Do you have control of your data?
How sure are you about that? As sure as the management teams of Target or JP Morgan before they were hacked, or the management of Lehman Brothers or Bear Stearns? These were all cases where failures in data management or policy played a part (sometimes a significant part) in the ultimate catastrophe.
Naturally, the catastrophes get the big headlines, but poorly controlled data is costly to organizations in a more subtle way. Great data is one of the ingredients that takes your product, business decisions, productivity and security from good to great, and without it you will most likely be another average competitor in the market. Of course if you have other reasons your product stands out from the competition then you can overcome that shortfall, but otherwise you may be destined to a life of mediocrity.
So if you step back and ask yourself that question again, what grade would you give yourself for how effectively the data that matters in your organization helps to take you from good to great?
Data Governance – over-stated and under-explained
“I have no idea what [Data Governance] is – it is too abstract for me”
Data Governance is the tool set we are told we need to get control of our data. But what does that actually mean? How do you get control of the data through some abstract concept? Is your reaction to the word the same at the user feedback above (a true quote)?
They say a picture is worth a thousand words but what if there’s a thousand different pictures? That’s my experience when I go and explore data governance – I’m presented with a hodge podge of governance schematics and varying definitions which fail to help me navigate the fundamental questions – like ‘What the heck is Data Governance’, “Why should I care” and “What are the benefits”.
So I’ve decided to throw my hat in the ring and answer those questions in layman terms – no pictures, just a bit of thinking it through and piecing it altogether from first principles.
If you’re a sceptic, stick with me a bit longer. Proper data governance doesn’t have to be cumbersome or expensive, and if your primary operating mode is “just do it”, I’d ask you to spend a few extra minutes to read on and think about some of the ideas here. Done well, it will help you execute better and in a more focused way.
Let’s tackle some of the basic questions.
What is Data Governance?
Data Governance is the system by which you govern data, just like the system of US governance is how the USA is governed.
Let’s use this analogy to help bring the Data Governance issue to life. So why do we need a system of governance for the USA?
The US political system is designed to do a few things, including:
- Ensure there is a process in place for maintaining and amending the guiding principles of the US system, embodied in the Constitution, the Bill or Rights and subsequent Amendments
- Ensure the US constitution is reflected in the laws of the land, including the way in which the laws are made, interpreted and enforced
- Ensure that government is for the people – including an appeals process that allows citizens to challenge the laws if they consider them in conflict with the constitution – and drive changes to the constitution as well.
And if we think about the benefits of this system they are pretty self-evident in my opinion:
- It protects the core underlying principles on which the US is governed with checks and balances
- It creates a system of delegated authority to speed up decision making and avoid everything escalating back up to the senior branches of government
- It provides a feedback mechanism to ensure decisions are implemented correctly and support evolution of the constitution and laws
Imagine the US without a system of governance – the outcomes would be the inverse of the point above – chaotic, random government, lack of ability to challenge decisions, slow (even slower! J) decision making
So why do we need to govern data, it’s not the USA?
Data may not need life, liberty and the pursuit of happiness, but if you are a business where data plays a key part in your continued survival (whether as key part of the product, for business decision making, or regulatory / legal reasons) there probably are some high level things you want to care about – some fundamental values you want to live by to avoid chaos. That’s Principle #1 – Be clear on what data really matter to your business success and survival. For example:
- Data used in my customer products and solutions should meet my customer promise
- Data that matters for business decision making should be reliable
- Data needed to fulfil my legal and regulatory obligation should be managed in accordance with those laws and regulations
- Data that allows me to run my business productively and successfully should be available on a timely basis and in a meaningful way
Establishing your fundamental values for what matters in your data world is the foundation for reaping the benefits of Data Governance – better product, better business decisions, fulfilment of legal and regulatory obligations and improved productivity and business success
Good data governance makes you better, faster and cheaper. It puts you in control.
Bad data governance, like bad political governance delivers random and chaotic outcomes. You might be lucky that everything works out but that doesn’t seem a very good basis for running your business.
At this point I expect the sceptics among you are thinking this is all baloney. How do high level truisms add any value to running your business better day to day? Fair challenge – and the short answer is that it is a question of a bottoms up vs top down view of the world. There’s a famous ad for the UK Guardian newspaper which makes this point beautifully – if you just look at the things hitting you in the face today, you miss the bigger picture and live in a reactive rather than plan-ful world. And just like the cost of a newspaper is incidental to your life, the cost of thinking through the top down picture is often very low but highly impactful.
The dry dish of Data Governance is best served over several courses, so in my next post I’m going to tackle that question in more detail, as well as introducing the conceptual hierarchy for data governance that I use, together with some practical examples.
In the meantime let me know what you think. What are your favorite examples of governance catastophes? Do you believe data governance has value? What data is most fundamental to your business success and how would you rate your control over it? What are the questions you most want answers to?