Doctor, my data needs first aid!
By: Esther Labrie
We’ve all been there: a sore throat has been bothering you for days, you feel a fever coming up, you keep saying “it will blow over, it’s just a cold” and yet you “swart to soumd like bis” because by now your nose is running faster than your dog when it chases the neighbour’s cat … At some point you just have to admit it: you are ill and you need to surrender yourself to bedrest and painkillers. And, because you pretended to be immune to a trifling cold, you now have a serious pneumonia that will keep you out of business for a good month at least…
The 1-10-100 rule
So you nursed your flu into a pneumonia, but what does that have to do with data? Well, it may sound strange, but data are not that different from people. Are you familiar with the 1-10-100 rule for the cost of bad data? The 1-10-100 rule states that it takes $1,- to make sure you get the quality of a new record right before it enters your system(s), $10,- when you have to correct it later, and $100,- or more as you have to deal with the ramifications later on when the data has had the chance to spread its wings and fly to every corner of your system. Or, as your doctor would put it: “An ounce of prevention is worth a pound of cure.”
Diagnose your data
So we’ve established that any data quality issue has the potency to go from bad to worse fast. It goes without saying that, like your doctor, Human Inference is in favor of prevention rather than cure. But that’s easier said than done. Because we also know how hard it is to admit you have a problem. Especially if you don’t know exactly what that problem is. When you are working with legacy systems and disparate sources that contain outdated or incomplete data, there is no way your business is going to reach its full potential. That’s why we encourage organisations to diagnose their data so they can take measures to improve the quality of their data and prevent data pollution at the source.
Do you suspect your data are still feeling a bit under the weather? Human Inference is here to help.
More blog items