Service 03 — ¥29,000
A clear picture of your administrative work before any automation decisions
A four-week review of recurring tasks in a department — what they are, how long they take, how much they vary, and how much judgement they require. The result is a proportionate view of where automation could help and where it could not.
Back to HomeWhat this engagement delivers
An honest account of your own work, with measured figures
Measured, not estimated
The task register records actual time figures from observation and interviews, not estimates from memory. Duration, frequency, and variability are captured across the four weeks, not inferred from a single conversation.
Proportionate view
The output states an honest proportion of the department's administrative time that could be handled by automation, alongside a clear account of what could not — and why.
Risk noted explicitly
Where automation would introduce more risk than it removes — because of variability, required judgement, or regulatory sensitivity — that is stated plainly in the register, not omitted to make the figures look better.
The situation this addresses
Broad claims about automation, and no clear picture of your own situation
Organisations considering AI frequently receive claims about what percentage of administrative work could be automated. These figures are drawn from surveys across industries and organisation sizes that may have little in common with yours. They are not wrong, exactly, but they do not describe your department.
The difficulty is that before commissioning any integration work, you need to know what is actually there — which tasks occur, how much time they consume across the team, and how much of that time involves decisions a person needs to make. Without that, any discussion of automation is speculative.
What tends to be unclear before a review
Actual time consumption
People underestimate how long routine tasks take, particularly tasks that interrupt other work. The aggregate across a team in a month is rarely known until it is measured.
Hidden variability
A task that looks repeatable often has a tail of exceptions. That variability matters for automation — a system handles the common case, not the exceptions, and the exceptions need to go somewhere.
Embedded judgement
Some tasks look administrative but contain a decision that depends on context the system would not have. Identifying those tasks before building anything saves a considerable amount of effort later.
The approach
Four weeks of observation and structured interviews
The review works by direct observation and structured interviews with the people who do the tasks. We record what each task involves, how often it occurs, how long it takes across the team, and what decisions — if any — it requires.
The observation period is four weeks, which captures monthly recurrences as well as weekly and daily ones. Tasks that only occur at month-end would be missed by a one-week review; this scope avoids that gap.
The output is a task register — a structured document that your team can use as a basis for any subsequent automation discussion, whether with us or with another supplier. It belongs to you and is written to be readable without specialist knowledge.
What the task register contains
Each task entry records
A plain description of what the task involves, written to be understood by someone who does not currently do it
How often the task occurs — daily, weekly, monthly, or triggered by another event
Measured time per occurrence and aggregate across the team per month
How often the task differs from its standard form, and what causes the deviation
Whether the task contains a decision that requires human reasoning, and if so, what kind
A plain note on automation suitability — including cases where it would introduce more risk than it removes
How the four weeks proceed
Observation first, then analysis, then a readable document
Week 1
Scope and access agreed
We confirm which department or teams are in scope, agree on observation access, and conduct initial interviews with two or three people to map the task landscape.
Weeks 2–3
Observation and measurement
Structured observation of tasks as they occur, with time recording and follow-up questions where a task's details are unclear. Monthly tasks captured as they fall within the window.
Week 4 — first half
Register draft and review
A draft task register is shared with your team for factual correction. If a task is described incorrectly or a figure looks wrong, this is the point to say so.
Week 4 — second half
Final register delivered
The completed register is delivered, including the proportion suited to automated handling and the note on tasks where automation would introduce more risk than it removes.
Investment
¥29,000
Fixed price for the full four-week engagement. Covers observation, structured interviews, task register construction, one round of factual review, and the final delivered document.
The scope covers one department or equivalent organisational unit. If the review needs to cover multiple departments with distinct task sets, we discuss that before confirming the engagement.
What is included
Four weeks of structured observation and interviews with your team
Task register with measured frequency, duration, variability, and judgement figures
Honest proportion of tasks suited to automated handling — not adjusted to appear favourable
Explicit note on tasks where automation would introduce more risk than it removes
One round of factual review by your team before the register is finalised
What the register produces
A document you can use independently of us
As a decision basis
The register gives you measured figures to use when evaluating any automation proposal, from any supplier. You are not dependent on a vendor's assessment of your own work.
As a prioritisation tool
The tasks that consume the most time and carry the least variability are visible in the register. If you choose to pursue automation, you have a clear starting point rather than having to guess which tasks to address first.
As an organisational record
Many organisations have no written record of what their administrative staff actually do and how long it takes. The register addresses that gap regardless of whether automation follows.
The review is suited to organisations that have received broad automation claims and want a specific picture of their own situation before making any commitment. It is not designed to produce a business case in favour of automation — it is designed to produce an accurate picture, which may or may not support one.
Our commitment
The register reflects what we observed, not what we expected to find
We do not produce task registers calibrated to support a particular outcome. If the review shows that most of your administrative work involves the kind of judgement that automation handles poorly, the register says so. If it shows a clear candidate for a scheduling or translation integration, it says that too.
The factual review step before finalisation is your opportunity to correct any task description or figure that does not match what your team knows. We revise where the facts require it; we do not revise assessments because they are inconvenient.
Initial conversation
This service is suited to organisations that have encountered claims about automation potential and want to verify those claims against their specific situation before acting on them.
If you are uncertain whether a task review is the right starting point — or whether you would be better served by moving directly to an integration engagement — we are glad to discuss that before any commitment is made.
Arrange a conversationGetting started
What happens after you get in touch
Step 1
You send a note
A brief description of the department or team, a rough sense of what kinds of tasks you expect are there, and what has prompted the question now.
Step 2
We respond within two days
An initial response on whether the scope sounds appropriate for this service, and a suggested time for a short call to confirm access arrangements.
Step 3
Scope is confirmed
A written statement of which department is in scope, the observation approach, and the timeline. No obligation until this is agreed by both parties.
Step 4
Review begins
Initial interviews in week one, followed by observation. Your team's participation is structured to take as little time from their regular work as possible.
Back-Office Task Review — ¥29,000
Before committing to automation, know what your work actually contains
If you have received broad claims about automation potential and want a measured view of your own situation, we are glad to discuss what a review would involve. No obligation attached to the conversation.
Get in touchOther services
Other areas we work in
Service 01 — ¥41,000
Scheduling System Implementation
Automated scheduling built around your actual constraints — shift rules, skill coverage, individual preferences — with a parallel comparison period. Ten weeks.
View serviceService 02 — ¥32,000
Translation Workflow Integration
Assisted translation for recurring business documents between Japanese and other languages, with a maintained terminology base and a defined review stage. Seven weeks.
View service