FSCP

A close status of 92% completetells nobody what to do next.

FSCP is an operating framework for the financial statement close: eight domains, 56 sub-processes and 204 metrics, every one of which answers a single question. Is there a close blocker here, and how severe is it. Not a percentage, not a task list, but a count, a value and a colour that a controller can act on this afternoon.

How Discover works

Scores your close, does not run it
Every metric is a blocker, not a statistic
Read-only on top of the ERP you already run

One close cockpit

Eight domains, one close cockpit

Eight domains → DataTwin → the register of close blockers

Live process

Eight domains, 56 sub-processes

IC28 KPIs
CF28 KPIs
GA39 KPIs
TF20 KPIs
RC33 KPIs
FP24 KPIs
PC15 KPIs
IR17 KPIs

DataTwin

Scores 204 metrics against your close

What is blocking this close

CF — Journals unposted14 of 140
TF — GR/IR unmatched82 items
GA — Accruals not raised9 of 61
IC — Confirmations open3 entities
RC — Sign-offs pendingclear

The dashboard rule Show only close blockers: pending, mismatch, difference, breach, missing reference or unresolved approval. Every metric declares whether there is a blocker, and how severe it is.

Read-only to start. Nothing changes in your ERP or your close calendar. FSCP runs on the same engine that powers the DARP Framework: acquire, process, govern, report.

The shape

The shape of the framework

Not a dashboard somebody designed to look busy. A structure with a fixed number of places a close can fail, so the same question is asked of every period and every entity.

Domains

8

Sub-processes

56

KPI metrics

204

Colour logic

R/O/G

Browse the 204 metrics

Open the catalogue

Every metric

Every metric declares the same six things

The base table stays deliberately simple. Percentage, severity and the message a controller reads are all derived from it, which is why a new metric is a new row rather than a new report.

Column	What it holds	Example
Domain	Which of the eight it belongs to	Core Finance
Close process	The sub-process inside that domain	General ledger
Metric	The issue being measured, always a blocker	Journals unposted at close
Value	The pending, differing or breaching amount	14
Total count	The denominator it is measured against	140
Colour	Severity, from thresholds you set	Red

Domain

What it holds: Which of the eight it belongs to
Example: Core Finance

Close process

What it holds: The sub-process inside that domain
Example: General ledger

Metric

What it holds: The issue being measured, always a blocker
Example: Journals unposted at close

Value

What it holds: The pending, differing or breaching amount
Example: 14

Total count

What it holds: The denominator it is measured against
Example: 140

Colour

What it holds: Severity, from thresholds you set
Example: Red

And four things are derived, never entered.

Issue percentage is value over total count. Issue exists is value above zero. Severity comes from your thresholds, and the message writes itself: 14 of 140 pending.

The framework

From the design rule to the closes it has changed

Part 1

The design rule: show only what is blocking the close

Most close dashboards report progress. Progress is a comfort, not an instruction. Every FSCP metric is a blocker measure, so anything showing on the cockpit is something a person has to decide about before the books close.

Capture the blockers

Into one cockpit

Pending, exception, mismatch, breach and ageing values from every domain brought into one place, at the same grain.

Pending items, mismatches, breaches and ageing, in one view
Every domain scored on the same day, on the same basis
Sourced from the systems, not from a status meeting

Prioritise by materiality

Value, not count alone

A count says how many. A value says how much. Together they separate small operational noise from something that will move the statements.

Count and value carried on every metric
Small operational issues separated from close-impacting risk
Severity from thresholds you set, not from a fixed scale

Resolve before final close

Each colour is a work item

Every red and orange becomes a work item for review, approval or correction, with an owner, rather than a line in a report.

Each blocker routed to the person who can clear it
Cleared blockers recorded against the period they blocked
What was accepted rather than fixed, and on whose authority

The eight domains

The eight domains, and what each one is watching for

Each domain owns a set of sub-processes and a set of blockers. The three listed under each are the headline ones; the full 204 are in the catalogue.

IC · 7 sub-processes · 28 KPIs

Intercompany Finance

Eliminate intercompany billing, booking, settlement and confirmation blockers.

Billing and AP booking pending
Settlement and cash application gaps
Balance confirmation differences

CF · 5 sub-processes · 28 KPIs

Core Finance

Protect the core books: GL, AP, AR, fixed assets, bank and cash.

Unposted journals and TB differences
Blocked invoices and overdue AR
Bank reconciliation gaps

GA · 13 sub-processes · 39 KPIs

GA Close

Drive period-end close activities from master data through to P&L review.

Sub-ledger close gaps
Accruals, provisions and reclass pending
P&L variance and close task breaches

TF · 4 sub-processes · 20 KPIs

Transaction Finance

Control transactional close flows across P2P, O2C, inventory and payroll.

Three-way match and GR/IR exceptions
Unbilled revenue and billing blocks
Inventory and payroll posting issues

RC · 8 sub-processes · 33 KPIs

Reporting & Compliance

Finalise post-close reports, schedules, controls and the consolidation package.

Reports and schedules pending
Control evidence and sign-off gaps
GAAP and elimination entries pending

FP · 8 sub-processes · 24 KPIs

Financial Planning

Close budget, forecast and review cycles with approved management inputs.

Budget inputs and approvals pending
Variance explanations open
Forecast accuracy and margin gaps

PC · 3 sub-processes · 15 KPIs

Planning & Control

Validate cost, profit-centre and FP&A control views before management close.

Unallocated costs and mapping errors
Profit-centre loss alerts
MIS packs and actions open

IR · 8 sub-processes · 17 KPIs

Intelligence & Risk

Surface anomalies, recovery opportunities and audit-trail risks early.

Critical anomalies detected
Recovery actions pending
Manual touch and unauthorised changes

The full 204-metric catalogue

The same four stages

The same four stages the rest of the platform runs on

FSCP is not a separate system with its own logic. It is the DARP Framework pointed at the close, which is why the rules that recovered money in Discover are the rules that score the close in Prevent.

1 · Discover

Ingest, cleanse and harmonise across every stream. Detect anomalies, quantify exposure and rank by materiality.

2 · Assess

Connect ERP, bank, operations and partner silos. Attribute root cause and separate systemic from one-off.

3 · Recover

Route exceptions to accountable owners. Corrections, holds and adjustments in flight, with confirmation logged.

4 · Prevent

Control redesign, early-warning signals and board-ready narratives, looping back into Discover.

Where AI is used here, and where it is not.

AI reads documents, detects anomalies, attributes root cause and writes the narrative. Deterministic rules decide every threshold, every severity and everything that posts to the books. Nothing posts because a model was confident.

How the DARP Framework works

Worked closes

Three closes, and what actually changed

These are engagements, described without naming the customers. The context matters more than the headline: a number without the shape of the business behind it is not evidence.

Accounts payable

Two entities, 25+ locations

Over 5,000 invoices a month on SAP S/4HANA, half PO-linked. Discrepancies took four to five days to resolve; the AP close took five to seven.

Three-way reconciliation across invoice, PO or contract, and receipt
Audit checklist run against every invoice, not a sample
Prepaid, provision reversal and period-end accounting automated

Accounts receivable

20+ entities, 200+ locations

Collections split across gateways, bank deposits, UPI and cash. Matching 20,000 enrolments a month to sales made AR close a ten-day job.

Screenshots, statements and gateway reports read into one pipeline
Match codes for each settlement scenario, across dimensions
End-to-end collection reconciliation and revenue recognition

Inventory

Multi-plant manufacturer

Matching stock movements to GL balances across plants took five to eight days by hand, and provisions lacked the evidence to sign off.

Opening plus receipts less issues plus corrections, against the GL
Ageing and policy-based provision, trued up material by material
Routing across store keeper, material owner, cost accountant, controller

What these three have in common.

None of them started with new software in the transaction path. Each started by rebuilding the reconciliation on history that already existed, which is what made the close time fall. The cycle-time figures are the customers’ own, measured before and after.

Start here

Before you adopt it, we score a close you have already done

FSCP is prevention, and prevention is a promise about the future. The way to test a promise like that is to point it backwards first, at a close you already signed, and see what it would have flagged.

2min

This is how long it takes to understand how much can be recovered.

Send us the close data for a period you have already closed. We score all 204 metrics across the full population, not a sample, and come back with the blockers that were live at the time and what they were worth. Before any contract.

The close data
A period already closed
All 204 metrics scored
The blockers that were live, and what they were worth

What was red

The blockers live on the day you signed, each with its count, value and denominator.

What was missed

Items that cleared after the close rather than before it, and the periods they should have landed in.

What repeats

Exceptions across three consecutive closes, where a control is missing rather than a task.

What it costs

The days spent on tasks a metric would have caught, and where the close calendar actually breaks.

The framework runs the same way for manufacturing, distribution, retail, edtech, pharma and multi-entity groups. The thresholds are yours; the 204 questions are the same.

See all 204 metrics