For many companies, there will be an information need that triggers a business intelligence type of project. This defines the question “What is the problem?” The very next step that the company decides upon is to buy a BI tool and hire consulting resources to implement a data warehouse solution costing anywhere from hundreds of thousands to millions of dollars, all with the idea that it was going to solve the business problem. This is “How do I solve the problem?” Read the rest of this entry »

As we enter the era of smaller-faster-better of everything where the Hummer is giving way to the Hybrids, the large unwieldy camcorders giving way to small, pocket sized, flash memory based camcorder, the humongous hard disks giving way to the small, compact, high-capacity flash drives, there are many interesting strides being made in BI tools to cater to the ever voracious appetite of the business community for meaningful dashboards, predictive analytics and data mining.

With hardware and high-speed memory becoming very affordable in the last decade,  BI tools have become more sophisticated and the large footprint bulky BI tools of the last two decades have given way to more nimble, faster, in-memory tools.  In-memory tools help business users analyze large volumes of data instantaneously on their desktops to see performance patterns, trending, perform sophisticated what-if analysis, and mine data.

What does in-memory mean and how does it affect you?

Yes, everything that runs on your desktop or on your server has to reside in the memory (RAM), but the in-memory here specifically means the ‘data’ resides in memory – empowering you to perform highly effective analysis on large volumes of data very rapidly. In traditional BI environments, a slow and laborious process was involved if you had to analyze say, five years of sales data to look at trending, Budget versus Actual or to perform a forecast based on trending.  The CPU had to fetch data from the hard disk, cache it, and keep going back and forth to the disk for accessing additional sets of data to analyze. With in-memory tools, the same task is done more effectively and with blazing speed.  The entire data is compressed in proprietary algorithms and stored in the memory of the desktop or servers, enhancing the effectiveness of users to run tens of what-if analysis, identify backcasts and forecasts based on the data trending.

This is a welcome relief to the analysts and executives that require heavy data analysis more effectively and efficiently. In-memory technology has also redefined the way dashboards are built for executives to monitor and manage their organizational performance. Dashboards can now be rendered on web-browsers seamlessly or on PDAs, for the executive on the move.

As there are many tools out there, selecting the best tool for your current and future needs is key to your BI success. Talk to various vendors, read about tool capabilities, get a tool bake-off and last but not least, take the help of a vendor-neutral consultant out there, that can help you decide the best way to move forward and implement the needed BI successfully for you.

For comments and questions, please email to eash.iyer@tychio.com or visit www.tychio.com

Nobody likes failed projects. Especially in this economic climate, when project funding is sparse and everyone is scrutinized like never before on their individual and team performances.

BI (Business Intelligence) projects are key for all businesses to make pragmatic business decisions. But we hear about many failed BI projects all around us. BI projects help organizations measure the business performance, but how do we explain so many failed BI projects – how do we prudently manage the performance of BI projects in the first place?

Below is a list of some of the top reasons why BI projects don’t deliver what they promise:

1)      No Executive sponsorship: Most BI projects are highly cross-functional in nature. It needs collaboration from various departments within an organization. Unless there is executive buy-in and commitment, seldom do departments come together in a timely fashion and collaborate, as every department has its own priorities and targets to meet.

2)      Underestimating the commitment levels, cost and effort: Let us be honest – to have a really meaningful BI solution, which helps strengthen the organization’s top line and bottom line, it takes sustained effort, which could translate into a considerable commitment and cost. Many BI projects are abandoned mid-way and deemed a failure as the executive sponsors are not briefed about the needed commitment in advance.

3)      Biting more than what can be chewed: The old adage “Don’t bite more than you can chew” is still valid today! Most BI teams get into a ‘monument building’ mode right from the word go. The best way to succeed in a BI project is to intelligently prioritize business needs and break them into manageable chunks.

4)      Making IT drive BI projects: The term BI stands for Business Intelligence – it means it is intimately related to the business! Ideally the business should be driving it with collaboration and support from IT. Many organizations let IT drive their BI projects from concept to completion with minimal involvement of the business. This is a sure recipe for mismatched expectations and unexpected results, often leading to failure.

5)      Trying to Fit a Square Peg into a Round Hole: This cliché is very much true in most cases where BI needs / requirements are not understood clearly and the project team tries to fit the business needs into an incorrect technology solution.

There is abundance knowledge out there in the marketplace. Select a team that has ‘been there and done that’ to help you out with your BI needs, to prioritize the requirements, to select the best technology solution and finally, to deliver a roadmap and solution – all in deep collaboration with your business and IT.

For comments and questions, please send an email to eash.iyer@tychio.com or visit www.tychio.com