A People Business: IRM UK Data Governance Conference and Master Data Management Summit 2019

In Data Governance, Master Data Management, Reference Data Management by IRM UKLeave a Comment

Print Friendly, PDF & Email

This year’s edition of the co-located Data Governance Conference and Master Data Management Summit saw a change of scene – from Portman Square to Tower Bridge, arguably a better panorama for the many continental European delegates. There were delegates from 25 countries who attended the event in the mid-May sunshine.

Jeremy Hall, Managing Director, IRM UK
Click on the following links for information on our 2019 Master Data Management Summit & Data Governance Conference Europe, London.  IRM UK’s next Master Data Management Summit & Data Governance Conference Europe will take place in London from 11-14 May 2020 – please let me know at [email protected] if you would like information on either attending or speaking at this event.

“Where do we go from here?” was the guiding question of the event’s closing plenary discussion, guided by Jan Henderyckx, from BearingPoint and Gerard Bartley, Jacobs Douwe Egberts. The answer to that is unlikely to be: “buy more technology”, going by the tenor of many of the sessions. For what came across loud and clear – in what are the mature, conjoint disciplines of DG and MDM – is that human factors and business purpose are more important that tools.

This was especially evident in the closing plenary, but also in the MDM keynote panel on the future of MDM, featuring Michael Evans, from Comma Group, Salah Kamel, Semarchy, Scott Schumacher, IBM, and Conrad Chuang, Tibco; and moderated by Malcolm Chisholm, First San Francisco Partners.

The keynote talks by Rick van der Lans and Mike Ferguson rounded out the bigger picture, in which technology does play a vital role. Nevertheless, as Mike is fond of saying, adapting Peter Drucker, “culture eats technology for breakfast”.

Where Do We Go From Here?

In the closing plenary session, Jan Henderyckx was keen to put across that every organisation is different. “What a data culture is varies from organisation to organisation. And it is different at strategic, tactical and operational levels, which means you have to work on multiple stories [that come together] so that everyone does the right thing”.

Asked, by an audience member, what to do, then, “when you have inconsistency between corporate strategy and tactics being used at a managerial level”, Jan said the “answer is often to make the problem a little bit bigger to make visible the lack of alignment. Just put it on the table that there is misalignment”.

Gerard Bartley had some advice for participants about how data governance and management professionals can interact productively both with peers and with executives at the topmost level of organisation. “If you get 10 minutes with the CEO, present a use case around where something has gone wrong. For example, there has been a customer problem due to a factory being shut down for two hours”. That sort of concrete problem, which could be headed off by better data management or governance, is the best way to focus attention with an executive at that level, he said.

“Clarity of message is key when talking to the business”, he said. “You have limited time”. Role playing workshops for the data governance team, ahead of presentations to the business is worth considering, he said. “That helps you to come up with a list of questions you are likely to be asked. And role play hostile audience members, too. Get the answers straight and you will have nothing to fear”. He also emphasized how important it is to “just be part of the business. Be a professional person, don’t come across as tech support”. And he advised to “recruit for attitude and train for skills”.

The Future Of MDM

This advice to focus on the needs of the business first and foremost, not on a technical agenda as such also came across strongly in the MDM keynote panel.

Scott Schumacher, from IBM, advised attendees that, when planning out an MDM strategy “[you] always have to be business focused. If there is no business use case, if you just have IT saying ‘a golden record would be a good thing’, that never works. How you show ROI [Return on Investment] is always the main thing, unless your sponsors have a lot of money and patience! And it is not good when an organisation is being forced to adopt a process by a technology”.

There was some dispute on the panel as to whether MDM programmes and projects should be labelled “MDM”, or whether the term itself is best avoided in favour of whatever is specific to the lexicon of the particular organisation.

Salah Kamel, from Semarchy said getting business value is “more about registering important data in a single place than the term MDM as such. MDM has created a lot of bad press so don’t call it MDM: focus on the business case, and start small and iterate. Prove the value to 10 people who are suffering”, he said.

However, Conrad Chuang, from Tibco said: “I don’t agree that should not call it MDM. If you can’t name something it is hard to keep moving forward. You need overall conceptual clarity, at least for the core [programme] team. Otherwise it is just an endless stream of projects. There can be security in an acronym”. Dr Schumacher, from IBM added: “We are stuck with what the analysts say we are. And so we can’t avoid the term. MDM is term we have to use because that is what Gartner says we are”.

Rick van der Lans: Enterprise Data Marketplaces

In his keynote, Rick van der Lans advocated an approach whereby data professionals can add value to their businesses by adopting what could be described as a publishing mind set. Publishers don’t always know what books will fly off the shelf. They have to take a punt, just as any commercial entity has to. A classic example is Steve Jobs at Apple launching the iPhone onto a market of users who didn’t know they needed a smart phone.

Rick advanced the concept of taking the model of data market places – where data is sold to the value of $18bn or so each year – and applying it internally across an enterprise. This is important, he said because the “biggest change in the last decade is usage – doing more with data”.

But in order to get to a place where enterprise data is enthusiastically and effectively consumed by business end users, we have to break with older data supply chain models, he said. “The long chain of the traditional data architecture is 30 years old and is still there. It is like a rigid assembly line, far too inflexible for today’s patterns of usage”.

It is also, he said: “too controlled by IT, so that final reports get copied into spreadsheets, then combined with other database, stored in Access and so on – because we haven’t given users want they wanted”.

Rick said there is a “challenge in building data products, where you have to do so in advance of knowing what the users want. There is still a chance you are wrong and it requires a deep understanding of the business”.

Why not, then, leave final consumption to the users? “Self-service BI tools are very popular, but are still hard to work with, in truth”, he said.

Instead, he advocated a “unified data delivery platform” obviating the myriad of separate data lakes, data warehouses, and (now) data marketplaces that characterize complex corporate organizations, littering them with siloed data delivery systems

For Rick, data virtualization is the critical technology. “It is the key technology for this, it can decouple users from sources – it’s about abstraction rather than ETL-ing [extract, transform and load]”.

Rick will be presenting 2 one-day workshops for IRM UK at our Data Ed Week Europe 2019 and collocated Enterprise Data & Business Intelligence and Analytics Conferences Europe 18-22 November 2019, London.  Practical Guidelines for Designing Modern Data Architectures and New Big Data Storage Technologies: From Hadoop to Graph Databases, and from NoSQL to NewSQL

Mike Ferguson: Unified Data Delivery

The current complexity and chaos of data delivery systems in big companies today was also the governing theme of Mike Ferguson’s keynote, ‘Unified data delivery – shortening time to value in a digital enterprise’.

“Digital transformation projects are creating more complexity than ever”, said Mike. “The answer to that has to be rationalization, the organisation of data lakes and data warehouses into a single logical data lake”.

Mike advocated a “publish once and re-use” principle.

He shared some survey results from data-focused strategy firm New Vantage Partners. This year, in 2019, nearly 88% of those surveyed in a ‘Big Data and AI Executive Survey’ by the firm said investing in Big Data/AI more urgent now than ever. But, in the same survey, it emerged that 72% of firms had no data culture; and 92.5% cited people and process as the main stumbling blocks, not technology.

The siloed nature of data management and BI continues to be a real problem, said Mike. And, if anything it is getting worse. Self-service data preparation has been hot in recent years, but ungoverned it can be one of the latest contributors to yet more data chaos.

Needed is a chief data officer or some such, he said. And, from a technical point of view, he advocated a logical data lake, where the data exists in many data stores but is managed, operated and governed as if it were centralized. Mike listed out data management platforms and data catalogue products that can be chosen and enlisted in this effort. And he outlined an information supply chain – ingest, discover, catalogue, prepare, analyse, decide, and then act on data.

But he also stressed the importance of getting the organisational design right, with a CDO at the top. Otherwise, complexity will continue to cripple corporate organizations.

Mike will be Presenting a two day Course on Unified Data Delivery for IRM UK from 10-11 October 2019 in London. The next data-focused IRM UK conference is our Enterprise Data Conference Europe & Business Intelligence & Analytics Conference Europe co-located with our Data Ed Week Europe 2019  taking place at the etc venues location at Monument in London from 18-22 November 2019. Mike will be presenting a full day workshop on Data Management in a Hybrid and Multi-Cloud Computing Environment and a keynote on Unified Data Delivery – Shortening Time To Value in a Digital Enterprise

About IRM UK; IRM UK specialises in strategic business and IT training for business and IT professionals and management. Since 1999 when IRM UK was founded, over 25,000 professionals from over 2,000 organisations worldwide have attended our events. Noted for the participation of middle management and top-level decision makers from both the corporate, user and vendor communities, and the lucid analysis of critical strategic and management issues, our events are condensed and rigorous combining technical explanations with management advice and discussions of future directions.

Copyright IRM UK

Leave a Comment