The buzz around data has turned into a steady hum, as it has long been considered the lifeblood of organisations. But every day we still talk to people who feel frustrated with data management and especially data governance, and see it as a thorn in their side. 

If you’re like most people, this post will help you start and manage a data governance project that will bring lasting improvements to your organisation. 

I’ve lifted the lid on the five most effective practices that I’ve used and recommend to customers.

From the bottom to the top of any given company, every person is now required to have a certain amount of data literacy. They need it, because it’s now almost impossible to avoid working directly with data. And that means more risk associated with data manipulation as more people add data to databases and data lakes, interrogate data and run reports. 

Some privacy regulations like GDPR add more complexity to the way that sensitive data has to be processed and stored.

So – you have extra risk because more people are touching the data, and extra rules around handling it. This highlights that data governance is important not just from a company performance view but also just to ensure that you don’t end up accidentally leaking data for example, and then getting hit with a massive fine and bad publicity. Not to mention the damage this could cause to individuals which is the main reason for the rules in the first place. 

Data governance is a set of principles and practices you should be using to ensure high quality through the complete lifecycle of your data.

The better your data governance practices, the better and cleaner the overall data. 

But having clean data isn’t just less stressful, a time saver or something to brag about (although it is all of those things). Your business decisions will now be based on valid data, which massively increases the chances of good results for your business.

Here are the five practices I recommend: 

  1. Start small. Pick one small, high-impact area to start with, and then build on the quantified benefits from that project with larger projects later. This gives you the chance to shower your project with attention. A common reason for failure is starting multiple projects at the same time which all end up in an unloved state of partial completeness.

    Adopt the principle of start small, fail fast and make incremental improvements.

  2. Establish Priorities. Start off by tackling the data that will be in highest demand from your user community. This will boost your user acceptance, which is one of the most important keys to success.

    For example – let’s say you are an eCommerce company. Prioritising Customer, Sales and Product data will give you the most benefit, not data like Catalog and Supplier Information which your users are less likely to frequently need.

  3. Apply a Hybrid approach to Data Governance. There has been a trend of applying data governance as a top-down programme but many companies are now dealing with the pain this brings around buy-in and adoption. At the same time, bottom-up data governance often doesn’t work well either – you don’t have enough visibility of your high-level goals.

    If you blend these two approaches it allows you to get stakeholders and executives working closely with Data Architects and Data Stewards in a more meaningful and egalitarian way.

  4. Monitor and measure your results the right way. Having a set of objective measures agreed upfront makes it easier to focus on the right things.

    But make sure you avoid vanity metrics such as the number of policies established, or meeting participation percentages and use ones that align with strategic business goals such as time and cost savings.

    I previously worked for an eCommerce company and we noticed that it would take one of our analysts three days to generate our monthly sales report.

    This was because all the data had to be pulled in from many different sources and stitched together by hand. Once we had put better Data Governance processes in place this made it much easier for them to collate the data and generate the report. In the end, it took them just two hours and they were able to put that time saved to better use on other tasks.

  5. Identify data domains and generate consistent information. Your business is made of many different elements such as Customers, Suppliers, Products and Orders.

    It is not enough to just identify them. Make sure you ask the questions below so that high quality information is available for each element.

  • Who owns the data for each element in your organisation?
  • What policies and standards exist already?
  • Do you have business glossaries available?
  • Has a Data Dictionary been defined? 
  • What business processes are you using this data within?
  • What applications do you have that use this data?

Take the time upfront to organise and structure all of the data above for each of the domains.

This will pay dividends. It will help identify who is actually in charge of each area and ensure that clearer ownership is defined. Ownership is absolutely essential for strengthening your data governance initiatives. 

Now you can see why data governance is important and can make your life better, try implementing some of these straightforward approaches. Let us know what worked best to improve your data governance processes and controls – keep the conversation going by contacting us.

Also, if you’re a data analyst looking for ways to make your data stories come to life for your stakeholders, check out this blog from one of our analysts!