Back to blog
Guide

The Rule You Broke An Hour After Making It

The hour after you write a rule is when you are most likely to break it. Not because you forgot — because you were looking at the thing you just fixed, not the next one. Here is why, and what actually holds.

You work out what went wrong, write the rule, tell everyone.

Within the hour you break it yourself.

⚠️ Not by forgetting. You would have recited it correctly if asked. You broke it because you were looking at the problem you had just solved, and the rule arrived in a situation that did not look like that problem.

The short version

  • The period right after making a rule is when it is most likely to be broken
  • Writing a rule feels like solving it, and that feeling replaces vigilance
  • Rules that live in memory fail; rules built into the step hold
  • The same rule written for others applies to you last
  • Count how often it is broken, not whether people know it

Why the hour after is the dangerous one

What you thinkWhat is happening
"Now we have a rule for this"The rule exists; the habit does not
"I've just been burned, I'll be careful"You are alert to that case, not its next shape
"Everyone's been told"Telling is the cheapest part

⚠️ Writing the rule produces the satisfaction of having fixed it. That satisfaction is the problem — it closes the file.

One: it arrives wearing different clothes

The reason you do not recognise it is that the next instance looks nothing like the first.

The rule    "read the address back to the guest"
Where you broke it   a booking taken by message, where nothing was read aloud
⇒ same rule, different shape, not recognised

⚠️ Your alertness attached to the surface, not the principle.

We did this three times in one day

Writing honestly about our own week: we made a rule about checking something before stating it, and then broke the same rule three times in the following hours — each time in a different form. Once in a written note, once in a process we had built ourselves, once in a file where we had literally typed the warning at the top.

Made the rule
Broke it in a note
Broke it in a tool we had built for that purpose
Broke it in a file where the warning was written above the line

⇒ The third is the instructive one. The warning was visible on screen and it did not stop the hand.

Two: a rule in memory is not a control

This is the practical heart of it.

Rule in your head   works when you remember and are not busy
Rule in the step    works always

⚠️ On a busy night, the first one is not there at all.

What "in the step" means

Memory   "always take a deposit for groups over six"
In step  the booking form will not save a group over six without a deposit field

⇒ In a venue this is usually physical or procedural, not software:

The float is counted into a bag that cannot close until the slip is signed
The key does not come off the hook without the sheet being filled
The till prompts for a reason before any discount

⇒ Related: who can say stop.

Three: a warning that reports is not a warning that stops

Worth separating, because they feel the same.

Reports   tells you afterwards that it happened
Stops     prevents it happening

⚠️ A report catches the problem after the cost. We had a checker that flagged the exact mistake we then made — three times — because flagging does not hold a hand.

Where each belongs

Stops  for anything with a cost you cannot undo (money out, deletions, sending)
Reports for things you genuinely want visibility of, not prevention

⇒ If you find yourself reading the same report entry repeatedly, that item needs moving from reports to stops.

Four: the rule you hand out applies to you last

Uncomfortable and consistent.

You write the rule because you saw others get it wrong
You are therefore the person least watching yourself for it

⚠️ Being the author feels like being exempt, not because you think you are above it but because your attention is on enforcement rather than compliance.

The check

Next time you write a rule, ask: what would breaking this look like for me?

⇒ Answer it in one sentence, in writing, next to the rule. That sentence is what you will recognise later.

Five: what to do in the first hour

Since that hour is the risk, use it deliberately.

One   Do the thing the rule requires, once, immediately
Two   Find where it lives in the actual process, not in the notice
Three Write down what breaking it looks like

⚠️ One matters most. A rule you have never once followed is a sentence, not a practice.

⇒ We announced a habit in a message and the very message announcing it did not follow the habit. Doing it once before announcing it would have caught that.

Six: counting instead of reminding

The instinct when a rule is broken is to restate it. Restating does very little.

Weak    say it again at the next meeting
Better  count how many times it was broken this month

⚠️ A count changes the conversation from character to design. Five breaches in a month is not five careless people; it is a step that does not hold.

What a count tells you

Broken once       - an individual moment
Broken repeatedly  - the rule does not fit the work
Broken by everyone - the rule is wrong, or the step is missing

⇒ The third is common and rarely admitted. Some rules are broken because following them makes the job impossible, and the honest answer is to change the rule.

Seven: rules that cannot be broken quietly

The best rules make breaking them visible without anyone policing it.

Two signatures on the cash-up sheet
The safe log where the last entry is always visible
A rota where changes are written in a different colour

⚠️ None of these prevent anything. They make the break obvious, which in practice is nearly as good and far cheaper.

⇒ On records that make changes traceable: who did it is not what they did.

Numbers worth keeping

One   Rules you have written this year
Two   How many are built into a step rather than a notice
Three How many times each was broken this month
Four  How many you have personally broken

⚠️ Four is the honest one and the one that tells you whether the rule was workable.

The card for the office

────────────────────────
  Just made a rule?

  1. Do it once yourself, now.
  2. Where does it live in the actual step?
  3. What does breaking it look like for me?
  4. Count breaches. Don't restate.

  The next hour is when it breaks.
────────────────────────

Common objections

"Everyone knows the rule"

⚠️ Knowing is not the failure point. The failure point is recognising the situation as one the rule covers.

"We just need to be more careful"

⇒ Care is a finite resource that runs out on the busy nights, which are exactly the nights it is needed.

"Putting it in the system takes too long"

⚠️ Sometimes true. Then make breaking it visible rather than impossible — a second signature, a different colour, a log.

"It only happened once"

⇒ Count for a month before concluding that. Most "once" turns out to be "once that was noticed".

"I wrote the rule, I'm not going to break it"

⇒ That belief is the specific thing that makes it likely. Authorship is not immunity.

What to do this week

One   List the rules you introduced this year.
Two   For each, say where it lives - notice or step.
Three Pick the most-broken one and move it into the step.
Four  Next rule you write: do it once yourself first.
Five  Write down what breaking it would look like for you.

Summary

  • The hour after making a rule is when it is most likely to be broken
  • Writing it produces a feeling of resolution that replaces attention
  • It reappears in a different shape and is not recognised
  • Rules in memory fail; rules in the step hold
  • Warnings that report do not stop hands — move the costly ones to stops
  • Count breaches rather than restating; a high count means the rule does not fit

The rule you just wrote is the one you are about to break. Not from carelessness — from having already filed it as solved.

Free, no signup, ~5 minutes

Map out your operations in 5 minutes

Eight questions cover reservations, customer management, shifts, and settlement. Results shown instantly with industry benchmark. Sales emails only if you request them.

Your answers are not stored. The assessment runs entirely in your browser.

Try tasteck free for 30 days

No credit card required. Full access to reservations, cast shifts, dispatch, and analytics.

  • No card required
  • Free data migration support
  • All features unlocked for 30 days