Data Governance Interview with Ed Mathia

Data Governance Interview with Ed Mathia

In this Data Governance Interview, I spoke to Ed Mathia. Ed is a Data Scientist who learned very quickly that data quality is paramount, and that the processes to make data right the first time increases its value.  When he finds incorrect data he works to change the data source to ensure the company has good information and makes good decisions. I’ve always enjoyed my conversations with Ed and in particular love his analogies when explaining Data Governance.

How long have you been working in Data Governance?

About 15 years ago, I became the Specification Manager for a semiconductor materials manufacturer.  My team was responsible for keeping the product specifications for the company.  The old-timers used to say that a specification mistake released to production was equivalent to buying a house.  We recalculated when I was there, and a mistake could easily be 2 million dollars – that is a nice house.

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

I have heard people talk about Data Governance as an unusual field, but I think that we are just early in the curve.  Data Governance reminds me of the Quality profession twenty years ago.  Companies understand Quality now, but it took Toyota and Motorola to show the benefits of great quality and the ISO-9000 standard to show the right processes.  I think twenty years from now that viewpoint will be strange.

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

I think it is critical to have a good understanding of data science and machine learning.  Companies have so much data, stored in a variety of systems, that it becomes hard to find the fixable issues and the impact of the issues on the business.  It is a lot like a Magic Eye poster.  If you don’t know how to look at the 3D image, then all you see is the repeating horizontal pattern that looks like nothing.  With the right pattern matching techniques, you can resolve the special 3D picture.  Finding those patterns in the business data means you can fix the right problems based on the impact.

Are there any particular books or resources that you would recommend as useful support for those starting out in Data Governance?

Nicola, I always found that your coaching calls are the most useful support.  Books like the DAMA manual are great, but they are generic and don’t help with the specifics of communicating with your company.  Being able to ask specific questions, to draw on your vast experience and get options very quickly is extremely helpful.  I always found the coaching sessions to be like Christmas – I look forward to them for a long time but they are over too soon.

What is the biggest challenge you have ever faced in a Data Governance implementation?

As in most areas of life, communicating the need for change is one of the biggest challenges.  Everyone knows that bad data causes pain and has to be fixed before getting the right decision.  Companies accept the pain when they think it is a small, easily-fixed issue – like a paper cut.  But if everyone accepts little inefficiencies in the data then you have a big problem.  A piranha only takes a small bite, but a lot of small bites can do a lot of damage.  That is why I think it helps to have a good understanding of data science.  It is hard to find the inefficiencies spread through the company but it is possible –  using data science I found 6 million dollars of expedited shipment fees and one hundred thousand hours of productivity loss due to poor master data settings.

Is there a company or industry you would particularly like to help implement Data Governance for and why?

I think one of the most beneficial areas are company supply chains.  In the US, financial services are awaking to the understanding of the need for data governance, but supply chains aren’t seeing the need yet.  However, every dollar the supply chain saves impacts profit directly, while financial services are predicting which customers and products might be successful.  Manufacturing is a field ripe for harvest.

What single piece of advice would you give someone just starting out in Data Governance?

Hang in there even when things seem tough.  Everybody is hoping to hire a pharmacist who will give them a pill to make them skinny.  Data Governance folks have to be personal trainers telling clients they need to eat right and exercise.  Even though it is the right way to lose weight, they won’t want to hear it.  Hang in there and be consistent.

Finally, I wondered if you could share a memorable data governance experience (either humorous or challenging)?

I once saw a data field that took on opposite to its original intent – sort of like the word “dust”.  Dust can mean either add or remove fine particles depending on whether you are talking about cleaning the house or making powdered doughnuts.  When I was managing the specifications at a silicon wafer manufacturer, one specification was how close to the edge the backside seal had to extend.  Several application engineers chose to check the “Edge-to-Edge” process specification instead of putting in the number of millimetres from the edge the seal could extend.  They were thinking that the “Edge-to-Edge” process sealed all the way to the edge, but it was actually a 15-year-old process that had the worst sealing coverage.  It really shows how important data governance is.  It would have been much clearer to focus on the specifications on the customer needs rather than which process to use. Then the process could change as long as it met the customer’s needs.  Specifying the process meant that we couldn’t give the customer a better product when new processes came along.

You can find out more about Ed and connect with him on LinkedIn by clicking here.

Have Your Say: