Enterprise picture · the transaction ledger
One week of documents, sampled
This is the work itself. 18,010 documents drawn as a 2 percent stratified sample from the 900,433 that Brightvale processed in the week of Aug 10, 2026 — every tower, every document type, in the published mix. Each row carries its counterparty, its state, how it was handled, what it cost in touches and hours, and whether it cleared inside the service level for its type.
Health
What the week looked like
Value is gross document value, not revenue. One sale is documented three or four times as it moves — order, delivery, invoice, cash — and record-to-report and legal documents carry notional amounts that never move cash.
1,535 documents raised an exception at some point in the week. That is a larger number than the 877 still open at the close, because most exceptions are raised and cleared inside the same week. A touchless document carries no named owner, which is why only 4,227 of the 18,010 have one.
Operations
Which tower did the work
Twelve towers processed the week. Open the tower to see its document types, its queues and the exceptions it raised.
By tower
12 towers · 18,010 sampled documents · $197.0m gross document value
| Tower | Share of the sample | Documents | Types | Value | Touches | Avg cycle | Not closed | |
|---|---|---|---|---|---|---|---|---|
| S2PSource-to-Pay | 2,842 | 6 | $39.3m | 1,258 | 9.7 h | 295 | Open | |
| H2RHire-to-Retire | 2,533 | 6 | $3.0m | 1,899 | 14.5 h | 383 | Open | |
| O2COrder-to-Cash | 2,151 | 7 | $59.1m | 1,073 | 9.3 h | 230 | Open | |
| R2RRecord-to-Report | 1,750 | 5 | $29.9m | 1,270 | 7.8 h | 288 | Open | |
| ITSIT Service | 1,706 | 4 | $505k | 829 | 10.0 h | 180 | Open | |
| M2DMake-to-Deliver | 1,622 | 6 | $25.8m | 866 | 8.5 h | 175 | Open | |
| AIOAI Operations | 1,369 | 3 | $0 | 720 | 2.4 h | 140 | Open | |
| C2LConcept-to-Launch | 1,029 | 4 | $6.0m | 691 | 2.0 d | 139 | Open | |
| A2LAwareness-to-Lead | 1,023 | 4 | $13.6m | 609 | 16.9 h | 139 | Open | |
| LGLLegal Operations | 813 | 4 | $8.0m | 610 | 44.1 h | 119 | Open | |
| D&AData and Analytics | 668 | 3 | $0 | 367 | 12.2 h | 67 | Open | |
| P2PPlan-to-Perform | 504 | 4 | $11.9m | 263 | 13.1 h | 48 | Open |
"Not closed" counts documents in any state other than completed — exception open, waiting on a person, in flight or failed. Across the sample that is 2,203 documents, 12.2% of the week.
Document types
The sixteen largest of 56 types in the sample
| Document | Tower | Sampled | Value | Average | Avg cycle |
|---|---|---|---|---|---|
| Payroll run item | H2R | 947 | $1.3m | $1k | 4.2 h |
| Supplier invoice | S2P | 856 | $8.9m | $10k | 12.0 h |
| Incident ticket | ITS | 606 | $0 | $0 | 2.0 h |
| Purchase order | S2P | 573 | $9.0m | $16k | 2.9 h |
| Agent run exception | AIO | 567 | $0 | $0 | 2.2 h |
| Service request | ITS | 566 | $505k | $891 | 12.1 h |
| Journal entry | R2R | 545 | $12.0m | $22k | 5.2 h |
| Sales order | O2C | 510 | $14.5m | $29k | 1.8 h |
| Customer invoice | O2C | 482 | $14.1m | $29k | 4.5 h |
| Purchase requisition | S2P | 460 | $7.0m | $15k | 3.8 h |
| Time correction | H2R | 452 | $68k | $151 | 7.4 h |
| Shipment | M2D | 435 | $8.3m | $19k | 4.3 h |
| Model routing decision | AIO | 417 | $0 | $0 | 58 min |
| Goods receipt | S2P | 408 | $5.5m | $13k | 5.8 h |
| Delivery note | O2C | 402 | $11.9m | $30k | 3.1 h |
| Prompt policy check | AIO | 385 | $0 | $0 | 4.2 h |
Live observability
Where the week came from
Region, source system and currency, each counted over the same sampled documents.
Region
An exhaustive partition of all 18,010 documents
Source system
21 systems produced the sample
These are the same system keys the estate registers on the systems of record page. 21 of them appear in the sampled week; the rest hold reference or master data that this week did not produce a document against.
Currency
20 currencies · amounts above are converted to US dollars
Every amount on this page is stated in US dollars. The document currency is carried on the row, but no rate table is modeled, so the dollar amount is the stored figure rather than a conversion anyone could re-perform.
Actions
What is waiting on a person
Every count below is a document sitting in a state that will not clear on its own. They are drawn from a two percent sample, so the real population behind each one is roughly fifty times larger.
Explain a missed commitment
Cycle time exceeded the service level recorded against that document type.
Find why a document went fully manual
No automation touched it. Each one is a repeatable case that was never built.
Work an open exception
The document stopped on a named exception code and nothing further happens until somebody resolves it.
Chase a document waiting on a person
Processing is complete up to a human step. The clock is running and the queue behind it is growing.
Watch a document still in flight
Moving through its queue at the close of the sample week. Not late yet.
Re-drive a failed document
The document failed outright. It is not in any queue and will not be retried without an instruction.
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 can read the sampled week: what was processed, by which tower, in what state, at what cost in touches and hours, and against which counterparty. It cannot post, reverse, approve or release any document. There is no write path from this platform into the modeled enterprise, and there is no connection to a real one.
Live observability
What the record shows right now
Document state at the close of the sample week, an exhaustive partition of all 18,010 sampled documents: every document carries exactly one state.
Current distribution
18,010 documents
Is policy and strategy coming to fruition
Whether the written intent is holding here
76.5% of the sampled week ran touchless and 86.9% cleared inside its commitment.
Nothing on the record settles this
The estate publishes a touchless rate and a set of service levels as its standing position on how work should run. Measured on the sample rather than asserted, 13,783 of 18,010 documents ran touchless, 2,588 were assisted and 1,639 were fully manual — 76.5% touchless, against the touchless rate the estate publishes. On commitment, 15,657 cleared inside the service level recorded for their document type and 2,353 did not. 1,535 documents raised an exception at some point across 91 distinct codes, which is a larger number than the 877 still open — most exceptions are raised and cleared inside the week. That agreement proves nothing. The generator was given the published document mix and the published touchless rate per tower as inputs, so the sample was always going to land near them — it corroborates the arithmetic, not the position. Nothing here is measurement of a real operation.
Brightvale Consumer Group is an invented company. Every brand, site, customer, supplier, item and document behind this page was generated from a fixed random seed, and no figure is drawn from any real company accounts, filings or systems. What is real is the shape: a monolithic consumer goods group with one ledger and one shared services organization, and a document mix and volume that matches the estate this platform already publishes.
The ledger is a sample, not a week. Each tower contributes two percent of the documents it processes in a week, drawn across its document types in the published mix, so the counts here are roughly one fiftieth of what actually moves. The queue day series beside it is different: it is full population daily arrival and service counts for every queue. Multiplying one by fifty does not turn it into the other.
Gross document value is $197.0m for the sample, which annualizes to a far larger number than the group revenue on the company page — and that is expected, not an error. A single sale is documented as an order, a delivery, an invoice and a cash application, so its value is counted four times here; record-to-report journals and legal documents carry notional amounts that never move cash at all. Read this figure as document throughput, never as revenue.