InetSoft Webinar: Going Down the Road of Implementing Agile BI

This is the transcript of a Webinar hosted by InetSoft on the topic of "Agile BI: What is it and why should you care?" The speaker is Mark Flaherty, CMO at InetSoft.

Now for the realistic side of agile BI. It's not a silver bullet. There are some potential bumps in the road as you go down this road of implementing agile BI, and I want to point those out. So we will talk about some of the do’s and don’ts. And then I will come back at a later point and talk about how we can get started with this agile BI environment.

So let's talk about the top do’s, the first do. Focus on the business needs. I can't say it enough. You need to understand what your business users want. You need to understand what they are trying to solve and then stick with it. The second sub-bullet here is something that I find a lot of BI implementers doing.

They try to not only implement the basics of a BI solution to meet the first level requirements, but they dream too big, trying to reach too far. Unfortunately, the scope can expand very rapidly in that kind of an environment, and it can get out of control very, very quickly.

So the first do for implementing agile BI is to maintain your focus, and that focus is on what does the business want. What is the solution that I am trying to build in this very short amount of time? The last bullet point there is also something that is critically important and is sometimes missing in a lot of BI projects, whether it's agile BI or traditional BI. You have to identify a main point of contact in your business user community.

Who is it that you are going to go to when you run into a question or an issue or a problem that you need to have solved. You have to have that main point of contact, and say to them, it's your responsibility to get me an answer to this, I am moving at light speed, please get me an answer as fast as you possibly can. So that’s the first one.

The second key to a successful BI implementation is to prepare your data. Unfortunately you can't get around that. The quality of the data is very important. It's almost as important as how you visualize the data, how your business users visualize it, how they access it. In other words, you have got one shot at making your first impression of data accuracy, and that’s it. One shot at making a good impression, so put your best foot forward. If you make a mistake here, users will forever distrust the data and the BI system.

You have to understand how good does the data have to be to solve the problem? Sometimes good enough data is good enough to solve the business problem. But again if it isn’t then you need to bring in your data quality processing, your data profiling, cleansing and so forth. So make sure you put your best foot forward. That means you will also have to rethink exactly how much data you can get into this iteration. If the need is for heavy duty, high quality data then you may have to shrink the size of the data that you are bringing in so that you can handle that data quality processing, the profiling and the cleansing and so forth. So make sure that you do understand those quality needs before you launch into this BI project.

View a 2-minute demonstration of InetSoft's easy, agile, and robust BI software.

The third secret to attaining agile BI success is utilizing technologies that provide a sustainable architecture. Yes you do need an architecture. The architecture has to be both conceptual as well as technical, and it is going to be your roadmap for each of these iterations. Each of these quick iterations has to fit in to the roadmap. They have to fit together, and you will see that in just a moment. In order for that to happen, you have to make sure that you have an overarching architecture to understand how all of these pieces do interact with each other and do work together.

Now that doesn’t mean that the architecture is cast in stone. No architecture is. Any time you do an iterative type of methodology, you are always going to have to revisit the architecture, both conceptual as well as technological and make sure that it’s still appropriate. You have to make sure that you don’t have to tweak it some way or change it in some way to make the next project fit in.

And of course, you must look at open technologies and open APIs to make sure that the architecture is extensible and that it is sustainable over the long haul. The long haul maybe 4 or 5 years down the road. Last of the top dos here is to create and maintain documentation. If you read a lot about agile methodology they say, skimp on the documentation. Well that may work for certain types of project initiatives, but it doesn’t work in a business intelligence environment. Yes, you do need BI documentation.