I'll Sort It Out When Things Calm Down

You know what your operation looks like. You're not in denial about it. The food comes out different depending on who's cooking. The close is a coin toss - some nights it's tight, some nights you walk in the next morning and the place feels like nobody finished. The new person behind the bar is making drinks from memory, and the memory is about two shifts old.

You can see all of this. You've probably got a mental list of everything that needs fixing, and it's long enough that the idea of writing it down feels like a waste of time you don't have. So when someone mentions a training tool, or a system, or anything that involves documenting how your business runs, the response is immediate and honest: there's nothing worth documenting yet. You'd be capturing a mess. Putting a frame on a bad painting.

I hear this from owners all the time, and the thing is, they're partially right. Training - the way most people picture it - genuinely would be useless here. If your prep is inconsistent and your close is sloppy, sitting someone down with a manual that describes the inconsistent prep and the sloppy close achieves nothing. You'd be formalising the problem. The instinct to reject that is correct.

Where it goes wrong is what happens next, which is usually nothing. The problems stay. The plan to fix them stays in your head. And the operation keeps running on whatever each person remembers from the last time someone showed them something.

The loop

There's a version of the future where things calm down and you finally have the time to step back, look at the whole operation, fix the things that need fixing, and then build systems around the improved version. Every owner I've spoken to has this version somewhere in the back of their mind. It's the reason they haven't done anything yet - they're waiting for the right moment.

The right moment doesn't come. There is always a no-show to cover for, a supplier who shorted you, a piece of equipment that broke at the worst possible time, a Friday night that went sideways enough to eat your entire Saturday fixing the fallout. You are permanently in the middle of something, and the idea of stepping back to work on the business instead of in it is a luxury that belongs to someone with more staff, more money, or fewer fires.

So you keep doing what you've been doing. You show up early. You stay late. You fix things yourself because it's faster than explaining it again. When someone does something wrong, you correct it in the moment and hope it sticks, knowing it probably won't stick with the next person, or the person after that.

It's the blunt knife problem. Every chef knows it. You're too busy chopping to stop and sharpen the blade, so every cut takes longer, so you fall further behind, so you definitely can't stop now. The knife gets duller. The chopping gets slower. You work harder and harder and the output gets worse and worse, and at no point does a gap open up where you can pause, because the blunt knife is the thing creating the pressure that stops you from sharpening it.

I spoke to a builder who was getting called back to the same kinds of jobs every few weeks - not because the work was bad, but because his crew on different sites kept making the same mistakes with waterproofing membranes. The steps weren't complicated, but they were specific, and every time a new labourer came on he'd explain it once on the first morning and then hope for the best. By the third site visit to fix the same problem he could have done the whole job himself in less time than he'd spent driving back and forth to patch other people's guesses.

A kitchen I talked to had the same prep failure every Monday. Whoever closed on Sunday night wasn't setting up the mise en place for Monday's lunch service - not because they were lazy, but because nobody had ever shown them what Monday's opener actually needed. The closer did their close. The opener came in and started from scratch. Every single week.

You can't work harder at the same things you're already doing and expect the pattern to break. Something structural has to change, and waiting for a quiet week to make that change is the thing that keeps the quiet week from ever arriving.

Forget the word "training"

Here's what happens when an owner hears the word "training." They picture onboarding day. A binder. Maybe a screen with modules and a quiz at the end. Someone sitting in a back office watching a video before their first shift, learning a version of reality that may or may not match what actually happens on the floor. They tried something like that once, or they looked into it, and it felt wrong because the operation wasn't ready for it. The product had to get better first, and then they'd worry about documenting it.

That instinct is sound, as far as it goes. Formal training programs - onboarding modules, pre-shift briefings, printed checklists laminated and stuck to the wall - assume that there's a stable, correct version of each task to teach. If you don't have that, if the way things get done changes depending on who's on shift and what kind of day it's been, then yes, building a training program on top of that is a waste of everyone's time.

But there's a different thing entirely, and it doesn't look anything like training. It looks like this: someone is standing in your kitchen at 10pm, halfway through the close, and they can't remember whether the fryer gets drained tonight or just filtered. Or someone's on a construction site looking at a junction detail and they're fairly sure they know which sealant goes where, but not sure enough to commit. Or the new person on the bar has a customer asking for a drink that was on last week's specials and they have no idea how it was made.

In each of those moments, what that person needs is something on their phone that shows them how to do the thing they're standing in front of. Step by step. Filmed in their workplace, with their equipment, showing the version their boss actually wants. The answer to one question: how do I do this?

That's execution support. It happens during the work, in the moment someone needs it, at the point where the alternative is guessing or skipping the step entirely. Something that puts the correct version of a task in front of someone while they're doing it changes which version of the task actually gets done on each shift. The operation gets better because fewer things go wrong each day, and fewer things go wrong because people stop guessing.

One person already does it right

Even in the roughest operation, there is usually someone who does each thing well. You've got one closer who actually does every step, every night, without being reminded. One tradesperson whose work holds up on the first inspection. One prep cook who labels everything the right way and sets up the station so the next person can walk in and start immediately.

The knowledge of how to do each task properly already exists in your business. It lives in that one person's hands. The problem is that when they're not on shift, or when they leave, that knowledge goes with them and everyone else is back to guessing. You've watched it happen - your best person takes a week off and the standard drops measurably, visually, within days. They come back and it recovers. They leave for good and it drops permanently.

The fix here has nothing to do with getting the whole operation perfect before you start capturing anything. You don't need every task sorted. You need one good version of each task, and you almost certainly already have it - it's just locked inside one person's head. The moment you get that version out of their head and into something the rest of the team can access, the floor rises. Everyone starts working from the same reference point instead of from whatever they remember from a busy Tuesday three weeks ago.

You film the closer who does it right. Now every closer after them has those steps on their phone. You film the tradesperson whose waterproofing holds. Now every labourer on every site has that process in front of them while they're doing the work. You don't need to have fixed everything first. You just need to have identified the person who already does it well and captured what they do.

What this looks like in practice

That's what I built Lattify to do. Your best person films the task once, on their phone, in your venue or on your site with your actual equipment. Lattify turns that recording into a step-by-step guide that lives on your team's phones. When the new closer gets to the fryer and can't remember the schedule, they open the guide and see exactly what your best closer does, in order, filmed in your kitchen. When the apprentice is looking at a junction detail, they pull up the guide and see your best tradesperson doing it correctly, on a site that looks like theirs, with the same materials.

The thing your team opens mid-shift when they're standing in front of a task and need to know what to do. The answer to the question they currently either guess at, skip, or interrupt someone else to ask.

Your operation doesn't need to be perfect before you start. You capture the best version you have right now - one task, one person, five minutes - and the floor gets better from the next shift onward. Then you do the next task. And the next one. Each one is a small piece of the mess that stops being a mess, because the person doing it has something better than a guess to work from.

Each one is a small piece of the loop that stops looping, because the person doing it has something better than a guess to work from.