How To Identify The Right Data Owners?

One of the items on my free Data Governance Checklist is “Define roles and assign responsibilities”. Now that is a generic statement that covers a number of different roles, but the role which I always start with (and believe that you should too) is that of Data Owner. By that, I mean the senior individuals in your organization who are accountable for the quality of one or more data sets. And don’t get worried if you have decided to call that role something else – that is perfectly acceptable and the naming of roles is something that I covered in this old blog: A Rose By Any Other Name.

At the Data Governance Clinics that I run one of the most frequent questions is how you find the right people to be the Data Owners in your organization. Anyone who has ever tried to implement Data Governance understands how important it is to have the correct people in these roles, but it can be hard to actually identify them. So I thought I would share a simple approach that I use for identifying the correct data owners:
I like to start by looking at the various departments and their relationship with a particular dataset. If I was trying to find the Data Owner for customer data, I would start by finding out which business area feels the most pain when customer data is wrong. I ask lots of questions like:

· Who cares when the data is wrong?
· Which team is likely to be the first to identify data quality issues with customer data?

Some people believe that the Data Owner should always represent the area that captures or enters the data. In my experience, this is sometimes the case, but you should not rule out the possibility that the owner may sit somewhere else within the organization.

Wherever they are, they must have an interest in the data, but this can be either as a data producer or as a data consumer. If they are neither of these, it is unlikely that they have sufficient interest in the quality of that data to undertake the role properly.

Once you have identified the department that seems to have the most interest in the data, you can then identify the individual best suited to take on the role of Data Owner. Remember that for them to be able to improve the quality of the data, the candidate Data Owner needs to be suitably senior and have resources at their disposal. So identify the individual in that area who has:

· The authority to change business processes and IT systems to improve data quality
· Access to budget and resources to be able to resolve data quality issues
· The ability to instigate data cleansing activities.

If you find the person who fits these criteria, they are very likely to be the right Data Owner. This approach has always worked for me, although sometimes it can take you in unexpected directions. For example, in one Personal Lines Insurance Company when I was trying to identify the Data Owner for Customer Data I ended up following this route:

First I approached the Underwriters – after all, they decide who gets an Insurance Policy and what data is needed on the customers for that decision to be made. However, they explained to me that as their company was a high volume personal lines firm, they did not get involved with individual policies and had no interest in specific data about individual customers.

Next, I tried the Service Department. These were the teams of people who speak to customers on the telephone and enter their details on the system. But again they had no real interest in the quality of the data. They did not decide what gets captured, nor did they use the data so did not feel any pain if it was wrong.

Finally after asking lots of people who used Customer data and who cared if it was wrong, I found myself meeting one of the Marketing Directors. I didn’t hold out much hope that they would be the Data Owner of Customer Data, but it turns out I was wrong. In that particular company, the Marketing Department was responsible for sending out renewal letters to customers. If the customer data was wrong the renewals did not reach the customers and there was a strong possibility that business would be lost. As soon as I explained that I was trying to identify the Data Owner for Customer Data, they immediately agreed it was them as they had such an interest in the quality of that data.

I hope this will help you identify the correct Data Owners for your organization, but remember that just because you have worked out who it should be, it does not mean that they will necessarily agree. Your next activity will be to practice your influencing and communication skills!

Remember that finding the right Data Owners is only one of the items on my free Data Governance Checklist, you can download the checklist here to see what other activities you need to be doing to implement Data Governance successfully.

Do You Need To Prioritise Your Data?

This year I've tried to stick to a regular schedule of creating and posting blogs. However, as many of you will have noticed, I failed abysmally a couple of weeks ago. A number of family illnesses and emergencies meant that I was unable to work as many hours as usual. And my blog writing time was spent looking after my husband and other family members.

It wasn't just my blog writing activities that got curtailed.  I am very grateful to my clients who very kindly agreed to reschedule planned workshops. They were fantastic and understood that I had to prioritise my family over my work for a short period of time.

What amazes me, though, is that so few people think to take the same approach to their data. After all, is all of your data equal in value? Would it be fair to say that some is “more equal than others”?

What I mean by this is that not all of your data needs to be as proactively managed, monitored and controlled. In fact, to do so, would make data governance a burden or a hurdle to people actually carrying out their day to day activities. This is never the point of data governance. I believe that the point of data governance is to identify the most important data and manage that data proportionately to the value it has to your company.

I first came across this concept when working in the insurance sector. One of their regulations is Solvency II.  It primarily deals with the capital adequacy of insurance firms, but at the same time asks for data governance to be put on place over all data being used in the capital adequacy calculations. However, the regulator realised a really important point, in these rather complex calculations, some data is very important, and other data is included just for context. Now, if that latter data is wrong or missing, it would have either no or a negligible impact upon the final calculation. Therefore, the regulator said they didn't expect you to put the same level of data governance in place over that data as opposed to the data which is really important and would actually result in the significantly wrong numbers being calculated.  

As soon as I started trying to work this out for the first insurance company I worked with, It made sense to me. I quickly realised that focusing your efforts on the most important data was the right thing to do for data governance. Since that time, I have encouraged every client, regardless of which sector they operate in, to adopt this approach.

 Solvency II gave a name for this approach - materialityi.e.  it is about identifying your most material data and managing that appropriately.  However, be aware that calling it “material data” may not work for you. In fact, I was told in no uncertain terms by a German manufacturing client of mine that the term materiality most definitely does not work if your company uses materials to make something as material data means something else entirely in that context!    

We agreed that this would cause confusion rather than be useful. And for that client, and many others since, we have chosen to call it critical data.

Identifying your critical or material data is a very sensible and pragmatic approach, but not necessarily an easy one. You will need to define some criteria for what each level of criticality means so that Data Owners can assess the data that they own against the criteria and decide whether it is critical or not. 

There is also the interesting question of how many levels of criticality you need?

 My preference is for three levels:

High criticality or high materiality is the data that is the most valuable to your business and would have the most negative impact if it was of poor quality.

Medium criticality or medium materiality is data that is important, but would not have such a large impact if it was of poor quality.

Non-material or Non-critical data is the data that is useful, and perhaps adds context but would not cause great problems if it was not of the best quality.

Over the years, some clients have preferred to go for just two levels of criticality i.e. it either is critical or it isn't. But that feels a bit too much like an all or nothing approach. Data either has lots of controls, standards, data quality monitoring and reporting in place, or it has nothing. 

One client, asked me to implement five levels of materiality. I'll be honest, I really struggled to differentiate between the different levels of data governance that you would apply across five categories and ultimately, we rationalised it back to three.

Whatever you call it, and however many levels you decide is right for your organisation, I really would encourage you to try this approach in your data governance initiative. Anyone who regularly reads my blogs will know that I advocate a pragmatic approach. You really can't manage all data perfectly. So why not identify the data that is the most important to your organisation, and manage that appropriately?

Taking such an approach makes it far easier to manage the workload, particularly when you're in the early stages of a data governance initiative. So If you look at the activities on my free data governance checklist, you might decide that some of this needs to be done for all data. But some of the activities you may only wish to do for things that are high or medium materiality data. In this way, you can agree a phased approach of appropriate management on your data according to how important it is to your organisation.

Why You Need Data Governance

In this blog I’m going to look at why you really should do data governance. When I tell people what I do, I get a mixed response. Some people seem genuinely surprised that everyone isn’t already doing Data Governance, and an awful lot of people ask why would you need that?

Now I’m biased, as I believe that every organization would benefit from implementing data governance. It may not solve all problems, but it really does provide a framework which can be used to proactively manage your data.

A few years ago the main driver of Data Governance initiatives was regulatory compliance and while that is definitely still a factor, there is a move towards companies embracing Data Governance for the business value which it can enable. For example if your organisation is starting a digital transformation or wants to become “data driven”, you are not going to be successful if your data is currently not well understood, managed and is of poor quality.

If you embrace Data Governance and achieve better quality data, all sorts of benefits start to be seen. But you don’t have to take my word for it; take the DAMA DMBoK Wheel for instance: 

As you can see, it lists all the Data Management disciplines around the outside of the wheel. There in the middle, at the heart of it all, is Data Governance.  Now it didn’t just get put in the middle because there were no more spaces on the outside of the wheel – it’s there for a reason. Data Governance provides the foundation for all other data management disciplines.

Let’s look at a few of these disciplines to illustrate the point:

Data Quality

Without Data Governance all data quality efforts tend to be tactical at best. This means a company will be constantly cleaning or fixing data, perhaps adding default values when a key field has been left blank. With Data Governance in place, you will have processes, roles, and responsibilities to ensure that the root causes of poor data quality are identified and fixed so that data cleansing is not necessary on an on-going basis.

Reference and Master Data

Anyone who has been involved in any master data projects will have no doubt heard or read numerous dire warnings about the dangers of attempting these without having Data Governance in place. While I am not a fan of wholesale scaremongering to get people to embrace Data Governance, these warnings are genuine. For master data projects to be successful, you need data owners identified and definitions of all the fields involved drafted and agreed, as well as processes for how suspect matches will be dealt with. Without these things (which of course Data Governance provides) you are likely to be faced with a mess of under, over or mismatching!

Data Security

Of course Data Security is primarily an IT managed area, but it makes things a lot easier to manage consistently if there are agreed Data Owners in place to make decisions on who should and should not have access to a given set of data.

I hope you agree that these examples and explanations make sense, but don’t forget that is theory; and explaining this in data management terms to your senior stakeholders in order to get agreement to start a Data Governance initiative is unlikely to be successful. Instead, you are going to need to explain it in terms of the benefits it will bring. The primary reason to do Data Governance is to improve the quality of data.  So the benefits of Data Governance are those things that will improve, if the quality of your data improves.  This can cover a whole myriad of areas including the following:

Improved Efficiency

Have a look around your company. How many “work-arounds” exist because of issues with data? What costs could be reduced if all the manual cleansing and fixing of data were reduced or even eliminated?

Better Decisions

We have to assume that the senior management in your organization intends to make the best decisions. But what happens if they make those decisions based on reports that contain poor quality data? Better quality data leads to more accurate reporting.

Compliance

Very few organizations operate in an industry that does not have to comply with some regulation, and many regulations now require that you manage your data better. Indeed, GDPR (the General Data Protection Regulation) impacts everyone who holds data on EU Citizens (customers and employees), and having a solid Data Governance Framework in place will enable you to manage your data better and meet regulatory requirements.

So, at this point you are probably thinking, “isn’t it just a generic best practice thing that everyone ought to do?” And the answer is, yes – I do believe that every organization could benefit from having a Data Governance Framework that is appropriate for its needs.

What Happens if you Don’t Have Data Governance?

Well I’ll leave that to you have a look around you and decide what the likely consequences for your company could be, but it is usually the opposite of the benefits that can be achieved.

Remember data is used for dealing with your customers, making decisions, generating reports, understanding revenue and expenditures. Everyone from the Customer Service Team to your Senior Executive Team use data and rely on it being good enough to use.

Data Governance provides the foundation so that everything else can work.  This will include obvious “data” activities like Master Data Management, Business Intelligence, Big Data Analytics, Machine Learning, and Artificial Intelligence.  But don’t get stuck thinking only in terms of data.  Lots of processes in your organization can go wrong if the data is wrong, leading to customer complaints, damaged stock, and halted production lines. Don’t limit your thinking to only data activities.

If your organization is using data (and to be honest, which companies aren’t?) you need Data Governance.  Some people may not believe that Data Governance is sexy, but it is important for everyone.  It need not (in fact it should not) be an overly complex burden that adds controls and obstacles to getting things done. Data Governance should be a practical thing, designed to proactively manage the data that is important to your organization.

Just one final word of advice: I hope that this article has convinced you that your organization needs to embrace Data Governance; but if that is the case, please don’t just spout the generic benefits and examples I have shared here in your efforts to gain stakeholder buy in. It is very important to spend time working out the specific reasons your company should be doing Data Governance. You can find more advice on that and how to engage your senior stakeholders here.

Does it have to be called Data Governance?

This is a question that I get asked fairly regularly. After all it is not an exciting title and in no way conveys the benefits that an organisation can achieve by implementing Data Governance. Sadly however, there is no easy yes or no answer. There are a number of reasons for this:

1. Data governance is a misunderstood and misused data management term

Naturally I am biased, but in my view, data governance is the foundation of all other data management disciplines (and of course therefore the most important). But the fact remains that despite an increasing focus on the topic, it remains a largely misunderstood discipline.

On top of this, it is a term which is frequently misused. A few years ago, a number of Data Security software vendors were using the term to describe their products. More recently the focus on meeting the EU GDPR requirements has led to a lot of confusion as to whether Data Protection and Data Governance are the same thing and I find that the terms are being used interchangeably. (For the record, having Data Governance in place does help you meet a chunk of the GDPR requirements, but they are not the same thing).

Having more people talking about Data Governance is definitely a good thing, but unless they are all meaning the same thing, it leads to much confusion over what data governance really is.

I explored this topic in a bit more detail in this blog: Why are there so many Data Governance Definitions?

In order to understand whether Data Governance is the right title for your organisation to call it, I would start with looking at how you define data governance. And this step leads nicely to the next item for consideration.

1. Sometimes it is right to include things which are not pure data governance in the scope of your data governance initiative.

This is a topic that I covered in my last blog which you can read here.

To summarize that article, it is just not possible to have one or more people focus purely on Data Governance in smaller organisations. It’s a luxury of large organizations to be able to have separate teams responsible for each different data management discipline (e.g. Data Architecture, Data Modelling or Data Security).  Going back to my point above, if data governance is the foundation for all other data management disciplines, it is only natural that the line between them can sometimes get a little blurred. As a result of this, the responsibilities of the Data Governance Team can get expanded.

So consider what is included within the scope of your data governance initiative and decide whether it be more appropriate to name the initiative and your team (either or both)  something that is more aligned to the wider scope of the initiative and activities of the team.

Is the name going to make cultural change harder to achieve?

Achieving a sustainable cultural change is one of the biggest challenges in implementing data governance and insisting on calling it “data governance” could make achieving that cultural change more difficult if the term doesn’t resonate within your organization. This is related to a topic that I explored in another old blog Do we have to call them Data Owners?

Whether we’re talking about the roles, the team, or even the initiative the same principles are true. It is better to choose a name that works for the culture in your organization than to waste considerable effort trying to convince people that the “correct” terminology is the only one to use.

It would be my preference to explain that the initiative is to design and implement a Data Governance Framework, but if the primary reason for implementing data governance is to improve the quality of your data, perhaps calling it the “Data Quality Team” and “Data Quality Initiative” would fit better? After all, that very much focuses on the outcome of what you’re doing.  It also addresses the question that everybody asks (or should ask) when approached to get involved in data governance of “why are we doing this,” which is usually followed by “what’s in it for me?”

When having these conversations, I explain the initiative in terms of its outcomes (e.g. better quality data which will lead to more efficient ways of working, reduced costs and better customer service). That is a far easier concept to sell rather than implementing a governance structure, which can sound dull and boring.

Is the name causing confusion?

In the early days of a data governance initiative, the talk is all about designing and implementing a data governance framework. Once this work has been achieved you start designing and implementing processes which have “Data Quality” in their titles:

• Data Quality Issue Resolution

• Data Quality Reporting

I have been fortunate enough to work with organizations in the past who have had both a Data Governance Team (supporting the Data Owners and Data Stewards) and a Data Quality Team (responsible for the processes mentioned above) but that is fairly unusual in my experience. It is more common for the Data Governance Team to support the above processes. So it is worth considering whether it would confuse people if they had to report data quality issues to the Data Governance Team?

In summary, I would not want to miss the opportunity to educate more people on what Data Governance really is. But the banner under which it is delivered can be altered to make your data governance implementation both more successful and more sustainable. So if having considered all the points above in respect of your organization and you want to call it something else, then that is fine with me.

Deciding what to call your initiative is only the start of many things you need to do to make your Data Governance initiative successful.   You can download a free checklist of the things you need to do here. (Don't forget this is a high level summary view, but everyone who attends either my face to face or online training gets  a copy of the complete detailed checklist which I use when working with my clients.)

What should you include in a Data Governance initiative?

One of the many challenges you will have to face when implementing Data Governance is agreeing the scope of the initial phase of your initiative. By this I don’t just mean which data domains or business functions are going to be in scope. I’m thinking of associated activities like data retention, end-user computing, and data protection. Being a bit of a Data Governance purist I maintain that such activities are most definitely NOT data governance. It is easy therefore to make the logical conclusion that they should not be in the scope of your initiative. So what I say next may surprise you:

Do not immediately go on the defensive and refuse to take any (or even all) of these activities into the scope of your initiative!

Now you may be wondering why someone who spends her time educating people on what Data Governance is would say that! Well, when I’m training and coaching people it is important that they understand what Data Governance is, but when I’m implementing Data Governance in practice, I take a pragmatic approach.

However, I would not want you to think that I would just say yes to an ever-expanding scope. There are a number of factors that would make me consider bringing these additional data activities into the scope of my data governance work, which include:

• If you work for a small organization that does not have the luxury of separate specialist teams to cover each data management discipline;

• If they overlap with other projects ongoing at the same time;

• Or if a senior stakeholder requests it.

Whilst you may become aware of other activities that you want to bring into scope, they are most likely to come to your attention through your senior stakeholders – so let’s consider this question:

How do you manage senior stakeholders who ask you to extend the scope of your initiative?

Now whilst it may be tempting to protect the scope of your initiative, remember they have their own agenda. They are not trying to derail your plans, they just have concerns of their own or issues that they need addressed. The first thing you are going to need to do is to listen and understand what their concerns are before you try to educate or influence them. After all, how can you properly allay their concerns if you don’t fully understand them?

But remember whilst it is imperative that you understand why they’re asking you to extend the scope, when I say educate or influence them, I don’t mean your initial stance is to say no! When talking to your senior stakeholder, ask lots of questions and constantly consider the following:

• What exactly does this person need done?

• Does it have any alignment or overlap with your data governance work?

• What will happen if this additional work does not get done? (And in particular will it cause a problem for your data governance initiative?)

Even if the answer to this last question is no, it may still be necessary for you to consider that if you say no, that this senior stakeholder could divert resources currently allocated to your initiative to address this other issue.

Are there benefits and/or efficiencies to be achieved by taking on this work? This can be especially true if you are talking to the same stakeholders.

My advice is to look for solutions that help everyone. This is not about you or them winning. This is about doing the right thing for your organization. Find out why he/she is concerned about these other topics. Is it because they are not being done, or is it that they are being done but are not visible or are being done but not well enough or quickly enough?

Now obviously I’m biased, but I truly believe that well implemented data governance can be the framework against which you align an awful lot of other activities in your organization (well at least those concerning data)! Once in place, you can use your data governance framework to coordinate, oversee, and escalate other data matters to the appropriate people. That said, it is not the answer to everything and you should resist taking on everything (unless of course you are Superman/Superwoman), or at least agree to timescales for adding additional scope once the implementation of your data governance framework has reached a certain stage.

If you do take on something that perhaps you feel is not in the area of your expertise, that is ok – just be honest and clear on the matter. Explain that whilst, for example, you may not be a data retention expert, you see how including that in your data governance initiative has benefits for the organization. Confirm that you are happy to do the necessary research and support the work if you are given the necessary expert support (for example from your Legal Department).

Remember that whether your data governance initiative is small and focused or has gained additional scope, stakeholder engagement is absolutely vital for success. You need to spend a lot of effort engaging your stakeholders. If you could lose their support by not addressing their other concerns, it’s got to be worth considering whether the additional work is something that you can take on.

Finally, if you want ideas on how to go about engaging your stakeholders, you can download my top tips on stakeholder engagement for free if you click here.

Originally posted on TDAN.com

Data Governance Interview - Bonny McClain

I haven’t done a Data Governance interview for quite a while, but while preparing for a webinar I am doing with Bonny on Data Governance in Healthcare, it became clear that I had to ask her to do an interview as she has so much expertise to share.

Bonny curates data from the intersection of health policy, health economics, and healthcare to create powerful storytelling narratives. Real insights come from the ability to hold tensions and bring multiple data sources to the conversation.

The data revolution is here, and her expertise is tackling industry specific problems-- rendering them solvable with data-- relying on a wide variety of tools like Python, R, SQL, and Tableau data visualization.

A big advocate of data literacy Bonny is a life-cycle data consultant. The ability to appreciate the overall concept while simultaneously thinking about detailed aspects of implementation allows different levels of abstraction to be curated creatively and empathetically.

How long have you been working in Data Governance?

Since I first began working with electronic medical record (EMR) systems. Working with relational databases it became painfully obvious that data systems and data assets were not being managed across their lifecycle. Data quality issues were impacted by non-existent information governance often exacerbated by the move from paper records and charts to electronic databases.

Some people view Data Governance as an unusual career choice, would you mind sharing how you got into this area of work?

In smaller community medical practices and even larger health systems—information was being generated and shared across brick and mortar institutions. As physicians became curious about data questions that extended beyond administrative concerns there was a palpable need to understand “data dictionaries” and the schema and architecture of data storage. Understanding data assets was a natural evolution and requirement to glean insights from the digital data being created.

What characteristics do you have that make you successful at Data Governance and why?

Because I work with the populations that rely on the quality of data—the emphasis is on usability but not at the expense of safety—I have the data analytic skills (recent certificate in applied data analytics from Columbia School of Engineering) to identify which measures and variables are needed to answer a data question. Knowing that pieces of information often live in different parts of a database generates concern and an evolution of skills to ensure that patient records are matched to avoid duplicate records, missing data, inappropriately merged records, current medications and procedures—all with an eye to seek out potential opportunities for harm if data is incomplete, incorrect, missing, or low value.

You work a lot with the Healthcare Industry – how mature would you say they are in Data Governance?

You should mention that I snorted and began laughing. Although in full disclosure I work with many smaller organizations that haven’t been able to prioritize data governance for one reason or another. Our conversations about many data governance policies being articulated as “best endeavors” or a best effort illuminated for me why many of the best intentioned guidance falls short without out a complete and accountable data governance strategy.  

I do stress scalability with new clients so they don’t feel overwhelmed and tempted to park the whole process until “later”. I meet clients where they are—and most make significant process relatively quickly. Healthcare is unique as bad data here—can lead to deleterious outcomes and harms.

 The typical data client actually has low data literacy and maturity and often answers a data question before the analyses have been initiated. They want data that shows what they actually believe to be true—and in the face of a contrary outcome—move on to the average analyst that will gladly only report the curated outcome they seek.

Given that this is the typical healthcare professional—you can imagine that governance strategies that sit upstream from information governance are not recognized or prioritized.

How clear do you believe the Healthcare Industry as a whole is on the difference between Data Governance and Information Governance?

I may be wrong but I feel the labels are not applied to the different behaviors correctly. Although I do believe that compliance, quality care (value), cost containment, evolving payment models, and safety are understood as information—and rely on proper custodial processes of information assets--the industry as a whole is at a crossroads of how data governance influences the rigor of information downstream from policies and processes. 

What’s the difference between Data Owners and Data Custodians?

I often get asked about the difference between Data Owners and Data Custodians.

 If you've read my other blogs, you'll know that I'm not fixated on sticking rigidly to the standard role names, but for the purposes of this blog, I’ll to stick with what I consider “best practice” role names and consider both roles in turn:

A Data Owner is a senior business stakeholder who is accountable for the quality of one or more data sets. They are usually a senior business person who has the resources, budget and authority to be able to make changes to that data if necessary. 

Data Custodians are very much an IT role. They are responsible for maintaining data on the IT infrastructure in accordance with business requirements. I think the confusion between the roles and who should be making decisions about data is rooted in a long term lack of Data Governance.   

Before a company has Data Governance in place, it's common that the business has not been trained in articulating their data requirements and therefore it is often down to IT to interpret or make decisions in order to help the business.  However, once you have a Data Governance framework in place, the business usually gets much better at articulating data requirements and IT’s job gets easier. 

Another difference between the roles is that for Data Owners, I'm looking for named individuals who own one or more data sets. When it comes to Data Custodians. However, I use that more loosely as a collective term for all of the IT department who are supporting your infrastructure.  That said, it's not impossible to have named custodians. I've worked with a number of clients where they have named their Data Custodians.  These have usually been smaller organisations, where there is only one subject matter expert for each system and that person has been named as the Data Custodian.

Another frequent question is whether IT is ever a Data Owner? Now this is an interesting questions as generally, I would say, no. They might own meta data, or performance data around the systems, but nothing more than that. However, recently, I have come across circumstances where IT might own some data. 

One such example is maybe your Data Security or Information Security Team who may be monitoring telephone calls or internet activity. This often isn't being done for any direct business purpose, but rather to protect the business as a whole. If this data isn't being collected to meet the requirements of a business Data Owner, then it could easily be argued that IT own that data.

I also often get asked if only IT can be Data Custodians?  Generally, it is the case that Data Custodians sit within IT, but there are instances where business teams or functions may also act as Data Custodians. For instance, you may have a BI or Analytics Team sitting in the business reporting line, who perhaps manage and support your data warehouse.  Because of the work they do it is quite common for such a team to be considered as the Data Owners for all data that's in the data warehouse. However, this is not a correct approach, because all that data has a Data Owner when it sits in the source system and that person still owns it even when it's on a data warehouse. Of course, in a data warehouse a lot of data is aggregated, combined and often new data is derived or calculated. It is common for the BI or Analytics Team to be involved in these activities, but it does not mean that they own the new or aggregated data. It is more appropriate to consider them as Data Custodians for the data, whilst it's being managed, manipulated and reported on by that team.  After all they should be carrying out these activities to meet business requirements and the person who approved those requirements is the Data Owner.

I hope that this has clarified the difference between Data Owners and Data Custodians.  Roles and responsibilities are only one of many things that you have to address as part of implementing Data Governance. If you are struggling to get your head around everything else you should be doing and the order in which to do it, please download my free checklist to help you plan your initiative.

Data Quality Issues - Who Is Responsible for Resolving Them?

One of the first processes, I believe that you should introduce in your Data Governance initiative is a Data Quality Issue Resolution process. In fact my last blog covered what you should include in a Data Quality Issue Log to help you get started with collating and resolving issues.

But the log itself is only part of the answer and I have been asked on numerous occasions to clarify what the Data Governance (or Data Quality) Team are responsible for doing when managing and resolving data quality issues.  This often gets asked by newly formed teams.

There seems to be a lot of confusion around who does what and I have often come across the expectation that the Data Governance Team will do or solve everything.  There are times when I truly wish that I had a magic wand and could simply fix all the data problems, but sadly that isn’t the case.  Over time, of course your Data Governance Team will develop knowledge and expertise about the data your organisation creates and uses, but they are not responsible for deciding what the remedial actions should be and especially not for undertaking any manual data cleansing that may be required.

However, I am not saying that they have no part to play in the process.  The best way to understand what the Data Governance Team are responsible for is to look at a high level simple data quality issue resolution process:

Raise Data Quality Issue

It will usually be a Data Consumer (business user of that data) who spots an issue and will be the ones to notify the Data Governance Team.

The Data Governance Team will then log the issue on the Data Quality Issue log and identify the data owner(s) of the data concerned.

The Data Governance Team notifies the Data Owner of the issue, who will advise whether or not they are the correct owner of the issue.

In addition, the Data Governance Team reports the current status of all open material data quality issues to the Data Governance Committee (usually as part of their regular agenda).

The Data Governance Committee reviews the open material data quality issues and prioritizes/directs on the remedial activities if needed.

Impact Assess and Root Cause Analysis

The business user who notified the issue, the Data Governance Team, and the Data Owner(s) assess and agree about the impact of the issue.  If the issue is agreed to have a material impact its resolution will be prioritised.

The Data Governance Team works with the Data Owner(s) to identify the cause of the issue.

The Data Owner(s) consider possible remedial actions to rectify the issue.

Remedial Action Plan

The Data Owner(s) proposes an approach to resolve the issue and prevent it from re-occurring.

The business user who raised the issue agrees whether the proposed action plan is appropriate (the Data Governance Team can facilitate discussions between the parties if needed).

The Data Governance Team updates the Data Quality Issue Log with the agreed actions and target dates.

The Data Owner(s) plans how and when the remedial activities will take place.

Monitor and Report on Action Plans

The Data Owner(s) and their team(s) undertake the agreed remedial actions (N.B. this may need the support of IT).

The Data Governance Team monitors progress on remedial actions against agreed target dates and reports on progress to both the impacted business user(s) and the Data Governance Committee.

The impacted business user(s) advise if timescales for resolution are not appropriate.

The Data Governance Committee then reviews progress and prioritizes/directs on the remedial activities if needed.

Of course, in practice, solving data quality issues takes more than the four steps listed above, but the additional steps will be sub-sets of the stages discussed above.  In addition, keeping it simple like this will help your stakeholders quickly understand who is responsible for what.

As with all things Data Governance, communication is key. I would recommend creating a simple high-level diagram of your data quality issue resolution process and using that in your communications to help people not only understand the process but also everyone’s role in it.

The Data Governance Team’s role in the data quality issue resolution process can be summarized as followed:

• Identifying which Data Owner is responsible for the data which needs fixing and liaising with them

• Maintaining the Data Quality Issue Log

• Monitoring and reporting on open issues and associated action plans

I hope that has clarified the situation for you and remember you can find out more about what to include in a Data Quality Issue log here or download a free template for an issue log here.

Setting up the Data Quality Issue Process is just one of many things you need to do when starting a Data Governance Initiative - you can get a summary of all the things you need to consider by downloading my free Data Governance Checklist here.