Run the Retention Math Once: What You Spend on Training vs. What You Lose to Turnover

Now layer in what the National Restaurant Association consistently reports: somewhere between 75 and 80 percent annual turnover industry-wide. If you've got 20 employees, that's 15 to 16 people walking out the door in a 12-month window. Run the number. Fifteen exits at $5,864 each is $87,960 gone. In a year. At a single location.

That's not hyperbole. That's the industry average, and most operators aren't tracking it this way because the cost never shows up as a single line item.

Here's the part that actually stings when you sit with it. If you're hitting that average turnover rate, you're already spending close to $88,000 a year on replacement costs and you almost certainly aren't thinking about it that way. You're thinking about it as a rough Tuesday, a frustrating Thursday, a text from a server at 6pm saying they're not coming in. The cost is real. You just don't see it assembled.

So let's flip the question. What would it cost to hold onto one more person per quarter?

One extra retained employee per quarter means four fewer replacements per year. Four times $5,864 is $23,456 you don't spend. That's the target. That's the math you're working against.

Now what does training actually cost? If you're doing it the old way, paper quizzes and tag-along shifts, the honest answer is that it costs you management time, inconsistency, and a guess that the information stuck. It doesn't cost much in out-of-pocket dollars. But it also doesn't work very well, which is why the turnover number stays where it is.

If you go the other direction and invest in something real, let's say you're spending a few hundred dollars a month on a training system, call it $300 as a round number. That's $3,600 for the year. To stay math-neutral, that system has to prevent you from losing one employee you would have otherwise lost. That's it. One person in twelve months stays longer because they felt prepared, felt like the job was organized, felt like someone thought about their experience in the first two weeks. One person.

Three thousand six hundred dollars spent. Five thousand eight hundred sixty-four dollars not lost. You're ahead by over two thousand dollars and you've barely started.

Now think about what actually drives early turnover. The Cornell research and a lot of what I've watched in 30 years of operating backs this up: the first 30 to 60 days are where you lose people. They show up, the training is disorganized, nobody can answer their questions consistently, they feel like dead weight on the floor, and they start scrolling job boards by week three. It's not usually about pay. It's about feeling set up to fail.

A server who walks into their first real Friday night rush knowing the menu cold, knowing which wings come with ranch and which ones are dry-rub only, knowing to flag the gluten question before it gets to the kitchen, that server doesn't feel like dead weight. They feel like they've got a shot. That's retention. Not a ping-pong table in the break room. Just basic competence built in fast.

I built the platform I built because I watched this cycle repeat at every concept I opened. Great hire, bad onboarding, frustrating first weeks, gone by week six. And we'd do the whole thing over again. After about the fifteenth time, you stop thinking it's a people problem and start realizing it's a systems problem.

The math doesn't care about the specific training system you use. It just cares about whether the training works. If what you're currently doing is retaining people, great. If you're hitting average turnover numbers, the formula above tells you that even a modest change, one that saves you one replacement per quarter, pays for itself fast. In most cases, inside 30 days.

Run it for your own headcount. Take your number of hourly employees, multiply by 0.75, that's your expected exits in the next 12 months at industry average. Multiply that by $5,864. That's the cost you're already absorbing. Then ask what 25 percent of that would buy you in actual training infrastructure. Whatever that number is, it's probably more than you're spending now and less than one quarter of what you're losing.

Do the math once. Put it somewhere you'll see it. Because the turnover problem doesn't feel like $87,960. It feels like a bad Tuesday. And that framing is exactly why so many of us keep solving it the wrong way.

Have a great day! — Terry Psaltakis
Your AI Restaurant Guy

For more on putting this into practice, see how ShiftTrained approaches the best menu training programs.