all the data they need from a common portal or dashboard, the lower the training required before they can use the BI platform productively.
3. Foster a culture of data-driven decision making. As with any technology intended for broad deployment, it helps to have a push from the top that says, "This is how we're going to do business." Further, if the CEO and other top executives are firmly committed to managing according to the key performance indicators (KPIs) tracked by the BI system, anyone who wants to get a raise or a promotion will have every incentive to consult the BI system, too. Celebrate decisions supported by hard data. Discourage decisions that run contrary to the available data -- or that are simply made without consulting it. Acknowledge that sometimes decisions must be made in the absence of data or with data that is unreliable or ambiguous. Make it clear that using data to guide better business decisions is an organizational goal.
4. Implement processes for continuous data quality improvement. Data quality will never be perfect, but it can always be better. Errors creep in through manual data entry as well as automated glitches in merging data from different sources. Create a systematic process for identifying and eliminating the majority of these errors. Tactics include improving validation at the data-entry stage, cross-checking reference data such as customer addresses against public databases, and employing data cleansing software that looks for anomalies and contradictions.
5. Demonstrate improved planning, operations and other outcomes. Nothing succeeds like success. In the beginning, the best way to justify investments and improvements in a BI system is to demonstrate its impact with clear business improvements. Look for areas of inefficiency that would benefit, quickly, from data-driven decision making. Show how the company can improve planning, tighten schedules and budgets, and boost productivity and profits.
6. Implement a formal KPI methodology (e.g., Balanced Scorecard, Six Sigma, etc.). Top Performers show a strong commitment to adopting a proven methodology for setting and tracking performance metrics, according to Gleanster research. In the end, it's not how many charts and reports you can display that matters, it's what you do with the information. For better results, build on what others have learned the hard way.
7. Deploy alerts/notifications. Giving managers the ability to retrieve data is not enough, particularly if they are being tasked to respond quickly to threats and opportunities. Consider proactively providing an alert when a threshold is exceeded, say by a spike in supplier prices or demand for a given product. Such a notification system can be mapped to existing roles within the company. Additionally, individual users can be granted the ability to configure their own alerts.
8. Implement employee training. While BI tools uniformly promise to be easy to use, some employees will require training. Make sure to budget for this, recognizing that employee learning styles will dictate whether this training should be in person, online, instructor-led or self-guided.
9. Improve analytics capabilities. Beyond the basics of streamlining routine reports and answering common questions, one can maximize the value of a BI investment by adding more sophisticated analytical tools and nurturing better analytical talent. Uncover subtle, unexpected or counterintuitive patterns by applying data mining and predictive analytics. Improve the interpretation of the results by developing staff with the requisite knowledge and experience.