Five enemies of Master Data Management
By Clinton Scott, Managing Director, TechSoft International
Pop Master Data Management (MDM) in your search engine and you will quickly unearth verbs such as uniformity, accuracy, stewardship, semantic consistency, and accountability. But, as anyone following the holy grail of MDM will tell you, the actual practice beyond the niceties in the verbiage, is fraught with implementation difficulties.
Many of your applications and systems were built with specific goals or functionalities in mind by their nature. It is the notion that "does not play well with others" that hinders the real value that can be derived from collating and streamlining your data.
After deciding to follow the MDM route in your business, it is as essential to know the challenges that lie ahead, as it is to keep your eye on the end goal. This article discusses five of the most common enemies that can unravel your efforts and turn an MDM implementation into a complete headache.
One: Little to no Business Engagement
If your business or users do not understand MDM or the goals you are trying to reach by adopting an MDM strategy or plan, the whole project will flop. Pooling your data assets for the greater good requires an overarching effort and a collective understanding. It is important to educate teams on why you are doing what you are doing, how you intend to do it, and when it will have the proposed impact.
By their nature MDM solutions can be clunky, which is in part why many of them fail, so also try and opt for a platform with attractive user interfaces and automated workflows that will then make it easier for your users to buy in to and use. This will elevate business engagement and oil the wheels of success.
Two: Outline short and long term goals
MDM will not happen overnight, in a week, or even a month. It is pretty standard for enterprise-scale MDM to be a multi-year journey as it is by no means a once-off event. If you have immediate needs, you must focus on those first to demonstrate a quick return on investment. You can then take the time to add aspects of the business to this journey and introduce new domains while you build your skills and reach for enterprise scale.
When looking for tools or platforms, make sure that you purchase a flexible multi-domain and multi-style solution where the vendor is committed to the future development of the solution. The last thing you need is to be halfway into a three-year journey to then find you have a platform that is no longer supported.
Three: Underestimating technical needs
This is implementation enemy number one. MDM is complex and requires a fair amount of understanding of, access to, and integration with data sources across your organisation. Today's modern digital and cloud-centric environment means that your IT systems and their associated data are distributed and span multiple repositories and locations.
Requiring your technical teams to continually build new models to get this data to come together is a behemoth and manual task. You can address this by purchasing agile data integration solutions that support these tasks' automation and ensure consistency in the end.
Four: Bad quality data
One could argue that bad quality data is the primary reason that MDM fails, but fixing this quality is also why it is needed. With MDM in play, you need to become more data disciplined and data-aware, because as you implement it, it will become abundantly clear as to why your data quality lacks lustre.
Striving for perfection will hold up the process, so it is better to aim for degrees of data efficacy. The quality of subsets of data can be worked on later, dependant on their overall significance, or you can change the process by which data assets are stored or collated in time. But to give your MDM project a head start, strive for efficacy as opposed to data perfection.
Five: Overlooking Compliance
Ensuring data compliance is oftentimes the primary reason that organisations elect to deploy MDM. Especially as customer, employee, and product master data are central to the accuracy of audits and reports. Failure is not an option, and collaboration between departments is key. There is a process that will help you meet compliance obligations. It starts with bringing together business users who understand regulatory requirements, engineers who can transform these requirements into master data assets, and lastly data stewards to perform ongoing quality control.
In our experience, customers are often so daunted by MDM that they continually put it on the backburner. But it needn't be a monstrous task if you focus on your immediate needs and allow it to evolve with your business. If you require to improve your reference data quality, then start there and let your MDM solution develop. When in place having a single point of truth for your data assets will prove to be invaluable in the quest for continual business reinvention.