Master Data Management (MDM) is a notoriously difficult, time-consuming, high-effort (but high-value) initiative. It's also an on-premises initiative - pretty much all the pieces of an MDM solution are old-fashioned on-prem software.
Any enterprise-class software that moves up into the Cloud ends up being tremendously easier to use, faster to implement, and simpler to maintain. Witness Siebel on-premises vs. Salesforce.com or any old-style on-premises integration product vs. a new Cloud-based integration stack.
So given how difficult MDM is, I've been wondering, "WHEN is someone going to come up with an all Cloud-based solution?" Well, today I have the answer to that. It's a partnership between SnapLogic - a leading vendor of Cloud-based Integration Solutions and Orchestra Networks - a leading Multidomain MDM software provider.
The solution incorporates data governance, product information management, reference data management and business intelligence dimensions on the back of Orchestra Networks powerful smartdatagovernance.com solution, leveraging SnapLogic's enterprise-grade Cloud-based integration, with a easy-to-use consumerized interface.
About Master Data Management
In short, MDM is about resolving the ongoing problem of conflicting, incomplete and inaccurate information concerning key entities (such as Customers, Products, Parts, Suppliers) across the entire organization. Specifically, MDM is about "reference data" rather than "transactional data" - for example, Customers (master data) buy Products (master data) which generates Orders (transactional data).
Fundamentally, MDM is about having a single view of the truth for key corporate information assets, to be used by all consumers of those assets (applications, processes, people). MDM is not simply a technological problem - in many cases fundamental changes to business processes are required and significant cross-organizational cooperation is likely necessary. MDM is as much about the maintenance of the Master Data as it is about the creation of the Master Data.
An MDM initiative may include a host of different technologies and capabilities, including operational applications, business intelligence/analytics, data/application integration middleware, data validation, data cleansing, repository/data store, workflow.
With so many moving parts, and with so many different parts of the organization involved - it's no wonder that it's an expensive affair. A typical MDM implementation can involve millions of dollars in software licenses and services revenue.
Unfortunately, because of the level of difficulty, MDM projects are also very prone to expensive and highly visible failure.
Why the Cloud Makes Sense for MDM
First of all, everything that moves to the Cloud gets easier to use. A lot easier.
It's not that the Cloud is full of magic pixie dust - rather, it's just that the software stacks that are built for the Cloud are new, more cleanly architected, free of baggage and legacy problems.
For example, many on-premises integration stacks have code dating back to the early 1990's, with user interfaces that would easily be at home in a Windows 95 environment. These products have been enhanced and extended with new features and layers of functionality dozens of times. Compromises and kludges abound. Functionality can be highly counter-intuitive. And as anyone who's ever been responsible for extending old creaky code, it's very difficult to add new capabilities - so new capabilities and features are slower to come to market. As a result, the products are significantly harder to use, productivity suffers, and maintenance efforts increase.
Secondly, increasingly more and more operational as well as analytical applications in the enterprise are moving to the Cloud. SaaS applications like NetSuite or Salesforce, and increasingly, Cloud-based analytics platforms are the default choice for those doing significant analysis and number crunching.
Unlike with an on-premises business application, a new SaaS application can be fired up almost instantaneously.
With the locus of applications and data moving increasingly and rapidly towards the Cloud, that's where Integration (mostly) belongs - and increasingly MDM as well.
Cloud-based Integration is Moving Up-Market
A few years ago, Cloud-based Integration was a curiosity. Then it became "cute" - and useful for small projects linking a small number SaaS applications or perhaps some limited integration with an on-premises system.
All that has completely changed. Just as SaaS and Cloud are becoming not only mainstream - but preferred choices for business applications, so has Cloud-based Integration.
In the process of making integration of SaaS applications and Cloud data easier, these vendors (of which SnapLogic is one) have made ALL integration (including on-premises applications and data) easier. This move into MDM is a definite demonstration of how Cloud Integration is enterprise-class.
Moving Forward with Cloud-based MDM
As with any new generation of technology, some organizations will jump on this bandwagon with great enthusiasm, and others will hold back - preferring to let others take the perceived risk first.
I would put forth the argument that for those data entities (for example, "Products"), where the applications and data sources may already be predominantly SaaS or Cloud-based, the risk is lowered by going with a new Cloud-based MDM solution.
A prudent course of action for organizations considering an MDM solution would be to consider a Cloud-based MDM solution - by looking at data entities which are already semi-public (such as products) and/or already heavily SaaS/Cloud-based (or moving in that direction). In many cases, these will be customer-centric data, although they may be Supply Chain-centric.
It took a few years for Cloud-based Integration to hit mainstream. It will take a bit of time for Cloud-based MDM to do the same. But it will happen quickly, and it will become the implementation architecture of choice for MDM projects.