Why Knowing Where To Start Your Data Governance Program Is a Critical Factor To Its Success

During my workshops, I often hear that whilst organisations want to start (or reinvigorate) a Data Governance initiative, they don’t know where to start.

I understand this.

There’s so much to cover that knowing the best place to begin is itself a major challenge which can lead to analysis paralysis. 

The IT vendors will tell you that the first thing you need to do is buy their tool.  After all, you can’t possibly begin to do data governance properly without a shiny and expensive tool set!

Then there’s the jargon that needs to be unravelled. 

·        CDEs

·        Lineage

·        Glossaries and Taxonomies

How do I know where I should focus my initial efforts?  I hear you say. 

Then there are those organisations that have taken the plunge and made a start.

The CDO team have asked the business to define their Critical Data Elements…but the business can’t answer this question. 

Or they have asked the business who should be data owners and stewards…to be met with stony silence or a “surely that’s someone else’s responsibility”.

Very often the business will nod and agree that data governance is important until it becomes time for them to play an active part.   

Just getting off the starting line is hard.

How do you overcome this inertia? 

Data Governance is Change Management

I’ve said this before, but it bears repeating. 

Data Governance isn’t done in a darkened room in isolation from the business whilst somehow magically being the panacea for solving data quality issues or meeting regulatory demands. 

Doing data governance in isolation from the business is why so many initiatives fail or become mere tick-box exercises that don’t deliver real, meaningful and lasting change.  

Asking people in the business if they would like to own data in the absence of bringing them step by step on a journey that leads to this point is futile. 

Asking them to define their CDEs if you’re not clear on what constitutes critical is also likely to set you up for failure. 

The business is either unsure and can’t name any, or they simply list out all data used by them regardless of context. 

The Journey

As with any journey then, knowing where to start is integral to ensuring you reach your destination. 

Before you start broaching roles and responsibilities, CDEs or flashy tools, there are certain steps you must take to be successful with your data governance initiative. 

Ultimately, you need to build stakeholder engagement by connecting your initiative to the business. 

I’m going to list some steps that I have consistently found to be invaluable to help you do this.

Organisational Strategy

This one is fundamental. 

You need to have a clear understanding of the direction that your organisation is going in.

You need to understand its short, medium and long-term goals. 

When I go into organisations and ask to look at the strategy, I’m frequently met with the response that it won’t help as “data” does not currently feature within it. 

But “data” will feature and increasingly, it will feature very heavily.  Whether explicitly or not.

So, you need to review the strategy.  For example:

1.        Are there specific financial goals?

2.        How does it intend to meet these goals?

3.        Does the organisation intend to diversify its core products in any way?

You need to ask what part data will play in these objectives. 

If the organisation is going to increase revenue, it will need to identify which products or countries/regions it intends to focus on. 

If so, does it already know or will it require data to make this determination?

Again, if it intends to diversify its core products, does it already know which products it will be focusing on?  If not, it will require data to do so.

If it intends to cut costs, how will it decide where to cut?  How will it be monitoring spend?

Does the organisation have a way of measuring how it’s doing against its strategic objectives?

What about the KPIs?  Are these in place?

Are they uniform throughout the organisation?  I’ve worked in organisations where a simple KPI, say, “Revenue”, was calculated in one way by Sales and another way by Finance. 

Hardly conducive to being able to measure the progress of your strategic endeavours!

You can see how reviewing the Strategy and asking the right questions can give you a whole host of information about what data is important in achieving it and where there may be challenges along the way.

This is data that is vital to the business. 

There is no room for the business to be lackadaisical about it. 

Once you understand what this data is, where it resides and who produces and uses it, you can start to focus your initiative and build stakeholder engagement. 

Data Quality Issues

On to our next port of call. 

Is there a centralised list of DQ issues?

More often than not, such a list will not exist.  However, you will probably find that scouring an extract from the IT ticket management system will reveal at least some data quality issues. 

These will give you a sense of some of the issues that your organisation is experiencing.

It’s worth mentioning that these issues probably don’t actually sit with IT.  We all know that most data quality issues arise from broken processes and not broken IT systems.

A little bit of investigation may reveal that issues raised by different teams seem to originate from the same area or concern similar data.  They may have a similar root cause. 

Clearly, they are a source of pain in the business and can therefore be used as a meaningful point of engagement.

Stakeholder Mapping

You can see how just these two simple actions can help garner a meaningful and relevant dialogue in a language the business understands; not just data for the sake of data.

So having done your preparation, it’s time to actually start talking to the business. 

Who do you talk to?

Sometimes it’s obvious who your initial stakeholders are.  If most of your work is in one industry, you will likely know which functions are most reliant on data and therefore have a vested interest in its quality.

As a general rule, the functional areas to start with will be those sitting downstream of where most of the primary data is created.

Think about how the organisation works to determine the data flow.  If we take an insurer, most of the core data originates from the Underwriting and Underwriting Operations teams.  From here it is used by the Claims and Reinsurance functions, who then augment it before it flows downstream to the Finance and Actuarial teams for further usage and augmentation.

Talk to a business analyst if you’re unsure of how the business works.  Or you could use a technique like Business Activity Modelling (BAM) to help you.  More on that in another article.

Using this approach, you can build a comprehensive and detailed stakeholder map.  You’ll find yourself referring to this regularly and adding to it as time goes on. 

I always try to start with the functional heads and then work my way down to the individual team managers for greater levels of insight.

What should I be covering in my initial stakeholder meetings?

As well as the points gleaned from the steps above, these meetings should be used to delve into the following:

·        Processes: Core processes the Function performs

·        Data: Data used and where it comes from

·        Pain: Where the pain points lie

Gaining a deeper understanding of the pain points will help you to further build out your list of data quality issues as well as provide further potential candidates for some quick wins to solve.    

The Foundations for Sustainability

The groundwork detailed in this article is essential to equip you with the information you need to understand the data landscape within your organisation; specifically, it will yield lots of rich information about what data is important to run the business and, therefore, where you may want to start your initiative for maximum impact.

Not only that, but you will have started to build those all-important relationships that will help sustain your initiative as you move from planning and design to implementation.

In the next article, we’ll focus on delving down at a deeper level in your stakeholder conversations to tease out which data drives core processes and where it comes from.

Subscribe to get future articles in this series.

--

Need Data Governance help? 

Book a call to discover how we can support you.