Pilot Multimedia has been in Business Intelligence business for 18 years. Back then it was known as DSS or Decision Support Systems. The BI mission has always been our organization mission viz; Pilot Multimedia is in the business of delivering solutions that take data and transforming it into insights and decisions.
The technologies that support BI include Data Warehousing, Data Integration or ETL, OLAP, Dashboards, Business Reporting, Data Quality, Metadata management and Data Visualization. Data Mining is at times classified as part of BI, we have regarded it as part of our Advanced Analytics offerings.
Data is only useful if it is consumed for analysis. Data does not add value if it is stored and never consumed. For data to be used purposeful, the data schema needs to be arranged in a format which supports analysis. A critical error made by CIOs or IT managers is to generate reports from operational data schemas. This is tantamount to asking an ostrich to compete in a flying competition. Operational systems are design for transactional operations and not for analysis. Data marts and data warehouses are data schemas optimized to support business intelligence reports. Without a well designed data mart or data warehouse, reports can take days to generate or perhaps never at all.
The use of Business Intelligence (BI) tools such as SAS Enterprise Guide (SAS EG), Oracle Business Intelligence (OBI), Business Objects, QlikView, Cognos are used as presentation layers to display the data. The advantage of such tools is that they allow business users to easily manipulate data. With greater data understanding, it is inevitable that requirements will evolve. With business intelligence software, users are empowered to modify the data on their own accord.
Different users, have different needs. Dashboards refer to highly summarized reports for senior management. Such reports summarize data across different business units. Flat reports typically refer to granular or detailed information that is useful for business operations.
OLAP stands for Online Analytical Processing. A well designed OLAP cube can allow for very fast processing of large data volumes. This is vital for a BI implementation as an incorrectly design cube can take days to generate a set of reports. At Pilot we have the experience and skills in properly designing cubes which balance between flexibility and performance.
There exist two challenges in every BI Implementation; getting the data in, and getting the correct reports out. The bottlenecks often occur at both ends.
The first challenge is in the data integration wherein data from a myriad of sources need to be consolidated in a single data mart or data warehouse. Additional issues such as poor data quality, multiple and inconsistent data definitions add to the project risk. Perhaps the hardest aspect of data integration is in the data mapping exercise. Data mapping involves a business analyst, studying the source data and mapping the fields correctly to a data model. This task is challenging because IT and business frequently communicate in two different languages. Miscommunication is inevitable. This process is slow and fraught with reworks and iterations.
The second challenge is in designing insightful reports that provide decision makers with useful knowledge. Ideally, decision makers should have a clear mind on the type of reports and analysis that they would like to have. In practice, this is unlikely to be true. The dictum that applies is “I can’t tell you what I want. But show it to me and I can confirm for you on whether I want it or not”. Without a clear understanding of what the end requirements will be, a typical BI project can be fraught with danger. Imagine all the hard effort, in integrating the data from multiple sources, large expensive purchases on servers, databases and BI software. All this effort expended only to fail due to poorly designed reports which provide trivial insights.
How can we help you? With 18 years of BI implementation experience, we can help steer through your BI implementation with success. Don’t suffer project risk. Don’t generate the wrong type of reports and analysis? Don’t write queries that perform poorly and take days to generate reports? Call us.
