Feature

What the week actually did

Created against closed, what the backlog did, who is carrying what, and how long things take. Delivered Monday morning, or on request - with a short list of what is worth fixing, each one showing the figure it came from.

Computed, not written Against last week Recommendations with evidence

Book a free demo See pricing

A weekly review showing created, closed, backlog and per-person figures

In short

The weekly review reports what the business did in a week: jobs created and closed, how the backlog moved, what is overdue, how long work takes to close, how many daily updates arrived, and a breakdown per person and per department - each compared with the week before. Alongside it sits a short list of recommendations, each one citing the figure that produced it and linking to the screen that fixes it.

Delivered: Monday morning, or on request
Compared with: The previous week
Breakdown: Per person and per department
Computed by: Database queries, not the model
Recommendations: Rule-based, evidence attached

What it reports

The figures that actually tell you something

Created against closed

Whether the list grew or shrank, and by how much. The single number most owners have never had.

How long things take

Average and median time to close, and how much was closed the same day.

What is ageing

The oldest open job, what is still overdue at week end, and what is blocked.

Per person

Closed, still open, overdue, average close time, how often they needed chasing, and missed updates.

Per department

The same, by the departments you defined - which is usually where the pattern is.

Daily update rate

What share of expected updates arrived. Everything else in the report depends on this one.

Why not let the model write it

Counts: From the database
Comparisons: From the database
Recommendations: Rule-based
The model's job: Understanding the request
Checked against: The job list, every release

The recommendations

Advice you can check the arithmetic on

Recommendations are rule-based, computed from the week's figures. That means the same data always produces the same advice, and every recommendation shows the number it came from - so you can disagree with it on the evidence rather than wondering where it came from.

They are ordered worst first, and each one links to the screen where you would act on it.

"More work is arriving than is being finished - 21 created against 2 closed, the list grew by 19"
"Daily updates are not coming in - 8% of expected updates arrived"
"A lot of open work has no due date - 7 of 19 open jobs have no deadline"
"Staff keep asking things the written procedures do not cover - 4 questions asked more than once"

Works together with

More of the assistant

Reads the daily plan Someone types what they are doing today, the way they always have. It becomes separate tracked jobs with owners and deadlines - no forms, no commands to learn. Keeps the photo with the job A picture of the damage, the reading or the finished work is attached to the job that message created - not lost in a chat nobody can search. Listens to voice notes A technician holds the button and talks. It is transcribed, handled exactly like a typed message, and repeated back so a mis-hearing is obvious straight away. Creates the work that repeats Temperature logs, cleaning, safety checks, calibration, a weekly recall list. You describe the pattern once and the job appears on its own, chased like any other. Reads the end-of-day update "The rotors came in, the Ford is stuck waiting on parts." Matched against that person's open jobs by meaning - one closed, one blocked with the reason recorded. Chases what is late Escalating reminders to the person, never the group. After your chosen number of attempts it stops nagging and tells the owner instead.

See every feature