Forecast your Japan pipeline on approval progress, not rep confidence

Your Japan pipeline is full of deals at 90%. The local team reports every conversation as positive. Then the quarter closes and half of it moves to next quarter.

The reps are not sandbagging or over-promising. Confidence is being set by their read of a person who cannot approve anything.

Run a second axis next to the sales stage that tracks how far the approval has actually climbed, and forecast on that.

Positive meetings are not decisions

Japanese B2B marketers describe the same gap from their side. In a survey by ProFuture and Macromill conducted on 3 and 4 February 2026 (n=103, employees working in B2B marketing roles in Japan; a self-run survey by a B2B marketing services provider with a small sample, so read it as directional), only 9.7% said their programmes reach decision-makers adequately.

The respondents are sellers rating their own marketing, not reps reporting on individual deals, so it does not tell you who your team has met. It does suggest that, seen from the selling side, the signer is the hard person to reach.

Your CRM says “in negotiation.” That does not mean anything has moved inside the account.

In Japan, the gap between evaluator and approver is structural

A joint survey by IDEATECH and Hiroyasu Kitagawa (published April 2026, n=307, limited to buyers who participated with at least one colleague in purchases of 3 million yen or more, so it reflects larger multi-stakeholder deals; it is also a vendor-sponsored survey) found that 60.9% of companies require two approval layers and more than 80% require two or more.

The functions involved are information systems at 45.6%, the business unit at 37.5%, and corporate planning at 29.0%.

Your champion’s yes sits below those layers. With two layers of approval, two more people still have to say yes, and headquarters is reading the first one as the deal.

Add the second axis

Do not try to calibrate rep confidence. Add an axis.

Define the approval ladder as stages your local team can observe: the working level agrees, the department head approves, the executive approves, the approval is complete. Keep the sales stage exactly as it is.

Take the lower of the two.

A deal at 90% sales confidence sitting at “department head approves” is carried at that rung’s weight, not at 90%. That single rule stops the higher number from setting the forecast.

Write the exclusion rule down by deal size, before you need it. If every deal must reach executive approval, deals small enough for a department head to sign drop out of the commit for no reason.

Set the required rung per deal-size band (this band closes at department head, this one at executive), then write: a deal that has not reached the rung its size requires is not in this quarter’s commit.

It stays in the pipeline, and it stays in the weighted forecast at its rung’s weight. It only leaves the commit.

If your CRM has forecast categories, express the rule there: below the required rung, never Commit.

Decide who updates the axis and when. The rep updates the approval rung on the same screen when moving the sales stage, and the manager checks rung and evidence in the weekly forecast call.

That adds one field for the rep, so look for an optional field you can retire at the same time.

Switch at the start of a quarter. Changing the logic mid-quarter breaks the comparison with past forecast accuracy.

For the first quarter, report the confidence-based number and the approval-based number side by side, check which landed closer, and only then retire the old one.

Stalled deals do not come back

Ebsta and Pavilion’s 2024 benchmark (n=4.2 million opportunities; an international study, not Japanese data) found that 44% of deals slip, and slipped deals win 67% less often.

Their 2025 study (n=655,000 opportunities, also international) puts the actual win rate at 13% for deals that slip by a month and 8% for deals that slip by three.

A deal held in the forecast while the approval sits still is not deferred revenue. It is revenue that is quietly going away.

What headquarters should ask instead

• Stop asking “what is the confidence on this deal.” Ask whose approval is missing. If the local team cannot name a person, the deal is not positive, it is unobserved.
• Stop treating a slipped Japanese deal as a coverage problem. More pipeline does not fix approvals that never climb. Adding leads adds champions, not signatures.
• Stop forcing the global stage model to carry both jobs. Sales stages track what your team did. The approval axis tracks what the buyer decided. One field cannot hold both, and merging them is why the Japan number is never right.

Build the approval axis into the local process alongside Japan GTM and messaging work, not as a reporting patch on top of it.

Record the buyer’s approval progress separately from the rep’s assessment. Agree which missing approvals exclude a deal from the committed forecast.

Related reading

What to put in front of the approver to move that second axis is covered in Japanese enterprise sales: references matter alongside features.

Other notes on the same problem are collected under Sales process.