bp Sphere Future Finance North Star

Four intelligence systems become the 2030 finance operating model.

This suite connects close, forecast, credit, working capital, controls, capital allocation, portfolio, performance, treasury, tax, carbon, and enterprise value into one governed decision runtime. The dashboard is not the product; the continuously operating decision intelligence underneath is the product.

TrustContinuous close and controls prove numbers before reporting.

PredictForecast and treasury missions update future outcomes continuously.

ProtectCredit and controls detect risk before it materializes.

OptimizeWorking capital, capital allocation, and portfolio missions improve cash and returns.

TransformEnterprise value intelligence links finance to strategic action.

Five Transformational Missions For The Workshop

Do not demo twenty use cases. Demo five runtime-backed missions that progressively prove the Enterprise Decision, Learning, Control, and Impact Runtime. Each mission must show ontology, decisions, controls, evidence, learning, impact, autonomy, and CFO rollups.

Demo 1

Continuous Close Command Center

June 2026 close is seven days from certification. The runtime detects intercompany mismatches, manual journal risk, GR/IR aging, accrual variance, BlackLine delays, and controls at risk before month-end.

Runtime flow: Financial Event -> Close Risk Detection -> Decision Engine -> Recommended Actions -> Execution -> Outcome -> Learning -> Impact

Credible data proof: 72% close readiness, 1.8 day forecasted delay, $9.4M UK Trading vs Singapore Trading mismatch, $42M manual journal risk, $18M open reconciliation exposure, 11 controls at risk.

Open proof surface

Demo 2

Dynamic Credit Intelligence

Customer requests $50M incremental exposure. The runtime evaluates payment history, exposure, commodity prices, external ratings, geopolitical risk, existing limits, collateral, and policy boundaries.

Runtime flow: Counterparty Signal -> Exposure Context -> Policy/Limit Evaluation -> Evidence -> Decision Alternatives -> Human Approval -> Outcome -> Learning -> Impact

Credible data proof: $42M exposure governed, $15M unsecured exposure flagged, $8.4M overdue collection action drafted.

Open proof surface

Demo 3

Intelligent Journal Entry Studio

Controller creates a $25M journal. The runtime checks historical journals, policy, controls, prior approvals, similar entries, materiality, SOD, support evidence, and close impact.

Runtime flow: Journal Draft -> Historical Decision Memory -> Policy/Control Evaluation -> Evidence Pack -> Risk Score -> Approval Recommendation -> Replay -> Learning -> Impact

Credible data proof: Material journal policy, SOD boundary, evidence pack, historical journal comparison, controller approval path, and replayable decision.

Open proof surface

Demo 4

Autonomous Spend Intelligence

Budget owner asks whether another $15M can be committed. The runtime evaluates budget, actuals, open commitments, pending invoices, forecast consumption, policy, cash timing, and supplier risk.

Runtime flow: Spend Request -> Budget/Actual/Commitment Context -> Forecast Consumption -> Policy Check -> Recommendation -> Action -> Outcome -> Learning -> Impact

Credible data proof: $86M future cash impact, $2.4M discount opportunity, DSO/DPO and working-capital tradeoffs, governed action path.

Open proof surface

Demo 5

Enterprise Finance Impact Cockpit

All mission outcomes roll into a CFO-grade impact layer. Value claims drill from enterprise total to domain, case, decision, evidence, outcome, formula, confidence, replay, and learning.

Runtime flow: Decision -> Outcome -> Impact Event -> Value Formula -> Evidence -> Confidence -> Rollup -> CFO Drilldown -> Learning

Credible data proof: $42M value created example: cost reduction, risk avoidance, working capital, control improvement, decision quality, learning effect, and autonomy progression.

Open proof surface

Mission wiring matrix

This table is the credibility guardrail: every mission has named source-data shapes, real agent roles, policies/controls, and measurable impact expectations. Reference rows must be wired to live source evidence before they are presented as attested production proof.

Mission	Source-data shape	Agents	Policies and controls	Measured impact
Continuous Close Command Center	SAP/CFIN journals, BlackLine tasks and reconciliations, GR/IR, accruals, close calendar, policy registry, evidence packs	Close Readiness Agent, Journal Risk Agent, Intercompany Agent, Accrual Agent, Reconciliation Agent, Evidence Agent, Policy Agent, Impact Agent, Learning Agent	R2R-IC-009, R2R-MAT-002, SOX-CLOSE-404, R2R-REC-014, SOD approval gates	Close duration: 8.2 days before, 4.1 days target after sustained continuous-risk resolution.
Dynamic Credit Intelligence	SAP AR, credit system, customer master, ratings feeds, commodity feeds, geopolitical feeds, treasury exposure	Credit Intelligence Agent, Exposure Intelligence Agent, Counterparty Risk Agent, Collection Intelligence Agent, External Risk Agent, Policy Intelligence Agent, Evidence Intelligence Agent	Credit limit authority, collateral requirement, concentration risk threshold, payment behavior policy, escalation policy	Credit losses avoided, exposure quality improved, DSO intervention prioritized.
Intelligent Journal Entry Studio	SAP ACDOCA/BKPF/BSEG, BlackLine journal tasks, approver matrix, prior journal outcomes, evidence vault	Journal Risk Agent, Policy Agent, Evidence Agent, Control Assurance Agent, Decision Memory Agent, Close Readiness Agent	Journal materiality threshold, SOD, approver authority, close certification, evidence retention	Close acceleration, reduced rework, stronger journal control evidence, fewer post-close corrections.
Autonomous Spend Intelligence	SAP actuals, Ariba commitments, pending invoices, PO/contract data, forecast, treasury windows, supplier master	Spend Intelligence Agent, Cash Intelligence Agent, Forecast Agent, Procurement Context Agent, Policy Agent, Evidence Agent, Impact Agent	Budget tolerance, commitment authority, cash timing, supplier contract policy, working-capital guardrail	Working capital improvement, spend reduction, forecast accuracy, budget-risk prevention.
Enterprise Finance Impact Cockpit	Decision ledger, outcome records, evidence packs, value events, control registry, learning store, enterprise ontology	Impact Agent, Value Attribution Agent, Learning Agent, Control Intelligence Agent, Evidence Agent, Replay Agent	Approved value formula, confidence classification, controller attestation, auditability rule, replay requirement	Traceable, auditable value realization from analyst to team to tower to FB&T to CFO.

Shared runtime capabilities exercised by every mission

Capability	How it participates in the finance decision
Ontology	Every mission resolves domain, process, activity, decision, risk, control, evidence, outcome, and owner.
Decisions	Each mission produces named decision objects with authority, alternatives, expected outcome, and human boundary.
Controls	Policies and controls are evaluated before recommendation or action; exceptions remain visible.
Evidence	Source records, documents, data products, and lineage are linked to every material recommendation.
Learning	Outcomes and overrides create governed learning candidates, not uncontrolled model changes.
Impact	Decisions produce impact events with measured, calculated, or estimated confidence.
Autonomy	Every action records human-only, agent-recommended, agent-assisted, or governed execution state.
CFO rollups	Impact rolls from analyst to supervisor to tower to FB&T to CFO with drilldown to evidence and replay.

Future-state CFO challenge

Memorable demo question: If we want to close in 3 days instead of 5, what must happen? The runtime should reason across journals, reconciliations, intercompany, treasury, controls, resources, policy, evidence, and owner capacity, then return probability, top risks, recommended actions, and expected impact.

Answer component	Expected runtime-backed answer
Close forecast	3-day close probability: 82%
Top risks	Intercompany mismatch, manual journal evidence, delayed BlackLine reconciliations
Recommended actions	Resolve IC mirror entry, clear high-risk journals, attach reconciliation evidence, rebalance controller workload, escalate controls at risk
Proof	Each action links to source data, policy, evidence pack, owner, SLA, replay, learning candidate, and CFO impact

Mission hierarchy

Layer	Missions	Purpose
Foundation	Continuous Close · Autonomous Controls	Trust the numbers and prevent control failure
Operational	Dynamic Credit · Working Capital	Manage risk and optimize cash
Planning	Continuous Forecast · Treasury	Predict what happens next and manage financial capacity
Strategic	Capital Allocation · Portfolio · Performance	Invest, optimize, and intervene across the enterprise
Executive	Enterprise Value	Maximize long-term value across capital, cash, earnings, risk, carbon, and assets

Finance operating model audit

Every mission must work vertically through the organization. If any layer is weak, BP will see a dashboard instead of a finance operating system.

Layer	Pass criterion
Agent Runtime	Does the work: detects, resolves context, gathers evidence, evaluates policies, forms decisions, drafts actions, stores replay, learns from outcomes.
Analyst	Can act: understands what happened, why it happened, and what to do next.
Supervisor	Can manage: sees workload, SLA risk, escalations, and agent/help needs.
Function Leader / Controller	Can optimize: sees root causes, controls, process performance, and business-unit risk.
FB&T Leader	Can orchestrate: sees mission health, value, risk, teams, trends, and intervention points.
CFO	Can steer enterprise value: sees strategic risk, enterprise value, scenarios, tradeoffs, and top actions.

FB&T Mission Orchestration Center audit

Capability	Required proof
Mission health	Continuous Close, Forecast, Credit, Working Capital, Controls, Treasury, Portfolio, Performance, Enterprise Value
Intervention	Missions requiring attention, underperforming teams, unresolved risks, value blockers
Drill-down	Mission → team → decision → evidence → policy → transaction → replay
Value	Value created, cash unlocked, risk reduced, controls strengthened, cycle time improved

CFO Enterprise Value Cockpit audit

Capability	Required proof
Enterprise value	What threatens or improves enterprise value
Strategic simulations	Oil, gas, rates, refinery outage, LNG demand, carbon tax, acquisition, divestment
Top actions	Expected value, risk reduction, cash impact, earnings impact, and evidence
Board explainability	Why this action, what evidence, what alternatives, what tradeoffs, what risk

Universal BP Executive Credibility Audit

Every Future Finance mission must demonstrate these proof dimensions. If a mission cannot show them, it is not yet a credible enterprise decision runtime.

Dimension	What must be visible
Signal	Business event or market movement is visible and linked to a mission decision.
Context	Entity, business unit, owner, system object, policy domain, and history are resolved.
Evidence	Source records, documents, approvals, and source-system snapshots are inspectable.
Policy	Authority, control, risk, accounting, tax, treasury, or investment policies are referenced.
Decision	The mission produces a named decision object, not just a metric or chart.
Recommendation	Recommended response and alternatives are shown with impact.
Human Override	Human approval, rejection, override, or escalation path is explicit.
Action	Action payload is traceable to workflow, system update, notification, hold, hedge, or approval.
Replay	Event, context, evidence, policy, recommendation, human judgment, action, and outcome can be replayed.
Learning	Accepted/rejected outcomes are captured as future pattern memory.
Business Impact	Financial, risk, cash, control, or value impact is quantified.
Enterprise Integration	Systems of record, data products, APIs, and future UDP context layer are named.
Source Drill-down	Users can drill from decision to source transaction, document, or curated data product.
Confidence	The decision carries confidence and uncertainty drivers.
Alternatives	At least two alternative actions or scenarios are available.
Audit Trail	Every material step is retained with timestamp, actor/agent, evidence, and policy result.
Executive Narrative	The mission can produce an executive answer: what happened, why, what to do, and expected outcome.
Mission KPI	A mission-level KPI changes when the scenario changes.
Readiness	Readiness or confidence updates when blockers, policies, or evidence status change.
Enterprise Impact	The mission shows enterprise-level consequence beyond one transaction.

Planning Intelligence

Continuous Forecast Intelligence

What is most likely to happen next?

Open mission

Operational Risk Intelligence

Dynamic Credit Intelligence

Should bp increase, maintain, reduce, or block exposure?

Open mission

Foundation Control Intelligence

Autonomous Control Intelligence

Should this transaction be allowed to proceed?

Open mission

Enterprise Cash Optimization

Enterprise Working Capital Intelligence

How should bp optimize cash across payables, receivables, inventory, commitments, and liquidity?

Open mission

Strategic Intelligence

Capital Allocation Intelligence

Where should the next dollar be invested?

Open mission

Strategic Portfolio Intelligence

Enterprise Portfolio Intelligence

Which assets should bp grow, optimize, divest, or transform?

Open mission

Executive Performance Intelligence

Enterprise Performance Intelligence

What actions should leadership take right now?

Open mission

Planning Intelligence

Treasury Intelligence

How should bp optimize liquidity, FX, debt, covenants, and hedging?

Open mission

Strategic Intelligence

Enterprise Tax Intelligence

How should bp manage tax risk, compliance, and strategic tax value continuously?

Open mission

Strategic Transition Intelligence

Carbon & Sustainability Intelligence

How do carbon exposure, emissions, and transition risk change enterprise decisions?

Open mission

Executive Layer

Enterprise Value Intelligence

Which decisions maximize long-term enterprise value?

Open mission