Archive for February, 2011

7 Ways to Give Your BI Some Guts

February 2, 2011

The power of business intelligence lies in its ability to provide users with the right data, in the right format, from the right sources, with the right integrity

Over the years, I have noticed that despite advancements in business intelligence (BI) technologies, as well as organizations’ ability to implement solutions in faster timeframes, the amount of time consumed in accessing and preparing the data still represents a significant portion of the effort. With the list of systems that collect and provide data growing by the day, the need for a solution that effectively provides the integration to access data sources is critical.

In other words, integration is the guts of any BI solution, and it should be built into the heart and soul of the solution if you are to get the most out of your data. A BI tool with a solid integration strategy can make the difference between the success and failure of your BI initiative and can have a huge impact on your business. At the end of the day, irrespective of the sexiness of the front-end dashboard, if the data being presented is unreliable or not available in a timely manner, then the solution is a failure.

In order to understand your requirements and the best solution for integrating desperate data sources into your BI solution, you must take a journey with your data elements. You need to understand how that pretty chart on your dashboard is created from the source data.

What pieces of information had to come together in order to calculate that one data point that you are seeing? Where did the originating pieces of data come from? What modifications were required to standardize and cleanse the data elements? What path did they have to take to get to the dashboard from the source systems? Where are the source systems located? How does that piece of information come to the source system? How often does it get there? And so on.

Once you make this journey, it will be easier to understand the power and need for an effective and solid integration strategy and solution for your BI environment.

As you consider your options, keep these issues in mind to understand the importance of data integration in making a better BI solution:

1. Getting the whole picture of your data. BI is only as good as the data it is connected to. Without an integration tool that can connect to all of the disparate data sources within your IT infrastructure, you will only get part of the picture. How are you supposed to make solid business decisions with only half of the information?

2. Access to the most current data, fast. Organizations collect more and more data every day, and they depend on having the most current information. This is why every BI solution should include integration architecture that automatically refreshes data from up-to-the minute sources. Also, high-performance integration won’t slow down when your business has an increase in data volume. The ideal integration solution will allow you to pull data from the source systems without impacting performance for production users.

3. Ease and speed of implementation. The faster you can implement a BI solution, the faster your organization can start benefiting from it. Businesses need rapid deployments and faster ROI cycles: The days of multi-year BI initiatives are so “legacy.” Integration tools that are easy for IT to learn and use mean they can get the solution up and running faster. No year-long waits.

4. Ability to transform data. Different systems may store data differently. An integration solution should allow transformation of data that does not meet the organization’s standards in order to ensure consistency in data arriving from different sources.

5. Data mapping. Mapping of data elements to some standard or common value is crucial for integration platforms. In particular, a GUI-based application to ease data mapping and transformation logic will make it much easier to integrate with new systems and ease the data transformation process.

6. Adaptability to change. Business can change at the blink of an eye; so can your IT infrastructure. If your organization experiences a merger or acquires a new company, the integration tool behind your BI solution needs to adapt quickly so those changes don’t affect your business’s decision making.

7. Higher-quality data. If you want high-quality answers about your organization, you need high-quality data. That’s why it’s important for your integration tool to have data profiling capabilities that will meet your requirements. Some data integration vendors partner with contact verification specialists to ensure that contact data is accurate and complete.

The good news is that the growth of options for integration tools over the last few years is making integration less of a challenge. Consider that in 2008, Aberdeen Group’s business intelligence survey data showed that the top two pressures driving companies to get a grip on BI costs were a need to move beyond data-integration challenges (42 percent) and to actually deliver BI tools to end users (29 percent). Compare that to Aberdeen Group’s 2009 follow-up survey, which showed that both data integration and delivering BI to end users had gone down to 31 percent, with ease of use for non-technical users taking the top spot with 39 percent.

Integration challenges are certainly being addressed with new tools from various specialty companies, such as Pervasive, as well as built-in tools and capabilities from companies such as QlikTech and IBM Cognos. However, integration is still a top challenge for a lot of companies due to the proliferation of new cloud-based applications and the significant growth in the sheer number of available applications and data sources.

Advertisements