Archive for data protection

Why It’s So Hard to Quantify Losses from Data Storage Problems

By Tom Fedro

Thomas Fedro quantifies loss from disaster recoverySome time ago, I read Out of the Crisis, by Dr. W. Edwards Deming, the man generally credited as the Father of the Total Quality Management movement.  This book (which came out in 1984) is just chock full of brilliant business insight, but I’ve carried one phrase with me throughout my career.   He wrote that the effects of some decisions were “unknown and unknowable.”

So much of his book was an exhortation to run companies using statistical process control that the line seemed out of place.  It wasn’t though.  He was illustrating that effects that may not have a visible dollar sign attached to them are still worthy of management consideration.  We tend to push aside concerns that we can’t label easily, and this is especially true in data protection and restoration.

Anyone in the tech sector hears about the nightmare situations where companies lose or compromise data that leads to gigantic expenses by way of lost contracts, fines, or settlements.  For most companies, though, data loss doesn’t represent something that can be negotiated and pinned down.  The truth is, several consequences of data loss have Dr. Deming’s unknown impacts.  Today, I want to focus on two.  There are a whole lot more, but those will come in later posts.

  • Rework.  If you bought motherboards from an overseas provider and they were delivered with flaws, you’d send them back.  The manufacturer would take the defects and put them into “rework.”  This is pretty straightforward.  “Re” is the Latin root that means “again.”  The work is done again.  Everybody gets that this is a bad thing in manufacturing.Somehow, when it’s done with soft work like typing, accounting, and marketing collateral; management tends to forget that it’s still doing work again.  Let me rephrase that—it’s doing work twice and getting the benefit once.  The problem is that a manufacturer will show the rework in the accounting reports.  There’s simply no way to do that effectively in other industries, it’s unknown.
  • Inefficiency.  This is one of my pet peeves.  I like to compare it to writing with a pencil and a piece of paper.  Imagine you had to get a five page document handwritten and you could get about two pages done per hour.  No problem, two and a half hours, right.  Okay, let’s add a curveball.  The pencil lead is going to break.  You’re going to need to wait a few minutes because only one or two people know how to operate the sharpener, and they’re sharpening other pencils right now.Let that happen a time or two and your two and a half hour job has taken an extra hour or two.  Essentially, it’s a hidden and unknown expense.  You can measure how many times the pencil breaks.  You can even measure how long it takes to be sharpened, but you can’t really measure how it impacts an employee’s ability to write.  It would be naïve to believe that the stopping and restarting doesn’t cause delay above the actual downtime.