00 / Discovery before allocation
Where could overlooked value come from?
Join operating and financial signals at the same asset, period, and activity level. Generate a hypothesis, test its cause, then let specialists validate the case. This offline prototype uses explicit formulas; it does not ingest company data or run AI discovery.
Cross-asset mismatch
Compare energy per occupied room or unit produced after adjusting for weather, occupancy, service level, and asset age. A cost gap is a lead, not proof of recoverable savings.
Contract × behaviour
Join purchase volumes and supplier terms, or channel fees and customer switching behaviour. Local departmental reports may not show the group-wide economics.
These methods are also available to accountants and specialists. Any advantage must be demonstrated against their actual process, on previously unseen cases, after investigation and implementation costs.
Adds or updates X1. Always UNVERIFIED. No approval is inferred.
Proposed deployment
—
—
Assumption-adjusted NPV
—
After capital cost · illustrative
Year 1 after-tax cash flow
—
Modeled incremental operating cash flow*
Downside NPV (extra 15%)
—
Sensitivity · not a guarantee
01 / Owner-level decision
Which opportunities deserve the next ringgit?
The allocation is recomputed with capital limits, source-evidence restrictions, and mutually exclusive benefit pools.
—
Same gates and budget; not a specialist-team benchmark
—
—
Portfolio NPV after constraints
—
—
02 / Opportunity register
Every decision has a reason
Select Proof file to inspect source placeholders, assumptions, NPV calculations and reasons for approval or rejection.
| Opportunity · sector | Capex | Annual EBITDA benefit | NPV | ROIC proxy | Evidence status | Decision | Trace |
|---|
03 / Investor-grade safeguards
What the finance team should be able to challenge
Trace the numbers
Assumptions → supporting source → forecast → NPV → allocation. Inputs and approval must be linked to source documents in a live deployment.
Stop benefit double counts
Alternative hotel yield programmes and alternative retail leasing plans cannot both claim the same uplift. The overlap constraints are explicitly defined in this demonstration.
Reject attractive but unsupported claims
A high return does not override an evidence gate. A proposal can be economically attractive and still remain unfunded.
Make decisions falsifiable
Pre-register the baseline and cash-flow targets, assign an accountable owner, reconcile realized uplift to finance records, and revoke benefits when evidence fails.
*Simplified cash flow: forecast EBITDA benefit is multiplied by a subjective realization haircut (not a probability or expected-value estimate) and downside factor. Straight-line depreciation is used solely for illustrative cash-tax and return calculations; tax is 24% and ignores group structures, deferred tax, financing, maintenance capital, working capital, and real-world operating complexities. ROIC proxy = illustrative incremental NOPAT / initial project capital, not audited company ROIC. No asset valuations are inferred.