11 Guiding Principles for a Successful Business Intelligence Implementation

Tibco Spotfire has one of the most prolific corporate blog in the BI area. Updated nearly every day, the blog, entitled Trends and Outliers, is a great source of news and information about Business Intelligence (Full disclosure: I already wrote guest posts for the TIBCO Spotfire's Business Intelligence Blog). Last week, they published a great post on Successful BI Implementation, where they listed 11 guiding principles for a successful business intelligence implementation from Booz & Co. Below is the list:


1. Drive Change From The Top Down and The Bottom Up
Like any new solution, business intelligence is only effective if people use it. For it to be completely adopted, it needs to be used not only by line managers but also by executives.

2. Create a Comprehensive Definition of Business Intelligence
Business intelligence puts emphasis on measuring performance against goals and establishing accountability for reaching those goals. Make sure you have the supporting processes, systems and change management protocols in place to support this new way of running the business.

3. Use an Agile, Modular Approach
You can achieve more flexible and more effective implementations of business intelligence faster with agile development and by focusing on specific areas that are guided by an integrated, overall strategy.

4. Focus On The Right Metrics
Metrics must be aligned with the company’s strategy and capabilities, including both internal and external inputs, and consisting of both leading and lagging indicators.

5. Keep It Simple
Even though technology might let you drill down 16 levels into the data or slice and dice it 100 different ways, that kind of analysis may be irrelevant and distracting. Be selective by including a few key metrics that are the most important and drill down to the top three or four levels.

6. Build a Unified BI System
Some of the most important insight from a BI system can come from discovering how interdependencies impact outcomes across an organization. Integrate data and data analytics across the organization to allow for custom analytics that can identify root causes of issues.

7. Launch Early
To gain acceptance and support you may need some early wins with a BI project. Start with high-priority areas that have high-quality metrics. Demonstrate the value of a BI solution to help build momentum for the project.

8. Create Detailed System Requirements and Select The Right Partners
Successful BI implementations require a partnership between IT and the lines of business. They also require strong project management skills, systems integration know-how and software tools. Make sure you’ve included and selected the right teams across all these areas.

9. Leverage Existing Infrastructure
Business Intelligence implementations should align with a company’s IT strategy and vision for how IT will support the business. BI should complement existing IT capability which can often be achieved by leveraging current IT infrastructure to provide the back end, and using business intelligence solutions to provide the front end.

10. Establish a Centralized Governance Structure
A business intelligence implementation can touch every area of an enterprise. It requires cooperation and shared ownership from the business and IT, new data management protocols, strong project management and ongoing analysis of the metrics used. For maximum success, you should create an overall governance structure led by the business and supported by IT.

11. Proactively Manage Change
Introducing business intelligence requires extensive change management. When evaluating performance against metrics, there should be clear accountability, consequences for not meeting goals and incentives for exceeding them. As with any initiative that requires change management, a successful BI implementation requires senior leadership support, training and communication throughout the enterprise.