Your Japan adoption number is green because it counts logins, not decisions

Published Aug 9, 2026

The global adoption dashboard shows Japan above 90% weekly active users, and the rollout is reported as complete.

In the same month, the regional forecast for Japan still cannot be read off the system. Someone has to get on a call with the country manager and collect it verbally.

Both are true at once, and they are not in tension. An active-user or login metric counts the screen being opened. It does not count what was decided from it, and in a Japan entity those two numbers diverge further than they do in most regions.

Ask whether the tool is used in Japan and almost everyone says yes

Any metric built on usage will pass in Japan. Local teams comply with instructions from HQ, so a compliance-shaped metric returns a compliant number.

A survey by Keywalker (fielded 8 to 10 October 2025, published 11 November 2025, online panel survey run by PRIZMA, n=1,034, frontline staff and managers in sales departments at Japanese companies that have already deployed an SFA, CRM or BI tool; Keywalker sells BI implementation services, so read it as a vendor survey) asked what people actually use their dashboards for. Only 2.2% of managers and 1.9% of frontline staff said they hardly use them at all. On a usage metric, 98% of the Japanese market passes.

What that 98% is doing is the part HQ does not see. The question allowed multiple answers, so the figures below sum to more than 100%.

Checking numbers for the sales meeting: managers 58.2% / frontline 48.6%
Reviewing their own or their team’s activity: 46.6% / 43.6%
Checking progress against target: 41.0% / 39.5%
Preparing reporting material for a superior or executive: 35.9% / 26.0%
Deciding the next action: 15.4% / 20.7%

The dominant uses are confirmation and reporting. Deciding what to do next is the least common use for managers, at 15.4%.

In Japan, opening the dashboard is usually a document-production step

The third and fourth lines above are worth separating out, because they are where the Japan entity differs structurally from a US or UK region.

Approval in Japan is layered. In a survey by IDEATECH with Hiroyasu Kitagawa (fielded 23 to 26 March 2026, published 16 April 2026, n=307, employees and executives who were among two or more people involved in evaluating a B2B purchase with an annual contract value or one-off implementation cost of 3 million yen or more; a survey-PR study), 60.9% reported a two-stage approval and more than 80% reported two stages or more.

That structure changes what a dashboard is for. The Japanese user of your CRM frequently is not the person who decides. Their job is to move a number upward in a document that another layer will read. So they open the dashboard, extract the figures, and paste them into a deck or a written proposal. From HQ’s telemetry, that session is indistinguishable from a manager reprioritising a territory.

The consequence is specific: the region where your adoption metric looks best can be the region where the fewest decisions are being made inside the system, and HQ has no signal that separates the two.

What to measure instead in a Japan entity

Three replacements. All three can be built from data you already hold.

1. Agendas that can be built from the record alone. Count how many items on the Japan weekly sales meeting agenda could be assembled from the CRM without asking a rep. An item that requires a verbal answer is an item the record did not support.

2. Staleness of open deals. Count open deals whose last meaningful update is older than seven days, as a count and as a share. A login metric is satisfied by opening the tool daily even if nothing has been updated for a fortnight. In the same Keywalker survey, 40.2% said they enter data immediately after each activity, 32.2% at the end of the day, 11.2% a few times a week, 3.9% a few times a month, and 12.5% said they either do not enter it or frequently miss entries.

Staleness deserves particular weight in Japan because the cycle hides it. The IDEATECH survey found that 63.6% of considerations ran three to eight months, with roughly 40% at six months or more. In a market with a 30-day cycle, a record that has not moved in three weeks is conspicuous. In a six-month Japanese cycle, it looks normal, and a deal that has quietly stopped is indistinguishable from one that is progressing slowly.

3. Deals that progressed outside the record. Count the deals discussed verbally on the regional call that have no corresponding update in the system. A high count means the record sits outside the decision path, whatever the login rate says.

Entering data without using it is a distinct and measurable state. In a survey by Hammock (fielded 30 September to 2 October 2021, n=305, executives and board members at Japanese companies with 300 or more employees that had deployed an SFA; an SFA vendor’s own survey, and several years old now), 7.5% said they only entered data. On an active-user metric, that 7.5% is indistinguishable from a team running its business out of the system.

Pick the metric that inconveniences the local user, not the one that reassures HQ

Of the three, staleness of open deals is the one that changes behaviour, and the reason is structural rather than cultural.

A login rate inconveniences nobody. It reassures the person writing the rollout report. A published count of open deals not updated in seven days inconveniences the rep who owns them, because those deals stop being supportable in the weekly meeting. Adoption moves when the cost of not recording lands on the person who wants the deal to move.

Choosing the metric is also a statement about why you are asking for the data. In the Keywalker survey, the top driver of willingness to enter data was “the purpose of the entry and how the data will be used is clear” at 54.8%, followed by “my own results and growth become visible” at 40.1% and “what I enter is connected to a KPI” at 36.5%. A region measured on login rate has been told, in effect, that the purpose of the data is supervision.

The principle

Measure Japan adoption by the number of decisions produced from the record and by how stale the record is, not by how often it was opened.

Before the next regional review, ask your Japan lead to build the whole agenda from the CRM. Whatever they cannot build is not a training gap. It is a field that no decision depends on, and you can retire it.

Getting the local operating model right, rather than shipping the global one and measuring compliance with it, is covered in Japan Market GTM and Messaging.

Once you are counting decisions rather than sessions, the next question is what each number has to reach before anyone acts. That is covered in before you add another Japan dashboard, write the threshold.

Other notes on the same problem are collected under Change management.