A paused Japan programme needs a review date and a restart condition

A regional review parks the Japan investment. Not cancelled. Held, until the numbers are good enough to justify it.

Nobody objects, because holding sounds more responsible than cutting. In the paused programmes we have seen, the position two quarters later is often unchanged, and the review has spent a slice of every meeting on it.

The pauses that did resume had one line in the plan: the date on which somebody checks a specific number. What follows is a working rule drawn from that experience, not a survey finding.

The waiting time you are absorbing is mostly not research time

Look back at a programme held “until the data improves” and you often find that little data arrived while it waited. Nobody works a paused programme, so no new numbers come in, and the old ones go stale before they grow.

What was being waited for, in most cases, was someone’s judgement.

When a Japan programme is held for more numbers, the numbers are rarely the binding constraint. The unresolved question is who explains the call if it goes wrong.

A similar pattern shows up on the buying side, though it is not about internal programmes.

In IDEATECH and Demagen Research’s second-wave study of large B2B purchases in Japan (published 9 June 2026, 330 valid responses; an online panel run by a survey-PR company), the leading reason purchases ran long (n=239) was waiting for executive judgement at 63.6%, ahead of information gathering at 45.2%.

That is a purchasing survey and does not prove anything about paused programmes.

It does match what we see: judgement, not information, is usually the longer queue.

Japan’s CRM does not fill itself in while the programme is parked

The other assumption inside a pause is that waiting produces data. Where records are already entered with gaps, it does not.

Incomplete data in that kind of organisation is not a temporary condition that a pause outlasts. It is the operating baseline. Waiting for a baseline to end sets no end date at all.

Three questions that show whether data is the actual constraint

Before accepting a hold, ask the person proposing it three things.

Which number will we look at. What does it have to reach for this to restart. Who does what on the day it restarts.

If all three come back, this is a decision. Set the date and wait with a clear conscience.

If one comes back empty, data is not what is missing. In practice the empty one is the second.

The answer is a version of “let’s watch it for a while,” which means no number will ever trigger a restart, because no threshold exists to cross.

For a Japan investment specifically, there is a further reason the second question stays empty. HQ often lacks a Japan baseline to set a threshold against, and reaches for a global one.

A conversion rate or sales-cycle target calibrated on markets where a single buyer signs will not describe a purchase where the executive queue is the largest source of elapsed time. Setting the threshold from Japan’s own prior quarters is less satisfying and considerably more useful.

Where the threshold itself is the problem, the sequence is covered separately in before you add another Japan dashboard, write the threshold. This note sits one step earlier, at the point where the work has not been allowed to start.

Write a restart date, or move it to “not this year”

Two rules are enough.

Any Japan initiative placed on hold gets a restart date, an owner and one number to check on that date. The date goes in a calendar, not in meeting minutes.

Anything that cannot be given those three moves out of “on hold” and into “not this year.”

Where “not this year” lives does not matter: the initiative tracker, a ticket board, a planning page.

What matters is that it exists as a status, and that the call is made by the executive who owns the budget for the initiative. Left to the person who proposed it, an initiative tends to stay “on hold.”

The reason to be strict is that held items are invisible in the plan.

Cancelled work counts as zero and stops consuming attention. Held work consumes a few minutes of every review, sits in the country manager’s head, and delivers the same result as cancellation with the revisiting cost added on top.

One approach that did not work: a standing list of held Japan initiatives, reviewed monthly. The review became the place where deciding was permanently optional. Dates attached to individual items worked better than a list attached to a meeting.

None of this argues against pausing.

If the commitment is reversible, act on partial information and correct later. If it is irreversible, waiting is worth real money. The test is whether you can stop and undo it if it turns out wrong.

Long contracts, hires and migrations cannot be undone and can justify waiting; campaign spend and channel allocation can be reversed next month and should not wait. Either way, the restart date still gets written.

That judgement turns on reversibility, not on how much data has accumulated.

Waiting for better Japan data is a decision only when a restart date and the number to check on that date are written down. Without them, the pause produces the same outcome as cancelling, plus the cost of revisiting it every month.

Pull the list of Japan initiatives currently on hold and check how many carry a date. The ones that do not are decisions you have already made without saying so.

Common mistakes

• Reading a Japanese “we need more data” as risk aversion. It usually describes an approval queue with no named owner. Ask who signs, not what else to send.
• Setting the restart threshold from a global benchmark. A number calibrated on single-signer markets will not be crossed in Japan, so the hold becomes permanent by construction.
• Holding the headcount while holding the programme. The team stays booked, the pipeline that would have produced the evidence never runs, and next quarter’s review has less to look at than this one.

Related reading

When the stated reason for holding is that the numbers cannot yet be allocated precisely, more precision will not change the call: Japan attribution: agree the decision before debating the weights.

Other notes on the same problem are collected under Revenue data.