Closing the loop
Getting to checked
148 strategy positions carry no verification check of any kind, out of 209. This page ranks every one of them by what the sub-function underneath it actually runs, lays the declared capacity of 5 checking parties over that order across 3 periods, and then does the arithmetic that says how far short of the target the whole plan still lands. Independent verification is measured at 24.7% right now, and nothing on this page has changed it.
From 24.7% to a stated figure
One measured figure and two pieces of arithmetic. The arithmetic is drawn in outline because it describes something that has not happened.
Who would do the checking
A party is independent of the team that authored the subject, not of the company. Declared capacity is a number written here, not a commitment anybody made.
The order of work, and why it is this order
Every unchecked position, ranked. Weighted 60% on weekly run cost, 25% on items a week and 15% on policy items written under the same sub-function.
When the plan says it would happen
Three whole periods ending on or before the target date. Capacity is filled from the top of the rank order down.
What this plan does not reach
The parts of the independence problem that no amount of position checking closes.
Where the plan and the estate now disagree
The plan was written against the estate at one moment. The rank is recomputed on every load and any disagreement is published here rather than quietly corrected.
Actions
What is waiting on a person
Every line below is work that has to be done by a person before a single measured figure on this platform changes. Naming a party against a position is not one of those things, and none of these can be closed from this page.
claimed benefit lines with no verification record against them
190 of 655 lines carry a check. The plan on this page does not touch them: it ranks and assigns strategy positions only. The independence target cannot be reached from positions alone, so most of the records that would move it have to come from here.
self-checks on the record that no later check takes back
A check signed inside the authoring team stays in the denominator forever. That is why reaching 60.0% needs 222 new independent records rather than 89, and why the arithmetic gets worse every time somebody signs their own work.
independent records still needed after every assignment in this plan completes
The plan holds 60 assignments. If all of them were completed as independent records the measured figure would read 39.2%, against a target of 60.0%. The plan is a start on the problem, not a solution to it, and the page says so in the same place it shows the projection.
unchecked positions the plan reaches nobody for
148 positions carry no check at all and the declared capacity of the five named parties across three periods is 60. The rest sit below the line with no party and no date, and they are still listed here in rank order so nobody has to pretend they were considered.
assignments named and not begun
0 are complete and 0 carry an evidence reference. There is no route on this platform by which a party accepts an assignment or files the record that would close one, so every row sits in the state the plan wrote.
Operations
What this desk is allowed to start
A surface that only reports is not operable. This is the work this page can set in motion, and the bound it runs into.
Trigger and bound
This page allocates nothing and completes nothing. It reads every unchecked position, ranks them by three figures taken off the sub-function each one governs, and lays a declared capacity over the top of that order. Writing the independent record is done outside this platform, by a person, and until that record exists the measured figure does not move by a fraction of a point.
Live observability
What the record shows right now
The gap in records, not in percentage points. Each bar is a count of verification records: what is independent today, what the plan would add, what would still be missing after that, and the self-checks sitting in the denominator that none of this work removes.
Current distribution
561 records
Is policy and strategy coming to fruition
Whether the written intent is holding here
No. 24.7% is the measured figure and nothing in this plan has moved it.
Not holding on the record
The policy here is that a position should be checked by somebody outside the team that wrote it, and the target attached to that policy is 60.0% by 2027-06-30, owned by Grace Abbott, Head of Internal Audit. It is measured at 24.7%, which is 62 of 251 checks, and 0 of 6 reach targets across the whole estate are at or above their number. Every one of those independent checks was signed by Grace Abbott, Internal Audit, and 4 of the 5 parties this plan names have never signed a verification check of any kind. So the strategy is not coming to fruition, and this wave did not make it come to fruition either: it worked out what would have to happen, in what order, and by whom, and then wrote that down. The ceiling is the part worth reading twice. If every one of the 148 unchecked positions were checked independently tomorrow the figure would read 52.6%, still short of the target, because the 189 self-checks already on the record never leave the denominator.
What this page is, and what it is not
An assignment is a name written against a piece of work. It is not a check. Naming Internal Audit against a position does not make that position verified, and it does not move the independent verification percentage by a fraction of a point. Only an independent verification record, written by a party outside the team that authored the subject, moves that figure, and this platform does not write one of those on its own. Every row in this plan is in the state the plan put it in. Not one is complete, no evidence reference is attached to any of them, and the completion column exists so that the day one of them is finished the record can say so.
The order of work is a choice, so here is the choice. Three figures are read off the sub-function each position governs: the weekly run cost, which is units a week multiplied by cost per unit; the number of items a week passing through it; and the number of policy items written under the same sub-function. Each is scaled against the largest value in the unchecked list and the three are combined at 60, 25 and 15 percent. Those weights were chosen by a person and are written here so somebody can disagree with them. The policy count is a count of rules written under the same sub-function; it is not a declared dependency on this position, because no dependency edge of that kind exists anywhere in the estate. None of these three figures is a measure of how likely the position is to be wrong. Nothing on this platform measures that.
The projected figure is arithmetic on an assumption that has not happened. It asks what the independent verification percentage would read if every assignment in this plan were completed as an independent record, and nothing more. It is not a forecast, nobody has committed to it, and it is drawn in outline everywhere it appears so that it is never mistaken for a measured figure. Until an independent record is written, the measured figure is the only figure.
Every independent check on this record was signed by one person. That is not a governance model, it is a single point of failure, and it is the reason this plan names other parties at all. The parties below that have never signed a check are marked that way, and naming them here is not their agreement to do the work: there is no route on this platform by which a party accepts an assignment, which is why every row sits in the same state.