This month, data management professionals from across Europe and beyond will converge on London for the Master Data Management Summit Europe 2012. It’s no coincidence that the MDM summit will be held in the same space as the Data Governance Conference Europe 2012 -- the two topics are intimately linked. Effective data governance requires accurate and consistent master data and, equally, a key goal of MDM projects is to improve and ease the process of data governance.
But the fact that these two concurrent events have separate billing is also telling: for despite the linkage between them, there remains a nagging disconnect. And although organisations have come to realise that MDM projects will not work effectively without good governance, IT vendors have been slow to offer customers the tools they need to do the job effectively.
Learn more about master data governance
Discover how MDM must start with data governance
Get the full picture on governance and MDM with SearchDataManagement.com’s learning guide on MDM
Aaron Zornes on how MDG can fuse MDM to BPM
MDM analyst Aaron Zornes, who is chairing the summit, has been banging the drum for more effective master data governance for some time. But because of vendors’ sluggish response, many customers have been left in the unenviable position of having to develop their own manual mechanisms and processes with scant help from the industry, in his view.
“Manual data governance is error-prone, time-consuming and unable to ensure compliance or measure business impact,” Zornes said. “What’s been missing are tools that allow the business to define both IT and business governance rules across their different master data domains and ensure these rules are proactively enforced and adhered to.” That, he added, raises the question of why vendors aren’t using workflow and business process management (BPM) functionality to link the data governance side with the operational MDM side.
Governance guru Jim Orr, a marketing executive at software vendor Information Builders and author of Data Governance for the Executive agreed with Zornes that the required tools aren’t much in evidence.
“From my perspective, MDM is plumbing, but if you want things to flow through that plumbing properly you need to have governance around it,” Orr said. “When it comes to information integrity and processes to enforce that, the industry has fallen flat on its face. We’ve done pretty well in pockets, but overall vendors don’t want to deal with it because it’s the harder part of the equation.”
The challenge is exacerbated by the breadth of the discipline, he said. “Governance involves many different aspects -- data modelling, quality, metadata management; the list goes on -- and it’s the goal of many organisations to be able to have a platform that can manage as much of that as possible from a single point.”
MDM, data governance gap
Before they can hope to close in on that goal, however, organisations need to come to terms with what data governance is about. Orr again: “Everybody has a different slant depending on their industry and role. Data architects see it differently to data stewards, data governance directors, information management directors and so on. To bring things into perspective, companies must understand how data governance holistically drives business performance. But there’s still a huge gap between technology and business processes.”
As a result, few -- if any -- organisations have reached their goals. “We’ve seen a spectrum of governance maturity across our customers,” said Steve Sarsfield, product marketing manager for data governance at open source MDM and data integration vendor Talend. “And depending on their level of maturity, the data breaches and potential impact of poor data management they suffer varies.”
Sarsfield added that although data governance has become a critical issue for many corporate executives, “reaching maturity is challenging because it requires a coordination of people, process and technology.”
Agile departmental MDM
Most organisations can’t grow up overnight, according to Sarsfield. “Very few companies go straight from chaotic, ad hoc data management to full-blown strategic MDM,” he said. “A more popular approach is an agile one, where MDM starts with departmental implementations and evolves to companywide implementations.” That enables “a natural evolution” towards formal governance processes, “and organisations that move up this scale can mitigate unfortunate circumstances much more effectively,” Sarsfield said.
Manual data governance is error-prone, time-consuming and unable to ensure compliance or measure business impact. What’s been missing are tools
Aaron Zornes, MDM Institute
Many companies with ongoing MDM programmes have put in place MDM hubs for managing different sets of master data, like customer, supplier, product and financial records (with about 80% of them using tools from megavendors IBM, SAP, Oracle and Informatica, according to Zornes). But generally, they don’t have the different hubs linked together in the way required for effective master data governance. Zornes noted that such “random acts of MDM” are creating the undesirable by-product of “master data silos.”
He said that, at a high level, reaching the next stage of governance maturity requires organisations to put in place four fundamentals: effective organisational processes, data policy management, proactive data stewardship and metrics for both business and IT. While it’s difficult to pinpoint any organisations that are there yet, Zornes said the financial services industry is making the biggest strides, largely because of the weight of regulation that mandates them to get governance right. “Banks and insurance companies have to prove they’re governing themselves effectively,” he said. “They must address compliance, anti-money laundering, lots of different regulatory requirements around the world. So they have a huge incentive.”
And in spite of his criticisms of the IT industry, Zornes, who is founder and chief research officer at the MDM Institute consultancy in the US, said the tools for the job are finally starting to emerge. Some smaller, niche vendors have for a while been offering products that can, at least to some extent, integrate master data from different MDM silos -- Kalido, Collibra, Orchestra and others. But Zornes thinks they might have missed their window of opportunity to capture a big slice of the market, since the market leaders are finally starting to respond with products of their own.
But there’s still some way to go. As Orr noted: “There’s clearly no single package that does everything an organisation wants to do. These are complicated solutions -- and the bigger the organisation, the more complex and difficult it becomes. And we need to accept that tools alone are never going to solve all the word’s data governance problems, which is another chasm many organisations have yet to cross.”