Guest blog by Corey Mellick, CEO at Amplifi
Is your business thinking about MDM? Have you already tried and failed? Is management beginning to question the value of the time and money being devoted to MDM? If you are like most businesses trying to address data management in an increasingly data-driven economy – the answer is probably yes to one or more of those questions.
In our experience, success or failure of an MDM initiative is determined fairly early on in the project. Having seen many attempts to implement MDM over the last decade, some common themes jump out that tend to predict success. These include the following:
- Pull up on the reins when you find yourself starting with technology. Too often, we have seen the best of intentions fail because the solution was driven from a technology perspective. A platform was chosen, something was designed, implemented, and “released” only to be rejected by the intended users as yet another IT data project. Data problems as they relate to MDM are usually not technical problems. They are more likely to be the result of business process evolution within the company. As the company has evolved, changes in the marketplace, changes to the company’s organizational structure, regulatory changes, etc. have all contributed to a collection of disparate processes, inconsistent data definitions and a lack of standardization across the organization. For example, years of change could easily result in a situation where something as critical as customer data is represented differently in sales, marketing, operations, and finance. Only after the business problems are well understood and prioritized should you begin thinking about technology options. So, pull up on the reins, build a comprehensive strategy with the participation of the business users, then select the technology that best fits what you are trying to accomplish.
- Focus on the business to start. This is probably apparent given 1 above, and several of our previous posts. MDM is all about solving business problems through the mastering of data and processes critical to the company’s operation. Start by identifying current business processes that are causing pain in the organization. You might think about this from the context of the different data domains, such as product, customer, supplier, etc. Once you identify the problem process, involve the current process participants in a discussion to get at the root causes of the issue. This will invariably lead to identifying the master data required. You’ll want to understand where that master data resides, who is currently responsible for it, why it causes problems, and what you plan to address during your MDM implementation. Next, identify the metrics or KPIs you will use to measure improvements in the process, and try to relate those metrics to impacts on the business (i.e., dollars saved, time saved, etc.). It’s this business first approach we have seen be most successful.
- Establish buy-in. MDM is more of a change management process than it is a data exercise within your business. As such, simply building it in the backroom and deploying it to the business users is not going to work. You need buy-in across the organization. If you followed our suggestions in 2 above, you are well on your way to gaining Stakeholder buy-in. To effect the change, you need the people who will be affected, the Stakeholders, to agree with the need for change, and the proposed change itself. Make sure to educate them on MDM and how it will enable the change that will make their work-life better. Next, you need Executive buy-in. Change will be resisted; it’s human nature. You need Executive management to reinforce the value of the initiative to the business consistently. To get that they need to understand and buy-in to the business need, the investment required, and the expected return in the form of measurable improvements. You should have what you need to articulate this from 2 above.
- Don’t boil the ocean. Nothing will sustain both Stakeholder and Executive buy-in like small wins along the way. To achieve this, start small and grow the initiative over time. And by starting small, we’re not talking about merely limiting your scope to a single data domain, such as product. Yes, you should focus on a single domain to begin with but go more granular and identify the process(es) in that domain to address first. With metrics for that process in hand, build something “good,” not necessarily “perfect.” Think iterative and collaborative. Most MDM platforms on the market are incredibly flexible, not bound by rigid data models that are difficult to change. They know you won’t get everything perfect out of the gate. So, take your best shot, deploy it to the Stakeholders who have bought in at 2 and 3 above, and iterate the specifics based on their feedback. Nothing causes an erosion of support like massive, lengthy, waterfall-like technology initiatives. You’ll lose buy-in before you ever get something fielded. Go small, show frequent wins, and you’ll have long term support for the project.
I’m sure these observations seem obvious to most. However, we still see the tendency to jump into technology first and figure the rest out later. We’ve all been there and done that, yet for some reason we see companies continuing to repeat many of those same mistakes when it comes to MDM. Unfortunately, taking an approach like that generally leads to rework, lengthy implementations, frustrated users, and lost executive interest.
The suggestions we’ve highlighted here come from seeing MDM done many, many times. Sometimes well, occasionally not so much. We believe taking a deliberate approach and a business-first perspective will give you the greatest opportunity for success in your MDM project. You need buy-in across the organization. If you followed our suggestions in 2 above, you are well on your way to gaining Stakeholder buy-in. To effect change you need the people who will be affected, the Stakeholders, to agree with the need for change, and the proposed change itself. Make sure to educate them on MDM and how it will enable the change that will make their work-life better. Next you need Executive buy-in. Change will be resisted, it’s human nature. You need Executive management to consistently reinforce the value of the initiative to the business. To get that they need to understand and buy-in to the business need, the investment required, and the expected return in the form of measurable improvements.