MDM & DG

This blog post at an initial glance looks to be about a jumble of letters.  One thing that perhaps all data management professionals are guilty of is using too many acronyms!  Now I hope if you are previous readers of my blog that you immediately realise that the DG stands for data governance (after all in a blog from me what else would you be expecting?)  And the MDM stands of course for master data management.  Now I say “of course” because I have been involved in a number of MDM initiatives over the years, however, several recent experiences and conversations have made it clear to me that this term is still not well understood. 

So let's start with the basics; according to the DAMA Dictionary of Data Management MDM means:

 Processes that control management of master data values to enable consistent, shared, contextual use across systems, of the most accurate, timely and relevant version of truth about essential business entities.

So if MDM is about the management of our master data, we need to be clear what our master data is.  At its simplest it is a single source of data that is critical for a business to operate and designed to be used across processes and systems. For example: product, customer, employees, suppliers etc.  Some people call this a golden record and if you're talking about mastering customer data, you often hear it referred to as a single customer view.

Now that we all understand what MDM is, I can share that my biggest concern is not the lack of understanding of what it is, but more worryingly that a significant proportion of people do not understand how critical implementing data governance is to the success of your MDM project.  Indeed this message was mentioned in several presentations I attended last week at the excellent Master Data Management and Data Governance Conference in London (there is a clue in their relationship in the fact that the two conferences are run jointly).

A few weeks ago I was asked if I could use an analogy to explain this relationship between MDM and DG.  I've never used one before and couldn't immediately think of one I felt entirely happy with.  Being a strong believer in the wisdom of crowds, I posed the question on LinkedIn and got a great response (including the usual warnings about the dangers of using analogies!) 

The best analogy we came up with is to imagine that your MDM system is the equivalent to the human body, that the data on that system is the equivalent of food/fuel for the body and data governance is the rules/commonsense applied to what food you do or do not consume.  As we all know if we consume good wholesome food, we feel healthier, have more energy and tend to make the right decisions.  If we have abandoned healthy habits for a period of time and consumed lots of junk food, then our bodies fail to work properly, we feel sluggish, tired and we start to experience problems with our health. 

If we translate this into the data world, it is often the case that organisations implement an MDM solution and migrate data into it with little attention to the quality of that data. It is fairly common for an MDM solution to be implemented without data governance.  This inevitably means that over time as poor quality data is consumed by the system, that it in turn causes problems with the processes that are using that data. 

But what does that mean in practice?  Well let's say, for example, you are implementing PIM (sorry another acronym, Product Information Management, which is MDM for product data).  As you know data governance is all about the people and processes that manage data, so if you implement PIM without DG the types of issues that can occur are:

·      It is likely that the PIM solution will viewed as a technology refresh only and it becomes more challenging to engage the business.

 ·      If the business are not engaged they will not change their approach to capture and manage product data

·      If existing processes of capturing and managing data do not change, the quality of the data will be no better than on the previous system/systems and,

·      All the issues you are experiencing with poor product data previously will continue.  The list of such issues is extensive and can include such things as customers receiving different products than those they ordered or not even being able to find them on your website and also issues with supply chain logistics because product dimensions have been incorrectly captured.

If you're spending a significant amount of money, time and energy implementing a PIM solution, then surely you would want the data quality to improve and with it improve the customer experience and service and also reduce costs at the very least?

This is where data governance has it’s part to play, in ensuring the success of your PIM solution.  Implementing a data governance framework is crucial in defining and articulating the roles that your business stakeholders have to undertake in defining, monitoring and managing your product data.  This framework in itself is a key component in the successful engagement of your stakeholders and transforming for example Product Managers, from people who thought they managed products (and the suppliers of those products) into people who understand how vital it is to get the correct data captured accurately about those products.

When the data in your PIM system is well structured, well defined and of good quality, then the potential benefits to your company are significant.  So what are you waiting for?  Please make sure that if your organisation is currently implementing PIM, or any other MDM solution, that you implement data governance at the same time.

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

Data Governance Interview - Nigel Light

I am pleased that this Data Governance Interview is with experienced Data Governance Practitioner Nigel Light who has been working hard to improve data quality at Ecclesiastical Insurance (and yes they insure churches!)

After completing a degree in Chemistry and Geology at the Univeristy of Leicester Nigel has since followed the traditional path of IT development and systems/business analysis (mostly in General Insurance), over the last 30 years. 

Outside of work he enjoys being outdoors, particularly cycling and walking, trying to tame his garden and is a long-suffering Gloucester Rugby fan/member (which, I feel, entitles me to the odd beer or 2 at times!).

How long have you been working in Data Governance?

 On and off, around 8 years.

How did you start working in Data Governance?

I first started working to improve our client data as part of a CRM project but have since been instrumental in implementing solutions to improve our decision making data.

What where your initial thoughts when you first fully understood what you had got into?

The anorak fitted!

I get satisfaction from making things better and delivering improvements. The link to the business benefits from improved data that can be made appealed to me and I soon became a strong advocate.

Are there any particular resources that you found useful support when you were starting out?

I have been lucky with the solution vendors that we have worked with in that they have been really keen to work with us,  understand our requirements and share their expertise.

I have learnt a lot from them and also from the Data Quality and Governance Community as a whole who are always helpful and forthcoming.

What is the biggest Data Governance challenge you have faced so far?

I think the challenge is yet to come. Setting up a Data Quality Programme is a no-brainer and most people buy into it; making it stick when other business distractions come along is much harder.

What have you implemented or solved so far that you are particularly proud of?

Apart from the technical solutions we have built and deployed, I am proud of the way that certain individuals within the organisation I work for have bought into the process; to the extent that one business area regularly reports zero Data Quality issues each month!

What single piece of advice would you give someone just starting out in Data Governance?

 Get away from your desk and spread the word.

Data Quality won’t truly become ‘Business As Usual’ within an organisation unless there is a shift-change in mentality and I try to take every opportunity I can to promote the benefits of the programme and support the business community who are actually responsible for capturing, correcting and improving the data – as well as deploying solutions to stop issues occurring in the first place.

This should be supported by some really meaningful KPI measurements to help monitor improvements which can be used to encourage, motivate and support change.

Finally, what do you wish you had known or done differently when you were just starting out in Data Governance?

I realised early on that all data issues could not be fixed in one project! It is a more of a slow burning, long-term aim with gradual improvements being made rather than a big-bang, one-off approach and on to the next task.

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

Data Governance Interview - Peter Aiken

For my blog this time, I am thrilled that Peter Aiken kindly agreed to be interviewed.  I was lucky enough to be able to find a time when we could speak to each to do this interview, so you will find that some of the answers are longer than usual for the interview series, but I’m sure that you will find that as interesting and insightful as I did.

Peter Aiken is acknowledged to be a top data management (DM) authority. As a practicing data consultant, author and researcher, he has been actively performing and studying DM for more than 30 years. His expertise has been sought by some of the world's most important organizations and his achievements have been recognized internationally.  He has held leadership positions and consulted with more than 75 organizations in 21 countries across numerous industries, including defense, banking, healthcare, telecommunications and manufacturing. He is a sought-after keynote speaker and author of multiple publications, including his latest the "Case for the CDO" & "Monetizing Data Management." Peter is the Founding Director of Data Blueprint, a consulting firm that helps organizations leverage data for competitive advantage and operational efficiencies.  He is also Associate Professor of Information Systems at Virginia Commonwealth University (VCU) and past President of the International Data Management Association (DAMA-I).

How long have you been working in data governance?

Well, it’s a relative question, so I would say my initial experience, I worked for the Department of Defense in the late 1980s and at that time were doing what we were today would be considered data governance.

I was very fortunate in that I was going through my terminology graduate program and I had a professor named Professor Al Davis, who sat us down one day and said “I am putting together a new special edition for a journal and we are going to cover the subject called Requirement Analysis.” As most people are aware requirements is the first and most important part of any problem solving activity. If you don’t get the requirements part right you can’t build a correct design to any specified proposal. He made us graduate students go through all the papers that had been submitted for the journal. And we were supposed to recommend him to which ones we would accept.

So we graduate students started conferring behind the scenes and saying “I am not seeing any value to most of these papers”.  But of course we did not want to say that to him because it was his friends and colleagues that had submitted the papers. And, he of course forced us to down in the room and said now we’re gonna play a little exercise. I want you to write down, first of all, on a little piece of paper how many of these things would you accept. We had to actually say none. And of course he said “I thought they were all a bunch of rubbish as well.”

We breathed a collective sigh of relief. The key was that most papers were so theoretically in nature that they bore no resemblance to the real world problems that he was trying to address. We then spent the rest of the semester looking at requirements and it was very easy to determine that data requirements were in fact the most objective requirements that one could state and of course one of the primary functions of data governance is to manage the data requirements.

You are an author yourself, but is there a particular book or resources you would recommend, particularly for those just starting out in Data Governance?

I’d like to encourage people to research is the professional association, like IAIDQ, the TDWI and of course we do not want to forget DAMA, as a good place to go. DAMA has put a renewed focus on the business side of things.   I would love to put in a plug for my latest called “The Case for the Chief Data Officer.”

I wondered if you could share perhaps the biggest challenge you’ve ever faced in Data Governance implementation?

There are two types of problems. The first one is of structure. If the structure isn’t right then some of the rest of the effort becomes incredibly difficult. So I’ve walked in to a new engagement the first day and said “I am sorry I don’t think this is going to work”. And people say “oh, how can you be so definitive at this early stage of the project?” I said “well we are down three levels in IT. If I want to see the responsible Executive, it’s going take me six weeks to get on their calendar”. There are not conditions for success for any organization. That’s what I would call a structural problem.

The other type of thing is practice-oriented problem and that is manifest if you have a Data Governance organization but perhaps the person who is running it may not have the necessary soft skills to do this.

Are there any industries that you haven’t worked in yet that you’d like to?

I don’t know if there is anything that I have wanted to get in to but have not yet. But I am sure there is some out there that just have to go back and look!

Is there one single piece of advice you would give someone just starting out in data governance?

I guess it would be to research it before you get into it. There are advantages in going into a new field. I guess if you are comfortable with uncertainty then it would be a good career field. If you are uncomfortable with uncertainty this probably is not the area you want to get into.

Is there a particularly memorable data governance experience amusing or otherwise that you would like to share?

Well, it wasn’t much of a funny one but very effective one. Data Blueprint was involved in the Army Suicide Mitigation Project early on. One of the things we had to do was call a Council of Colonels. This was a group of people who were all trying to contribute data to software to solve the problem of service men and women who harmed themselves after action. A very noble and  very worthwhile project. The Secretary of the Army attended one of these meetings and after the 3rd person got up and said “can this data can be used for this purpose?” he put his notebook down on the table rather loudly with a thump that got everybody’s attention. And he said “ladies and gentlemen I have an announcement to make. From now on we are going to refer to this as my data and everybody else in the room went oooooohhhhh. He then said that anybody that wants to come and tell me why my data shouldn’t be used to save the men and women that are in the armed forces from causing harm to themselves, then make an appointment with me. Of course at that point everybody in the room said ok. I don’t have to worry about covering my butt. The Secretary of the Army has got my back and is going to support me with whatever we do because all we are doing is trying to save lives.

I’ve told this story in a number of corporations and urged leadership in those organizations to help out. If they would take the same sort of courageous stance with that I could save the hundreds of millions or billions, in some instances.

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

Why are there so many different Data Governance definitions?

Don't you think it's strange that a data management discipline that, amongst other activities actively promotes defining data, has so many definitions of what the discipline itself is?  I think it's bewildering and to be honest very frustrating.  I think that it is vital that we all mean the same thing when we are talking about data governance, but sadly this is not yet the case. It means that whenever I am presenting or running one of my courses on data governance, I start each and every one with a section on what data governance is, but I don't feel the need to do the same if the session is about Data Quality!

So why are there so many different Data Governance definitions?   Pick up a Data Governance book (not that there are many to chose from) and very early on you will get a definition of data governance or an explanation of the concepts behind it.  But if you pick up a Data Quality book they don’t define data quality, they dive into what causes poor data quality and why we may want to improve it. So why as a data management profession have we not yet agreed and universally understood what Data Governance is?

I don't know the definitive answer, but I believe that it could be down to the following factors:

Name

It could be due to the choice of name of the discipline after all the definition of just "governance" is fairly vague:

the action or manner of governing a state, organization, etc.

So from that rather woolly definition we can infer that "data governance" is the action of governing data, but that can mean any number of things and I believe this is a reason why many take it to mean governing access to data and hence the common misunderstanding that it is another term for data security.  Others take it to mean governing how long we hold data for and confuse it with data retention.

I think (hope) that the term is used to mean the correct thing the majority of the time.  It's inclusion in such things as the DAMA DMBoK and various regulations (including Solvency II), has got to be helping.  However, while a number of Data Security Software vendors persist in labeling their products "data governance" confusion will persist.

Foundation / Framework

Data Governance is a framework or foundation that underpins or supports the other data management disciplines (indeed it features at the centre of the DAMA DMBoK wheel). Although there are activities and processes to follow, Data Governance is all about putting in place a framework for who does what to manage data.  Other disciplines are much more about a specific set of activities e.g:  data modeling or data profiling and the focus is much more on the “what” rather than the “how”.

Relationship with other disciplines

As I mentioned in the point above, whereas many data management activities can be viewed on a standalone basis, data governance is rarely carried on for it’s own sake. It enjoys a symbiotic relationship with data quality (i.e. you would not want to do one without the other) and Master Data Management initiatives are just not successful if data governance is not implemented at the same time.

Also data governance’s relationship is not just limited to those two data management disciplines - earlier on I said that it is not Data Security or Data Retention but a Data Owner (a key Data Governance Role) is usually the person who is best placed to agree who should/should not have access to their data and to agree how long it should be held for (how long it should/can be held for of course needs input from your Legal or Compliance Teams).

I don't think this issue is going to disappear overnight, after all even data management professionals can't agree on the definition. Over the years I have had a number of discussions with a well respected Data Management author who despite my best endeavours still firmly believes that Data Governance is a term interchangeable with Data Management!  We must however all strive for consistency in the use of the term and it's meaning.  That will make life easier for all of us.  There will be one less hurdle in selling and implementing the concept if there is more of a common understanding of what it is in the first place.

What do you think?  Are there are other reasons that I've missed and what do you think we as a profession can do to help clarify the situation and get a clear consistent message of what we do, more widely shared?

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

Data Governance Interview - Garry Manser

Another week, another Data Governance Interview and coincidentally another Garry!  This time I was fortunate that Garry Manser agreed to be interviewed.  When I first met Garry his enthusiasm for Data Governance was obvious as well as his determination to succeed in implementing it.  Read on to find out more…

Garry has worked in financial services for over 28 years in a number of different roles. He has been involved with data since 1998 and was introduced to the worlds of governance and quality in 2005. During his career he has worked both in industry and consultancy, across both banking and insurance.

Achievements include introducing and supporting a number of:

·      Successful governance frameworks across organisations both at a local level and on a global scale.

·      Quality initiatives, from the initial monitoring, through root cause analysis and onto remediation, with one role involving introducing circa 200 front end controls to improve data quality at capture.

He is currently involved in embedding a data governance and stewardship network across an international financial services provider, in support of various regulatory requirements and a driven business wide desire to succeed.

How long have you been working in Data Governance?

10 years I began with a Data Quality role in 2005 but quickly realised the benefit of bringing the two disciplines together.

How did you start working in Data Governance?

I moved from heading up an analytics team for a major US credit card distributor into an Operational Risk role for a UK Bank. This was looking at the data used in the Basel II risk models and it became clear early on that the quality of data in the models was an important part of the role. A conference and a course later and I was leading a team delivering initially data quality monitoring reports but then remediation, both supporting a team manually and by introducing process changes and front end controls.

What where your initial thoughts when you first fully understood what you had got into?

My initial thought was that this was an easy role, I had a regulatory requirement that needed to be met, I had the support of people above me and I was fresh from an Information Management seminar. Then I found myself being asked to leave a meeting I was presenting at, in a very impolite manner, and I realised that this was going to be tougher than I had imagined. It took me 18 months to get back around that meeting table, by then I had built some success in the organisation and had changed the way I opened my pitch. The positive that came from that was that the person that asked me to leave became one of the top champions of what we were trying to do.

Are there any particular resources that you found useful support when you were starting out?

My first data seminar was hosted by Larry English and I continued to work with him for about two years afterwards, his book “Improving Data Warehouse and Business Information Quality” is still on my desk today. People wise, listening to people like Nigel Turner and Daragh O’Brien speak gave me the confidence that people were delivering solutions. I found the IAIDQ website at the time very useful as well.

What is the biggest Data Governance challenge you have faced so far?

There have been a number of challenges over the years. On one occasion I was doing some initial consultancy for an Asset Servicer which included supporting them in setting up their Governance structure and identifying Data Owners. However due to internal rules we were not allowed to recommend anyone for a role and had to spend time compiling interview questions and then interviewing each senior manager in the supply chain to “help” the right person identify they were the person for the role. Was very satisfying at the end and each person identified had made the connection and understood where they were going. One that wasn’t as successful was writing a data policy and associated standard for a company and the first feedback received was that both documents were 75 pages too short as the company expected policies to be at least 100 pages!

What have you implemented or solved so far that you are particularly proud of?

Introducing circa 200 controls into a bank front end system during 2008. We knew it was the right thing to do, but funding was not available for changes to the system. However one IT colleague found a solution that allowed us to put them into a web front end that mirrored the tool. We put an initial 30 controls in place over a weekend and then came in really early on the Monday to ensure the system still worked. Within days we were being asked to introduce more illustrating the success. It’s also great to see data Governance assignments being extended upon their initial remit, celebrating the success of the work of the people involved. The people side of Data Governance is often forgotten but it just takes one person in an organisation with the right spirit to make a difference. 

What single piece of advice would you give someone just starting out in Data Governance?

Take your time and remember this is delivered by people, not by policies.

Finally, what do you wish you had known or done differently when you were just starting out in Data Governance?

I wish I had taken time to understand people more and how they would support delivery of both the governance and quality frameworks. I started with regulation, and then policy then eventually talked to people. As I did that I was able to see that anyone was willing to support the initiatives once they understood what the benefit was to them.

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

Data Governance Interview - Garry Ure

Continuing with the Data Governance interviews, I was very pleased that Garry Ure agreed to share his experiences in this blog.  He is indeed one of the generous members of the Data Governance Community that he mentions in this interview, always happy to share advice and experience.

Garry is currently Head of Data Governance for a leading bank in the UAE. Prior to that he worked in the UK for various banking and financial institutions and in various data management roles.

How long have you been working in Data Governance?

I’ve been working in Data Governance in some capacity for just over 11 years; all of which have been in the finance sector.

How did you start working in Data Governance?

I sort of fell into a data management role straight from university (where I’d actually studied photography) and found the subject interesting and challenging enough to turn it into a career.

That first role was actually as part of a support function responsible for the development and maintenance of a ‘single customer view’ as part of a Basel II programme. It gave me first hand experience of a number of Data Governance aspects, from issues of ownership and accountability through to the impacts of poor quality data on different parts of the business.

What where your initial thoughts when you first fully understood what you had got into?

I think I’m still trying to understand fully what I’ve gotten into! I find myself discovering new challenges on a daily basis; whether it’s how to best support the latest business strategy or how to satisfy a new regulatory imperative. There’s no real end to things that need to be understood and given the speed of change in some industries it can be detrimental to make the assumption that you do fully understand something.

Saying all that, my recurring thought is usually “I’m going to need a bigger team” when I see the list of things to do!

Are there any particular resources that you found useful support when you were starting out?

I did a lot of research online. I got to know who the recognised thought leaders were and bought their books and pretty much stalked them on social media! In the early days I think just asking questions, observing and learning from mistakes were important for me. I always found it useful to have a mentor of sorts in my organisation; someone that was well experienced and could offer support that was not necessarily DG related, but could help me manage the various stakeholders and situations that inevitably come up.

These days people are spoiled for choice in terms of useful support materials; there are various organisations that provide high quality publications on a regular basis and I now struggle to keep up to date with the numerous blogs, white papers, and webinars on the subject of Data Governance.

What is the biggest Data Governance challenge you have faced so far?

I think my current role has been my biggest challenge so far. Obviously relocating as a family from the UK to the Middle East has it own set of challenges but combining these personal challenges with those associated with building a Data Governance programme from the ground up in a relatively immature (in a DG sense) environment is a real test.

What have you implemented or solved so far that you are particularly proud of?

Again I think it is my current role is giving me a lot of pride on a regular basis. I think moving the company from zero formal Data Governance activity to a point where we are beginning to see genuine business understanding, interest, and engagement is a real achievement. We have just recently implemented a best in class Data Governance tool and I’ve been encouraged to see the excitement shown from the business at the possibilities it offers us. Hopefully we can keep up the momentum!

What single piece of advice would you give someone just starting out in Data Governance?

I think it’s hard to pick just one piece of advice. But I think I’d stress that I find the Data Governance community in general to be a very generous bunch both in terms of their time and their knowledge; there are so many channels now to have direct contact with some of the best thought leaders in the industry. I’d recommend networking both virtually and in person; get on social media and get along to whatever events are happening near you.

Finally, what do you wish you had known or done differently when you were just starting out in Data Governance?

I’d have made fewer assumptions, asked many more questions and worked harder on ensuring continued business engagement. 

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

Data Governance Template?

Of all the various questions I get asked, one of the most frequent is "Do you have a template for a data governance strategy that I could have?"

When I respond that I don't, I sometimes get the impression that the individual doesn't believe me and thinks that I am being deliberately unhelpful.  But the truth is that I don't advocate using a template for creating a data governance strategy or copying other examples, so I don't keep any.

The reason for this is that for a data governance strategy to be successful it has to be aligned to the overall strategy of the organisation that it is for.  By this I mean that it has to support achieving that strategy. It needs to be specifically written for each company with reference to its strategy and goals.

No two companies are identical (not even within the same industry) so you would not want to even rework or copy other examples.  Getting your data governance strategy well defined and aligned with your corporate strategy is a key foundation for successful data governance and if you "cheat" this step, it is likely to have negative impact on the resulting data governance initiative and the benefits achieved. 

Understanding how data governance will help your organisation achieve its strategy (or in the very least not hamper or distract from achieving it) is a key step to implementing data governance successfully, but it is only one part.  If you are interested in learning more about how to develop your own data governance strategy and how to successfully design and implement a data governance framework, I am running my Getting Started in Data Governance course in London on the 25th March.  In this one day workshop we will work through my proven methodology for implementing data governance. As an added bonus all attendees on the course will get a free Data Governance Coaching call, enabling them to start working out how to apply my methodology and framework in their organisation.  For more details and to register for the course click here or feel free to contact me if you have queries about the workshop.

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report

New Year, New Interviews

Firstly I'd like to start off this short blog post by wishing all my readers a happy New Year!  The beginning of each year is traditionally a time when people take the opportunity to plan their activities for the year ahead, so I do hope that you are all planning to implement, embed or improve data governance in your organisations.

I too have been planning for the year ahead and I’m considering some interesting developments for my business, but rest assured that although I sometimes struggle to find time to blog as frequently as I would like, my blog has not been overlooked in my planning for 2015. 

I have received such great feedback on the Data Governance Guru interview series and I will of course be continuing with that, but I'd also like to take a slightly different tack and include interviews with Data Governance Practitioners i.e. those on the ground doing data governance for their organisation.

I know from the conversations that inevitably arise when I run courses and tutorials, as well as in discussions at conferences and networking events, that anyone who has been involved in data governance has valuable experiences and lessons they have learnt that would be of benefit to others in the same position.  I have a few people in mind that I would like to interview, but if you think that you have an interesting story or case study that you are happy to share please do get in touch.

I'm looking forward to some interesting interviews this year and wish you all good luck in your data governance ventures in 2015!

My free report reveals why companies struggle to successfully implement data governance. Discover how to quickly get you data governance initiative on track by downloading this free report