Bad Data Quality is Like Raw Chicken

Recently, I was traveling internationally and I was served chicken nuggets at a dinner. I was pleased by this as the fried chicken is usually relatively safe to consume. Imagine my surprise when I took a bite and the chicken was raw.

Yesterday, I sat on a panel at FIMA US 2016 in Boston where the topic was "Proactive Approaches To Working Toward Better Data Quality for Better Decision Making". The chicken example was poignant to me as it is a perfect example of being reactive. I had to consume bad chicken (data) and then complain about it. How often do we do this in our organizations?I believe that we can do better. A long time ago restaurants developed food quality tests using such basic instruments as a food thermometer to ensure "doneness" of meat. We can use this same approach by instrumenting the data each step of the way.

The first step is to understand your data flows to the extent that you can begin to inject instrumentation. At first, all of your data may be “raw” but you can begin to create consistency. Your downstream consumers will begin to have trust in the data, being able to make decisions whether it is fit for consumption.

There is much to be said about this concept. We recommend using a data diagnostic approach, as the diagnostics will provide a consistent and clear temperature of your data. By doing this well we will always serve our data consumers “fully baked” data.