Runs in your browser
ROI calculator
Model the governance ROI hypotheses from the commercial kit with your own inputs.
Loading the tool…
Hosted port of content/commercial/roi_calculator.py from the VERITRAIL Python platform.
Runs in your browser
Model the governance ROI hypotheses from the commercial kit with your own inputs.
Loading the tool…
Hosted port of content/commercial/roi_calculator.py from the VERITRAIL Python platform.
Illustrative estimate. VERITRAIL has no measured customer results yet: values labelled VERITRAIL hypothesis — validate in pilot are planning hypotheses, not measurements. Results are not guaranteed savings and are not financial advice. Your inputs stay in this browser.
Pre-filled with the synthetic worked example (fictional “Example Co”; not a customer).
| Measure | Value |
|---|---|
| Gross modelled annual benefit | 160,980 |
| Year-1 realized benefit | 120,735 |
| Year-1 cost | 105,000 |
| Year-1 net benefit | 15,735 |
| Year-1 ROI | 15.0% |
| Steady-state annual net benefit | 85,980 |
| Steady-state ROI | 114.6% |
| Payback | 10.4 months |
| Governance hours freed (modelled) | 684.0 h |
| Efficiency savings share of gross benefit | 40.4% |
Every VERITRAIL effect hypothesis is scaled by 0.5 / 1.0 / 1.5. Present ranges, never a single headline number.
| Scenario | Gross benefit | Year-1 net | Year-1 ROI | Payback |
|---|---|---|---|---|
| Conservative (hypotheses × 0.5) | 80,490 | -44,632 | -42.5% | 20.9 months |
| Base (hypotheses × 1.0) | 160,980 | 15,735 | 15.0% | 10.4 months |
| Optimistic (hypotheses × 1.5, capped at 0.9) | 241,470 | 76,102 | 72.5% | 7.0 months |
| Line | Item | Formula | Value |
|---|---|---|---|
| L1 | Assessment effort today (hours/yr) | ai_systems × assessment_hours_per_system × assessments_per_system_per_year | 960.0 h |
| L2 | Assessment cost today | L1 × hourly_cost | 91,200 |
| L3 | Assessment hours saved | L1 × assessment_effort_reduction | 384.0 h |
| L4 | Assessment cost saving | L2 × assessment_effort_reduction | 36,480 |
| L5 | Audit-preparation cost today | audit_prep_hours_per_year × hourly_cost | 57,000 |
| L6 | Audit-preparation hours saved | audit_prep_hours_per_year × audit_prep_reduction | 300.0 h |
| L7 | Audit-preparation cost saving | L5 × audit_prep_reduction | 28,500 |
| L8 | Approval days saved per use case | approval_cycle_days × approval_cycle_reduction | 13.5 d |
| L9 | Time-to-value benefit | new_use_cases_per_year × L8 × delay_cost_per_day | 81,000 |
| L10 | Expected annual incident loss today | annual_incident_probability × expected_incident_cost | 75,000 |
| L11 | Risk-reduction value (expected) | L10 × incident_likelihood_reduction | 15,000 |
| L12 | Gross modelled annual benefit | L4 + L7 + L9 + L11 | 160,980 |
| L13 | Year-1 realized benefit | L12 × realization_rate | 120,735 |
| L14 | Governance hours freed per year (modelled) | L3 + L6 | 684.0 h |
| L15 | Year-1 cost | veritrail_annual_cost + implementation_cost | 105,000 |
| L16 | Year-1 net benefit | L13 − L15 | 15,735 |
| L17 | Steady-state annual net benefit | L12 − veritrail_annual_cost | 85,980 |
| Assumption | Value | Source | Note |
|---|---|---|---|
ai_systems | 40 | Customer input | Number of AI systems in scope (inventory count or best estimate). |
assessment_hours_per_system | 24 | Customer input | Person-hours to complete one risk/impact assessment today. |
hourly_cost | 95 | Customer input | Blended fully loaded hourly cost of people doing governance work. |
audit_prep_hours_per_year | 600 | Customer input | Person-hours per year assembling AI evidence for audits, customers and regulators. |
approval_cycle_days | 45 | Customer input | Average calendar days from use-case intake to approval decision today. |
assessments_per_system_per_year | 1 | Customer input | Assessments or reassessments per system per year (initial, change-triggered, periodic). |
new_use_cases_per_year | 12 | Customer input | New AI use cases going through approval per year. |
delay_cost_per_day | 500 | Customer estimate | Business value lost per use case per day of approval delay. Default 0 excludes this benefit. |
annual_incident_probability | 0.1 | Customer estimate | Probability of at least one material AI incident in a year. |
expected_incident_cost | 750000 | Customer estimate | Expected cost of a material AI incident (response, remediation, legal, customer impact). |
veritrail_annual_cost | 75000 | Customer input | Annual VERITRAIL subscription from the commercial proposal (pricing is itself a hypothesis). |
implementation_cost | 30000 | Customer input | One-time cost, e.g. Governance Baseline fee and internal implementation effort. |
assessment_effort_reduction | 0.4 | VERITRAIL hypothesis — validate in pilot | Share of assessment effort removed by agent-assisted, pre-populated assessments. |
audit_prep_reduction | 0.5 | VERITRAIL hypothesis — validate in pilot | Share of audit-preparation effort removed by continuous evidence capture and evidence packs. |
approval_cycle_reduction | 0.3 | VERITRAIL hypothesis — validate in pilot | Share of approval cycle time removed by triage, routing and complete submissions. |
incident_likelihood_reduction | 0.2 | VERITRAIL hypothesis — validate in pilot | Relative reduction in material-incident likelihood from runtime guardrails, testing and oversight. |
realization_rate | 0.75 | VERITRAIL hypothesis — validate in pilot | Share of modelled benefit realized in year 1, reflecting adoption ramp. |
Hypotheses to validate in pilots: assessment_effort_reduction, audit_prep_reduction, approval_cycle_reduction, incident_likelihood_reduction, realization_rate.