When Japan reps ask the last owner instead of reading the CRM
HQ asks a reasonable question. The Japan CRM has deal history. It has AI-generated call summaries attached to every meeting.
So why, when an account changes hands, does the new owner message the previous rep instead of opening the record?
Because a summary of what was said is not what the next owner needs. They need to know what to do next, and the summary does not say.
The summary is accurate and still useless for the next move
Opening the CRM should be the safer route. The previous rep may not remember everything, and the record has fewer gaps.
The new owner asks anyway, and not out of laziness. After reading the summary, they still do not know what to do on Monday.
A call summary organises the whole conversation evenly. The moment the buyer leaned forward, or the point the rep knew could not be dropped, sits in the same weight as every other line.
Sales is a job of reading temperature to choose the next move, so this gap shows up quickly.
Ask the previous rep and you get the useful part in thirty seconds: “Talk to IT before the department head. He is still waiting to see.”
Fifteen minutes with the summary may never reach that sentence.
Once a record fails to answer, people stop opening it, and then stop writing in it
The new owner learns from a failed attempt. Open the record a few times, fail to decide, and next time go straight to a person. That is rational.
Then the record is no longer referenced. A record nobody references keeps charging the people who write it.
When writers see that nothing is read, they stop writing. More logging turns into more unused logging.
Japanese survey data points the same way, with caveats.
Mazrica’s SFA/CRM usage survey (fielded by IDEATECH, 25 to 28 November 2024, managers and team leads in the sales departments of B2B companies, n=101) found 39.6% naming “necessary information is not shared properly” as a problem with SFA/CRM.
It is a vendor survey from a company that sells SFA, and the sample is small, so read the figure as an order of magnitude.
It does show that a sizeable group has the tool and still lacks the information.
A second source: Keywalker (online survey by PRIZMA, 8 to 10 October 2025, n=1,034, sales managers and field staff at companies that have adopted SFA, CRM or BI tools; the sponsor sells BI implementation support, so it is also a vendor survey).
Asked what they use dashboards for, 15.4% of managers and 20.7% of field staff chose “deciding the next action”. The top answer for managers, “checking numbers in the sales meeting”, was 58.2%.
Records get used to confirm figures. They are rarely used to choose the next move.
The survey asked about dashboards, not about handoffs, so it does not explain why your Japan team asks the last owner. It does point the same way: the record is not where the next move gets decided.
Keep three lines in one fixed place, and let the person who held the deal write them
The fix is not more transcripts or longer summaries. It is a state in which the next owner can decide in three minutes.
On the deal record, in a fixed place, keep three lines. HubSpot and Salesforce both have a standard next-step field on the deal; use it for the first line and add custom fields only for the other two:
- What to check next, with a date if possible.
- Who holds approval. A role is enough if the name is not known.
- Where it stalled last time. What was missing, and who is waiting for what.
The summary can stay. But with these three lines in place there is a reason to open the record, and one successful decision is what makes the next opening likely.
The person who writes them is the rep holding the deal, at the moment they move the stage or book the next meeting, on the same screen. Written later from memory, the temperature is gone.
Left optional, the lines are blank within six months. Make them a condition for moving the deal to the next stage, not a requirement at creation.
Approval holders are not always known early, so allow “not yet confirmed” as an answer; a deal full of those is telling you what to check next.
Overwrite the lines to keep them current. Their job is to show the present state in three minutes; reading the history is a different job.
If you want to see how the stall point moved, paste the same three lines into the activity note each time the stage changes, and the earlier versions stay.
In the weekly pipeline review, the manager starts with deals where the lines are empty or still “not yet confirmed”.
We covered the same constraint for contract handoffs in What your Japan handoff drops is the reason the contract was approved: only the person who held the deal can write it.
Placement matters more than it sounds. The rep opens the deal screen, the manager opens the pipeline list, customer success opens the contract screen. Put the same three lines where each of those roles actually looks.
Asking AI instead of opening the screen does not remove the need for the three lines
More teams now query the CRM through an AI assistant in plain language, and for some questions that is faster than any screen.
It does not make the three lines unnecessary. It makes them more necessary.
An assistant can only return what the record contains. If the record holds call summaries, the assistant summarises the summaries.
Only when a line says “talk to IT first, the department head is still waiting to see” can the assistant say it too.
Whether the next owner reads the screen or asks through AI, the result depends on whether the decision material sits in a fixed place.
Telling a Japan team to “check the CRM first” will not work
An instruction to look at the record first does little. People pick the fastest way.
The only way to get the CRM chosen is to make opening it, or asking it, faster than messaging the previous rep.
A CRM that is a place to read will not be used. It has to be a place to decide.
A test for this week: pick three Japan deals whose owner changed recently. Ask the new owner to say the next move in three minutes using only the record.
If they cannot, they will message the previous rep, and what that rep says is the content of the three lines.
Related reading
Before reducing how much people enter, make a place where what they enter gets used. That argument is in Japan CRM adoption: examine the work, the records and their use.
Other notes on the same problem are collected under CRM adoption.