The importance of Data Discovery in MDM projects
In this in-depth article series, based on a several interviews, we explore the most valuable lessons learned from our recent projects. We share both successes and challenges, to give you an unique insight into how MDM can revolutionize your business's Data Management and especially from the perspective of performing an ERP system change or upgrade.
Insights, challenges and lessons learned from real MDM Projects
Data Discovery is about finding data and identifying how it is connected within the organization. It is important to understand the relationships between one's data in order to make a smooth implementation of Master Data Management.
Our colleague Peter Karlsson has been deeply involved in one of Fiwe's recent MDM projects. We've asked him a bunch of interesting questions around this topic.
at Fiwe.
What were the biggest challenges you encountered during the project?
The biggest challenge was understanding the current situation. Information was often fragmented, spread across many systems and formats, and no one had an overview of the entire picture. It was difficult to create a complete map of the situation.
How did this affect you as a project provider?
It made it difficult to understand the whole. This understanding is crucial for a smooth implementation. Many questions remained unanswered for a long time. It often made it challenging to move forward, and to make sure we were working in the right direction.
How did the dialogue and communication work between you as a supplier and the customer organisation? Did you get to talk to the right people and the right roles?
In the beginning it was difficult. Those responsible on the customer side were selected based on their technical competence. As a result, discussions were often bogged down in technical details, and we did not get enough exhaustive answers. It is important to understand the difference between “what” and “why”, then technical competence alone is insufficient. One finds out the “what”, but all too rarely the “why”.
How did this affect the success of the project?
“We knew what was being done, but not always why it was done. This made it difficult to draw correct conclusions and make the right decisions for the next stage of the project.
How could you have avoided this problem?
“We should have been clearer with which participants were needed, especially from the business side.
The success of an implementation project is often increased if one gets access to an “omniscient” from the ordering organization. Someone who has an “umbrella” knowledge, not necessarily in detail but of sufficient breadth to be able to provide information about relationships and dependencies across the entire system flora. Did you have access to such a person in this project?
No, there was no such person, and it was a big challenge.
How did you guys work it out?
“We collaborated on another sub-project, which started a similar journey 1 year before our project. Their journey was smaller and more focused on that particular subsystem, but otherwise there were several common denominators. We used their conclusions as a basis, and although there were still many gaps, it gave us a good starting point.
How did you ensure that the information you received was interpreted correctly?
It was a challenge. This is something we will see in black and white only when we arrive at the testing phase.
Okay, so the project isn't quite finished yet?
“Parts of the delivery are ready and operational in the organization, but the major integration points are not yet operational.
“The 'Data Catalog & Governance' tool, which is part of Informatica MDM-suite, has been an invaluable help. It has proven to be a very effective way to map and document the solution.”
— Peter Karlsson, Senior Solution Specialist
Documentation is an important source of information, was there enough documented to get a good start on the project?
“No, the documentation was flawed in some cases, and it was difficult to know what was still valid in the material received. Changes may well have occurred after the documentation was written and without it being updated. Several of the systems we interact with have been in operation for more than 30 years.
How was the timetable affected by these shortcomings?
“Our ambition at the start of the project was to be able to deliver the first parts in 3 months, but it was delayed. The primary reason for this is that there have been a lot more changes and corrections than we expected. Largely due to a lack of overview of the existing data structure.
Now that we are conducting this interview, it has been 10 months, what is the situation at the moment?
“Today we have a much better overall picture of the data structure and its scope, and we look very brightly at the final parts of the project.
What is the single biggest take-away you have received from the Data Discovery project?
“The Data Catalog & Governance tool, which is part of Informatica's MDM suite, has been an invaluable tool. It has proven to be a very effective way to map and document the solution. With the data directory being the heart of the configuration itself, it becomes self-documented. This in turn leads to the fact that it is not possible to “forget about” documenting a change. The maintenance cost of documentation is therefore minimal, after all, it is already ready at the moment when you save your configuration changes.
