The Thing You Stopped Doing

There was a Friday afternoon call to confirm the weekend's large bookings. It happened every week for two years.

Nobody can remember when it stopped, and nobody decided to.

How practices die

Not by decision. By three specific mechanisms:

The person who did it left. It was theirs, informally, and it was not in any handover because it was not in any job description.

One busy week. It got skipped, nothing terrible happened, and the skip became the new default. ⚠️ Nothing terrible happens the first time, which is exactly why it stops.

The reason became invisible. The practice was preventing something. Prevention has no visible output, so after a year it looks like an unnecessary step.

Why lapsed practices are expensive

The problem returns slowly. Not the week after — three months after, by which point the connection to the lapsed practice is unrecoverable.

Nobody investigates. The rising no-show rate gets attributed to the market, the crowd, the season. The actual cause is a phone call that stopped.

⚠️ This is the single hardest failure to diagnose after the fact, because the evidence is an absence.

Finding them

Ask three long-serving people one question:

"What did we used to do that we do not do any more?"

They will produce a list in about four minutes. ⚠️ And they will not have mentioned it before, because everyone assumes the change was deliberate.

That assumption is the whole problem. Nobody questions a change they believe somebody decided.

Sorting the list

Not everything should come back. Three categories:

Correctly dropped. It was solving a problem you no longer have, or it was never useful. Say so explicitly — that closes it.

Replaced by something better. Fine, and worth confirming that the replacement actually covers what the original did rather than the part everyone remembers.

Lapsed and still needed. ⚠️ These are the valuable finds, and there are usually two or three. They are frequently the cheap preventive things — a check, a call, a briefing.

Restarting one

Give it an owner by role, not by name. That is what killed it the first time.

Give it a trigger. "Friday afternoon" is a trigger. "Regularly" is not, and will lapse again within two months.

Write down what it prevents. ⚠️ This is the part that was missing originally. A practice with a stated purpose survives a busy week; a practice that is just a step does not.

The check that stops the next one

Once a quarter, the same question.

Four minutes, three people. It catches practices in the months after they lapse rather than the years, which is the difference between a small correction and an unexplained decline.

What this says about new processes

If you are adding something now, add the three things that would have saved the old one:

Whose role owns it. What triggers it. What it prevents.

⚠️ Without the third, it will lapse the first busy week — and in two years somebody will find it on a list and wonder why it stopped.

A lapsed practice has no date of death

Decided to stop  reversible, because there is a decision to revisit
Simply stopped  nobody remembers it existed

⚠️ Skipped once in a busy week, skipped again the next, and by the third week it has left no trace — including in the memory of the people who used to do it.

---

Why this is expensive rather than merely untidy

The problem it solved  comes back quietly
It is treated as new  and solved again

⚠️ You pay twice for the same thing, and the second payment is larger because nobody remembers the first solution.

---

Finding them without relying on memory

Do not try to remember
Do read six-month-old procedures, notices, and handover records

⚠️ The signal is the reaction: "we do not do that any more." 　 ⇒ That sentence, said out loud while reading an old document, is the 　 　 entire discovery method, and it takes an afternoon.

---

Sort what you find into three

Right to have stopped
Should be restarted
Should come back in a different form

⚠️ The first group is the largest. Do not restart everything — a practice that lapsed with no consequence was not doing anything.

---

Do not restart it in its original form

The original form  is the form that disappeared

⚠️ Bring it back with one change: a named owner. 　 ⇒ What lapses is almost always what belonged to everybody, and 　 　 restarting it unchanged schedules its second disappearance.

---

The check that prevents the next one

The week after a busy week  "what did we skip last week?"

⚠️ Written down at that moment, the third-week disappearance does not happen. 　 ⇒ The question costs nothing and it is the only point at which a lapse 　 　 is still visible to the people who caused it.

And it applies to anything new

⇒ ⚠️ Before starting a new practice, decide two things: whose it is, and where a skip gets recorded. A process with both survives busy weeks; a process with neither is already scheduled to vanish.

---

---

Three to hold

Practices identified as lapsed, quarterly. The list.

Of those, how many were deliberate. ⚠️ Usually very few, which is the finding.

Restarted practices still running after three months. Whether the owner-trigger-purpose fix actually worked.

Where the record has to sit

A practice with no record leaves no trace when it stops. That is why the question has to be asked of people rather than of a system — but the answer, once found, belongs somewhere it will outlive whoever is currently doing it.

tasteck keeps recurring tasks and notes on the staff-facing screens, so a weekly check has a place that survives the person who introduced it.

Nobody decided to stop. That is precisely why it has never been reviewed.

---

Read next

• The Report You Should Read Every Monday
• The Two Weeks After You Change Something
• What Two Managers Do Differently
• The Second Time Something Breaks
• The Thing Somebody Said Once
• The Report Nobody Reads