Improvement Insights Blog

Measure failures not success. Here’s why:

“Hi, I’m Jay Arthur, author of “Lean Six Sigma Demystified” and QI Macros [software].

“Recently one of our users called up and asked me about showing percent compliance. You know, present success (in this case it was neonatal intubations – this is for healthcare). But the goal obviously is to get to 100% first-time success at intubating a preemie, right? So that would be a good thing.

“Now what most people do is they try and focus on the success side of this, so if you’re down here, you want to get up that-a-way, right? You want your target lines up here at 100%, and as you go up, you’re going to get to a point where, “Gee, it’s starting to look pretty good: 95% or 96%.” Well, that still means you have 5% failure, all right?

“Now, one of my senseis when i was first learning Quality said, “Once you get above 80 percent success, you may want to start focusing on failure.” You want to know what your failure rate is. So right here our failure rate might have been… if this was 45 it might have been up here, and it’s going down like that. If we’re going to track things, what we want to do is start tracking our failure rate, you see, because Lean Six Sigma is a problem solving methodology; it’s not a goal achievement thing, right? By reducing delays, defects and deviation we improve our outcomes. Patient outcomes, quality of our products and services, delivery schedules… all of that stuff gets better, all right? So this is a byproduct of reducing that. If we focus on this, guess what? We’re going to get in a much better position to actually solve the problems that we see.

“So if you’re trying to focus on success, flip the metric over. Figure out what the percent failure rate is and then we can track that as a control chart. This is not hard. With QI Macros we can tell how it’s doing and we can tell if we actually moved the needle.

“When you’re doing problem solving most people fail. They forget the one thing they have to do is put in place a control system to monitor performance so if it starts getting bad you can go fix it. I cannot tell you how many people call me up and say, “Oh yeah, we tried process improvement. We made an improvement but six months later it was gone.” That’s because you failed to put in a control system. If you’re not going to be willing to do that, stop doing improvements. Just stop it. You’re wasting your time and everybody else’s, right? So you’re going to need some tools at the end to be able to do this.

“So that’s my Improvement Insight for this week: Focus on the failure rate, not the success rate. The success rate will follow reductions in failure rates. That’s pretty simple. Now I believe that anyone can learn how to do this, and they can learn how to do it quickly.

“You don’t have to know everything to improve anything.

“Let’s create a hassle-free America. Let’s go ahead and improve something this week.”