“We can’t solve problems by using the same kind of thinking we used when we created them.” – The famous quote attributed to Albert Einstein applies as much to Business Intelligence & Analytics as it does to other things. Many organizations that turn to BI&A for help on strategic business concerns such as increasing customer churn, drop in quality levels, missed revenue opportunities face disappointment. One of the important reasons for this is that the data that can provide such insights is just not there. For example, to understand the poor sales performance in a particular region during a year, it will not just help to have data about our sales plan, activities, opportunities, conversions and sales achieved / missed, it will also require understanding of other disruptive forces such as competitors promotions, change in customer preferences, new entrants or alternatives.
Thomas Davenport, a household name in the Business Intelligence & Analytics community, in his book ‘Analytics at Work’, explains the analytical DELTA (Data, Enterprise, Leadership, Targets and Analysts), a framework that organizations could adopt to implement analytics effectively for better business decisions and results. He emphasizes that besides the necessity of having clean, integrated and enterprise-wide data in a warehouse, it is also important that the data enables to measure something new and important.
Now, measuring something new and important cannot just be arbitrary. It requires be in line with the organizational strategy so that this measurement will have an impact on strategic decision-making. A proactive approach to Data Warehousing must then include such measurements and identify the necessary datasets that enable the measurement. For instance, an important element of a company’s strategy to keep its cost down could be to standardize on a selected few suppliers. To identify the right suppliers and make this consolidation work, it is important to analyze procurement history, which under normal circumstances might be treated like a throw-away operational Accounts Payable data whose value expires once paid. It is even possible that an organization does not currently have (or) have access to the necessary data, but this knowledge is essential to guide the efforts and initiatives of data warehousing.
To summarize, building an effective data warehouse requires a proactive approach. A proactive approach essentially implies that the organization makes a conscious effort to understand the business imperatives for the data warehouse; identify new metrics that best represent the objectives and proactively seek the data that is necessary to support the metrics. This approach can produce radically different results compared to the reactive approach of analyzing the data that is routinely available.