Every prompt is scored for intent, efficiency, and model fit, then attributed to the org chart. One number tells you whether the spend is working; the rows below tell you where it is not.
Loading organization…
AI literacy
–
Loading score…
Loading rubric and arithmetic…
Example recommendation
Route short, low-context requests to the standard model.
Use a bounded gateway rule while leaving high-value prompts unchanged.
Monthly baseline
$7,430
Projected savings
$5,200 / month
Accountable role
Core Services platform director
Confidence
High · 760-query scored sample
Bundled synthetic sample dataThis is an illustrative recommendation—not live analysis, customer data, or realized savings.
Select one or more manifest-compatible provider exports and one HRIS mapping. Two contiguous, equal-length provider periods unlock a like-for-like trend; a single period keeps the current decision brief unchanged.
◇Example data
Your files do not leave this tab.No upload · no credentials · no network transfer · no browser storage · results disappear on refresh. Column headers and totals are shown; source rows and cell values are never rendered, announced, or exported.
Choose files together or add them in batches. Provider periods must use the same source; all processing and state stay in this tab.
Ready for local files.The bundled example analysis remains visible until both compatible exports are ready.
Imported result
Local FinOps decision brief
The question this answers
—
—
—
Prioritized next action
—
Recommended department action
Waiting for a supported pair…
◆Confidence unavailable
Quantified impact
—
Example dataBundled synthetic sample — not your import and not realized savings.
Department
—
Benchmark
Unavailable
Data provenance and redaction
Waiting for an imported result.
Only column headers, declared contract fields, and aggregate totals are read into this page. No cell value, record identifier, or file name is rendered, announced, or written to an export.
Material trend
—History not yet analyzed
Benchmark
◇Unavailable
Progressive disclosure
Department evidence
Expand a department finding to inspect impact, trend, confidence, and provenance in that order.
Period-level detailMapping assumptions
Data-quality warnings
Benchmark and trend limits
Recommendation evidence
Loading bundled analysis…The local import tools remain available while the synthetic executive view loads.
◇ Example data — bundled synthetic sample, replaced by your figures only after a local import.
AI spend · period
–
Across all providers
Recoverable spend
–
Down-routing, training, and leakage
High-value share
–
Of scored spend
Peer position
–
Anonymized industry cohort
Finance leader action portfolio
Turn projected savings into verified results.
One prioritized action per department, with lifecycle and evidence kept explicit.
Savings targets and realized results will appear when the action lifecycle is ready.
Defensible evaluation · static fixture
Inspect how a recommendation earns its score.
Every point maps to a labelled criterion, evidence statement, and documented weight assumption. Unsafe content cannot be averaged into approval.
Bundled synthetic fixture · deterministic rubric · no stored prompts, credentials, customer data, or live provider calls.
Loading inspectable fixture scores…
Decision 1 of 3 · intervention
Which department needs help?
Lowest eligible performance score first. Unavailable samples are not scored or ranked as poor performance.
Loading sample provenance…
Loading departments…
Selected department
Loading…
Loading sampling status…
–
Priority 01 · recommended intervention
Loading static action…
Loading fixture
Reading the bundled demo result. No live analysis is running.
Expected impact
Loading…
Confidence
Loading…
Accountable role
Loading…
Provenance
Bundled static fixture
Baseline—
Target—
Estimated savings—
Simulated realized—
Why this action:Loading diagnosis…
Decision 2 of 3 · trajectory
Is cost/performance worsening?
Loading comparison…
Decision 3 of 3 · comparator & evidence
How does it compare with the benchmark?
Loading comparator…
Inspect supporting scored evidence
Quality engine
Where the money goes
Spend split by what the prompt was actually asking for, not by token count alone.
Loading spend mix…
Privacy & compliance
Scored on shape, never on secrets
Prompts pass through redaction before any judge model or score record sees them. Structure survives; identity and credentials do not.
Sample data only This tab renders hand-authored sample data. No production gateway, HRIS connection, customer prompt, or telemetry store is read by this page.
Gateway unavailable until the bundled sample starts.