ComputerWorld.in

Though considered simple and drab, information management can become quite a tricky exercise. That’s because it is fairly easy to end up being enamored by the tools and technologies that enable enterprise-wide information management and embark upon a project that could become a one-way ticket to failure.

Why do I think so?

Simple. At Asian Paints, we have engaged in rolling out an information management ecosystem. And, with a little prodding it is easy to figure out the way people generally look and approach a subject. I think people often focus too much on introducing the tools to address the issues they face with managing structured and unstructured data. But, if you ask me, this approach can actually complicate the entire exercise. And many have burnt themselves going down this road.

We looked at this exercise beyond an attempt to classify information. I believe it should all start with one question: who will consume what type of information? Simply put, who is the consumer of certain information and what will they do with it when they get it? These are key questions that one needs to get the right answers to before planning an enterprise-wide project. I believe that you can get data classification right, but still end up with wrong information or data that’s not in the form your business users want. And then you would not have done a great job.

Rather, I would worry more about data integrity. If classification goes wrong, somebody can point it out later and you can recover from the problem fairly easily. But, if you get data integrity wrong (in terms of a source not being validated) you may start introducing gibberish into your information management system. I think, knowing where information is coming from and how you will clean it is critical in handling external sources of data. Hence, it is paramount to be proactive in figuring out the integrity of data.

This brings us to the subject of managing unstructured data that can be found all over the enterprise. One thing that was missing before service-oriented architecture came to the real world was relating the content of unstructured data to relevant processes. A process engine plays a very important role in this space because if I want to go beyond mere ‘dash-boarding’ and reporting, I will need to empower users with certain tools. These tools should enable data to be seen in a certain context and help users initiate a change or take a decision. You need to link information back to action to be able to bring about relevant changes, changes required by any dynamic business. By marrying data to a process, information can lead to transformation.

Obviously, the other piece I would want to add to this puzzle is collaboration. If I am not keen in tying information with a process to effect a big change, I may want to look at certain information together in a context and share it in such a way that allows people, functions and processes to work together. This entire gamut of information management tools, process tools, and collaboration tools put together form a basis of what I term an information-led transformation strategy. We’re embarking on that journey at our organization.