• X

    Latest News

    How to succeed in business driven Data Governance — start right!

    In so many of the Data Governance programs I’ve rescued, I’m always frightened and shocked to see how many are down the road with big technology purchases, assigning names to roles, “standing up” capabilities, and {fill in your buzz word of choice here}, before they have completed one of the most fundamental and critical aspects of any enterprise data initiative. All of the things just mentioned will absolutely and in an emphatic way meet disaster if this one thing isn’t done before all of those are even considered. What is that one thing, it’s the business data inventory.

    Contrary to most of the technology integration expert opinions, it is highly probable and logical to gain an understanding of the unique business data elements of an organization prior to pursuing any data initiative, which in a lot of industries that means before any significant technology project. It is also important to note, if you are in financial services, 90% of all projects are “data initiatives”, and over the next 10 years every industry will be heavily contingent on data as a key organizational ecosystem variable.

    Regardless of your preferred approach towards establishing your business data inventory, getting to unique business data elements that are used as the variables of your business metrics or KPIs, and used to slice and distribute those metrics/KPIs across the various aspects of your business is the key to having one version of truth — surprise; surprise one version of truth doesn’t involved an $8MM investment with {fill in top four system integrator} and another $4MM {fill in technology pitch of choice}, which will actually cost $35MM all in because neither of those two accurately accounted for data issues.

    A business data inventory starts with business folks providing insight into the data that is critical to their everyday jobs, and then someone actually has to facilitate and capture that information to share across the organization — it’s not rocket science but it is painful and time consuming. We have created a guide that will help those of you that are struggling with engaging your business colleagues or partners, and I highly recommend downloading that here.

    [space height=”20″]

    [maxbutton id=”1″]

    [space height=”20″]

    Treating data like an asset is actually like playing with legos if you do things in the right order. It becomes a heck of a lot easier to assign someone’s name to a finite list of possible business data elements when you can then go and talk with that person about whether or not they are the right person to steward, as opposed to abstractly assigning names to large buckets of data (i.e. subject area level assignments).

    Having “the business inventory list” leads to a real and concrete conversation, as opposed to corporate speak about shared responsibilities, joint accountability, and all of the other stuff that gets discussed because the leadership team hasn’t transitioned out of abstract land. Establishing real accountability is only the first, small step for stewardship and a giant leap for data governance, because the next legos we can begin stacking up are in the variety of the physical instances of data. If you establish ‘customer name’ as a key business data element (which every company on the planet should have that in their glossary by now, that’s 90s), you can now link the physical version (databases, columns, tables, ETLs, etc.) to the customer name business data element. If you are like us, you take it one step further and create an algorithm that aggregates the quality scores and outcomes for all of those physical versions, grab the people that are accountable for those databases, ETLs, repositories, etc. and store it right next to their scores, and then rank everybody in a visual analysis to isolate poor performance. Anyone of those people that fall in the bottom 20%, highlight and send a notice out to the whole data community creating paranoia and shock. All of these super cool things are only possible because we have started with one view of business data, and related all of the supporting technological and architectural aspects to the business data element.

    While many of the tactics I’ve illustrated are aggressive and should only be deployed from a position of strong support by the business, they absolutely work. Regardless of how aggressive you decide to attack data, you must (I repeat MUST) start with an inventory of what you have first, before you even mention the acquisition of tools to fix data. While I’m a firm believer of DIY to a lot of organizational improvement, there are ways to expedite this process that can be explored in a free coaching session with our Data Strategy team  (click here for more information).