About a year ago I had a little tiny light bulb pop off in my head. I don’t remember exactly what set it off, but it was likely while I was on one of my many sailing trips on the sometimes rocky seas of metrics and numbers. I knew it was counter-intuitive, and I knew it’d be tough to quantify. You know that feeling when you’re trying to think of a word, or the name of an actor, or the answer to some useless trivia question – and you have it on the very tip of your tongue, but you can’t quite find it? It was like that. I spent many idle hours on planes, walking my dog, and at the gym trying to piece it together. But I couldn’t.
Ultimately I just kind of forgot about it – for a while. Then, as I gave up and decided to take a nap under a tree, the metaphorical apple fell and bonked me right on the head. The “apple” being a conversation with a client in health care about the cost of turnover.
This particular client had quantified the cost of losing a nurse at roughly $35,000. Some of that, about $10k, was general administrative costs to exit the employee, plus the cost of recruiting a replacement. But the majority of it, $25k, was overtime cost. Their average time-to-fill for a nurse was 60 days. Their in-house training program was also 60 days. Filling this 120 day vacancy cost them a total of $25k in overtime costs above and beyond what they would pay a normal nurse. Overtime, PRN (essentially temp labor) nurses, and other costly staffing options had to be used to fill in the gap.
One-Up Theory
(Who doesn’t have time for a gratuitous 80′s pop-culture reference?)
This is where the theory comes into play. Let’s run a hypothetical. Let’s say that as of today, the hospital is fully staffed. But instead of waiting until a nurse exits to go hire one, they hire one right now. Today. Making them “overstaffed” by 1. Now, I’m sure they could make use of this nurse, but just to play devil’s advocate to my own theory, let’s assume this nurse doesn’t work at all. Our “plus-one” nurse just sits at home, checks Facebook and plays Angry Birds, being paid $62,100 salary (averagefor this hospital’s city) for being “on call”.
The next piece to this theory is to find our break-even point. If someone exits, our Facebooking, Angry Birds-playing nurse is ready to go. He or she steps in and fills the void left by the FTE that exited. And there is no overtime. $25k saved. How long can our Facebook nurse collect paychecks from the couch before we lose money (where the “wasted salary” overcomes the saved OT costs)? It’s 4 months, 6 days.
Finally, we just have to tie a ribbon around this package and call it a day. That 4 month break-even point means that if the org is constantly “overstaffed” by one, any time two terminations are less than 4 months apart, the one-up theory saves us money – because we saved more money avoiding overtime costs than we spent on the idle nurse’s salary!
This particular organization turned over 20+ nurses per year –more than one per month. Some basic math shows that they would save hundreds of thousands of dollars in overtime cost by applying the theory. In fact, this particular organization would have benefitted from a “two-up” theory, but that’s a rabbit hole I won’t go down today (though if you want to hear more about it, I’ll gladly talk your head off with theory).
Obviously this is not the case in every job, department or organization. In many cases there may be logistical issues with running “one-up”. This doesn’t apply in exempt workforces. It doesn’t apply if your turnover rates don’t catch up to your overtime cost like the above example does. But there’s merit to the theory – it does offer some serious food for thought. This practice, or pieces of it applied in organizational context, can be beneficial. Would a manager turn down extra staff? Would a finance person turn down quantifiable ROI? If the answer to either of those is “yes”, well… I have many friends that work for the government too 
.
In part 2, we’ll take a look at potentially redefining what “optimally staffed” truly means. We’ll also look at some real world examples that work, and others that don’t. Further, I’ll provide an Excel template where you can test this theory with your own numbers. Stay tuned.