# Formula library

> Every value formula Roiva ships: what it calculates, its arithmetic, and the reading, baseline or rate each input comes from.
>
> Source: https://roiva-staging.com/docs/reference/formulas
> Generated from what Roiva ships, on every deploy.

The formulas an initiative template brings with it. Adopting a template clones these onto the initiative, where the account can change them; a clone nobody has touched follows the library when the library changes.

Formulas that suit any industry come first, grouped by the function they serve. The rest are written for one industry and grouped by it.

What a formula reads, when it runs and what its result has to clear before it counts is in [How Roiva calculates value](https://roiva-staging.com/docs/how-value-is-calculated).

## Engineering
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| AI Cost per Million Output Tokens Reduction; Measures the spend avoided when each million output tokens costs less, all in, than it did before optimization — for a provider that reports tokens but no request count, such as Anthropic. Output tokens in the period × the drop in spend per million output tokens, with the current figure read straight from the provider's usage and spend. | ai_output_tokens / 1000000 * (baseline_ai_cost_per_million_output_tokens_usd - current_ai_cost_per_million_output_tokens_usd) * 100 | ai_spend_usd — reading of ai.cost.spend; ai_output_tokens — reading of ai.usage.output_tokens; current_ai_cost_per_million_output_tokens_usd; baseline_ai_cost_per_million_output_tokens_usd — baseline of ai.cost.per_million_output_tokens |
| AI Cost per Request Reduction; Measures the spend avoided when each AI request costs less than it did before optimization. Requests in the period × the drop in cost per request, with the current cost per request read straight from the provider's usage and spend. | ai_requests * (baseline_ai_cost_per_request_usd - current_ai_cost_per_request_usd) * 100 | ai_requests — reading of ai.usage.request_count; ai_spend_usd — reading of ai.cost.spend; current_ai_cost_per_request_usd; baseline_ai_cost_per_request_usd — baseline of ai.cost.per_request |
| Asset Audit & Reconciliation Labor Savings; Monetizes IT team time recovered by replacing manual software inventory audits and license reconciliation with automated discovery. (Hours per audit before − after) × the audits in the period (a year's audits spread evenly over its months) × labor rate. | (baseline_audit_hours_per_cycle - current_audit_hours_per_cycle) * audit_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; audit_cycles_per_year — assumption; labor_rate_cents_per_hour — assumption; current_audit_hours_per_cycle — reading of it.audits.hours_per_cycle; baseline_audit_hours_per_cycle — baseline of it.audits.hours_per_cycle |
| Breach Exposure Cost Avoidance; Estimates the expected breach cost avoided by reducing mean time to detect. Breaches that are detected faster have materially lower total costs — earlier containment limits data exfiltration, regulatory exposure, and remediation scope. Incidents in the period × hours of exposure saved per incident × cost per hour of exposure. | incidents_in_period * (baseline_mttd_hours - current_mttd_hours) * avg_breach_cost_per_hour_of_exposure_cents | current_mttd_hours — reading of it.security.mean_time_to_detect_hours; baseline_mttd_hours — baseline of it.security.mean_time_to_detect_hours; incidents_in_period — reading of it.security.incident_count; avg_breach_cost_per_hour_of_exposure_cents — assumption |
| Deployment Engineering Time Savings; Monetizes the engineering time recovered by automating deployment pipelines, testing, and rollout processes. Deployments × hours saved per deployment × engineering labor rate. | deployments_per_period * (baseline_hours_per_deployment - current_hours_per_deployment) * labor_rate_cents_per_hour | deployments_per_period — reading of it.deployments.count; labor_rate_cents_per_hour — assumption; current_hours_per_deployment — reading of it.deployments.hours_per_deployment; baseline_hours_per_deployment — baseline of it.deployments.hours_per_deployment |
| Developer Time Saved by an AI Coding Assistant; Values the developer time an AI coding assistant saves: suggestions developers accepted in the period × minutes saved per acceptance × the fully loaded developer hourly rate. Acceptances come from whichever assistant reports them; the minutes and the rate are Assumptions an account can override. | acceptances * minutes_per_acceptance * developer_hourly_rate / 60 | acceptances — reading of engineering.coding_assistant.acceptances; developer_hourly_rate — assumption; minutes_per_acceptance — assumption |
| Developer Time Saved by an AI Coding Assistant (by lines of code); Values the developer time an AI coding assistant saves where the assistant reports no acceptance count: lines of code it wrote that developers kept in the period × minutes saved per line × the fully loaded developer hourly rate. For an assistant that does count acceptances, the acceptances formula is the better measure and an initiative gets that one instead. | lines_accepted * minutes_per_line * developer_hourly_rate / 60 | lines_accepted — reading of engineering.coding_assistant.lines_accepted; minutes_per_line — assumption; developer_hourly_rate — assumption |
| Developer Time Saved by GitHub Copilot and Claude Code; Values the developer time two coding assistants save together, each in its own unit: Copilot acceptances × minutes saved per Copilot acceptance, plus Claude Code accepted edits × minutes saved per Claude Code edit, at the fully loaded developer hourly rate. The two are priced separately because a Copilot acceptance can bundle several completions while a Claude Code one is a single edit. | (copilot_acceptances * minutes_per_copilot_acceptance + claude_code_acceptances * minutes_per_claude_code_acceptance) * developer_hourly_rate / 60 | copilot_acceptances — reading of github.copilot.suggestions_accepted; developer_hourly_rate — assumption; claude_code_acceptances — reading of anthropic.claude_code.acceptances; minutes_per_copilot_acceptance — assumption; minutes_per_claude_code_acceptance — assumption |
| Developer Time Saved by GitHub Copilot and Claude Code on Claude Enterprise; Values the developer time two coding assistants save together, each in its own unit: Copilot acceptances × minutes saved per Copilot acceptance, plus edits accepted in Claude Code across the Claude Enterprise organization × minutes saved per Claude Code edit, at the fully loaded developer hourly rate. The two are priced separately because a Copilot acceptance can bundle several completions while a Claude Code one is a single edit. | (copilot_acceptances * minutes_per_copilot_acceptance + claude_code_acceptances * minutes_per_claude_code_acceptance) * developer_hourly_rate / 60 | copilot_acceptances — reading of github.copilot.suggestions_accepted; developer_hourly_rate — assumption; claude_code_acceptances — reading of claude_enterprise.claude_code.acceptances; minutes_per_copilot_acceptance — assumption; minutes_per_claude_code_acceptance — assumption |
| Direct Cloud Spend Reduction; Measures the direct reduction in cloud infrastructure spend from rightsizing, reserved capacity, and waste elimination. Baseline spend minus current spend, compounded over the measurement period. | (baseline_monthly_cloud_spend_usd - current_monthly_cloud_spend_usd) * 100 * months_in_period | months_in_period; current_monthly_cloud_spend_usd — reading of cloud.cost.total_spend; baseline_monthly_cloud_spend_usd — baseline of cloud.cost.total_spend |
| Failed Deployment Incident Cost Reduction; Estimates the cost avoided when a lower change failure rate means fewer production incidents. Deployments × reduction in failure rate × average cost to investigate and remediate one production incident. | deployments_per_period * (baseline_change_failure_rate_pct - current_change_failure_rate_pct) / 100 * avg_incident_cost_cents | deployments_per_period — reading of it.deployments.count; avg_incident_cost_cents — assumption; current_change_failure_rate_pct — reading of it.deployments.change_failure_rate; baseline_change_failure_rate_pct — baseline of it.deployments.change_failure_rate |
| Incident Response Labor Savings; Monetizes the security analyst time recovered by automated detection and response playbooks. Faster containment means fewer analyst hours per incident. Incidents × hours saved per incident × blended SOC rate. | incidents_per_period * (baseline_mttc_hours - current_mttc_hours) * labor_rate_cents_per_hour | current_mttc_hours — reading of it.security.mean_time_to_contain_hours; baseline_mttc_hours — baseline of it.security.mean_time_to_contain_hours; incidents_per_period — reading of it.security.incident_count; labor_rate_cents_per_hour — assumption |
| MTTR Reduction End-User Productivity Value; Estimates the productivity value recovered by resolving IT issues faster. Every hour of resolution time cut per incident frees end-users to do productive work. Incidents × hours reduced × users affected × hourly productivity value. Both resolution times are the help desk's average days from ticket creation to resolution, so the before-value is the account's own months before the initiative. | incidents_per_period * (baseline_resolution_days - current_resolution_days) * 24 * avg_affected_users_per_incident * avg_hourly_productivity_value_cents | incidents_per_period — reading of support.tickets.volume; current_resolution_days — reading of support.tickets.avg_resolution_days; baseline_resolution_days — baseline of support.tickets.avg_resolution_days; avg_affected_users_per_incident — reading of it.helpdesk.avg_users_per_incident; avg_hourly_productivity_value_cents — assumption |
| Software License Waste Eliminated; Monetizes the direct savings from eliminating unused, duplicate, or over-licensed software identified through automated asset discovery. Annual software spend × reduction in unused license percentage × the period's share of the year. | annual_software_spend_usd * (baseline_unused_license_pct - current_unused_license_pct) / 100 * months_in_period / 12 * 100 | months_in_period; annual_software_spend_usd — baseline of it.software.annual_spend; current_unused_license_pct — reading of it.software.unused_license_share; baseline_unused_license_pct — baseline of it.software.unused_license_share |

## Finance
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Adjustment & Restatement Rework Cost Reduction; Monetizes the reduction in revenue recognition errors requiring correction. Fewer adjustments in the period × average fully-loaded cost to identify, correct, and document one recognition adjustment. | (baseline_adjustment_count - current_adjustment_count) * avg_adjustment_cost_cents | current_adjustment_count — reading of finance.revenue_recognition.adjustment_count; avg_adjustment_cost_cents — assumption; baseline_adjustment_count — baseline of finance.revenue_recognition.adjustment_count |
| Cash Collection Acceleration Value; Estimates the working capital value of resolving disputes faster. Disputed invoices represent delayed cash. Reduction in dispute resolution days × volume of disputed invoices × average invoice value × daily cost of capital. | invoices_sent * (baseline_error_rate_pct / 100) * avg_invoice_value_cents * (baseline_dispute_days - current_dispute_days) * daily_cost_of_capital_decimal | invoices_sent — reading of finance.invoices.count; invoices_volume_usd — reading of finance.invoices.volume; current_dispute_days — reading of finance.disputes.avg_resolution_days; baseline_dispute_days — baseline of finance.disputes.avg_resolution_days; avg_invoice_value_cents; baseline_error_rate_pct — baseline of finance.invoices.error_rate; daily_cost_of_capital_decimal — assumption |
| Close Cycle Compression Value; Estimates the business value of a faster close. Each day of close cycle reduction allows management to act on financial data sooner and reduces the cost of maintaining month-end crunch staffing. Days reduced × the close cycles in the period (a year's cycles spread evenly over its months) × estimated daily cost of delayed reporting. | (baseline_close_days - current_close_days) * close_cycles_per_year * months_in_period / 12 * daily_cost_of_delayed_reporting_cents | months_in_period; current_close_days — reading of finance.close.cycle_days; baseline_close_days — baseline of finance.close.cycle_days; close_cycles_per_year — assumption; daily_cost_of_delayed_reporting_cents — assumption |
| Dispute Resolution Cost Savings; Monetizes the reduction in billing dispute work — fewer errors means fewer credit memos, customer calls, and correction cycles. Invoices sent × reduction in error rate × average cost to resolve one billing dispute. | invoices_sent * (baseline_error_rate_pct - current_error_rate_pct) / 100 * avg_dispute_resolution_cost_cents | invoices_sent — reading of finance.invoices.count; current_error_rate_pct — reading of finance.invoices.error_rate; baseline_error_rate_pct — baseline of finance.invoices.error_rate; avg_dispute_resolution_cost_cents — assumption |
| Forecast Cycle Time Savings; Monetizes the analyst time recovered by automating the forecast production process. | (baseline_cycle_hours - current_cycle_hours) * forecast_runs_per_period * blended_labor_rate_cents_per_hour | current_cycle_hours — reading of operations.forecasting.hours_per_cycle; baseline_cycle_hours — baseline of operations.forecasting.hours_per_cycle; forecast_runs_per_period — reading of operations.forecasting.cycles; blended_labor_rate_cents_per_hour — assumption |
| Inventory Cost Avoidance; Estimates the carrying cost and write-off reduction from holding less excess inventory due to improved forecast accuracy. Excess inventory reduced × the period's share of the annual holding cost rate. | (baseline_excess_inventory_usd - current_excess_inventory_usd) * holding_cost_rate_decimal * months_in_period / 12 * 100 | months_in_period; holding_cost_rate_decimal — assumption; current_excess_inventory_usd — reading of supply_chain.inventory.excess_value; baseline_excess_inventory_usd — baseline of supply_chain.inventory.excess_value |
| Invoice Error & Rework Cost Reduction; Monetizes the reduction in manual entry errors, duplicate payments, and reconciliation rework on the vendor invoices AP processes. Vendor bills × reduction in the bill error rate × average cost to identify, correct, and reconcile one error. The error rate is AP's, on the bills it enters — not the billing error rate on invoices the company sends. | invoices_processed * (baseline_error_rate_pct - current_error_rate_pct) / 100 * avg_error_resolution_cost_cents | invoices_processed — reading of finance.bills.count; current_error_rate_pct — reading of finance.bills.error_rate; baseline_error_rate_pct — baseline of finance.bills.error_rate; avg_error_resolution_cost_cents — assumption |
| Invoice Processing Labor Savings; Monetizes the AP team time recovered by automating invoice extraction, validation, and 3-way PO matching. Invoice volume × minutes saved per invoice × blended AP labor rate. | invoices_processed * (baseline_minutes_per_invoice - current_minutes_per_invoice) * labor_rate_cents_per_minute | invoices_processed — reading of finance.bills.count; labor_rate_cents_per_hour — assumption; current_minutes_per_invoice — reading of finance.ap.minutes_per_invoice; labor_rate_cents_per_minute; baseline_minutes_per_invoice — baseline of finance.ap.minutes_per_invoice |
| Reconciliation Labor Savings; Monetizes the accounting staff time recovered by automating account reconciliations and journal entries. Hours saved per close cycle × the close cycles in the period (a year's cycles spread evenly over its months) × blended accounting rate. | (baseline_reconciliation_hours - current_reconciliation_hours) * close_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; close_cycles_per_year — assumption; labor_rate_cents_per_hour — assumption; current_reconciliation_hours — reading of finance.reconciliation.hours; baseline_reconciliation_hours — baseline of finance.reconciliation.hours |
| Revenue Recognition Labor Savings; Monetizes accounting team time recovered by replacing manual ASC 606 schedule building with automated recognition logic. Hours saved in the period × accounting labor rate. | (baseline_rev_rec_hours - current_rev_rec_hours) * labor_rate_cents_per_hour | current_rev_rec_hours — reading of finance.revenue_recognition.manual_hours; baseline_rev_rec_hours — baseline of finance.revenue_recognition.manual_hours; labor_rate_cents_per_hour — assumption |
| Treasury Forecast Assembly Labor Savings; Monetizes the treasury team time recovered by replacing manual spreadsheet forecasting with AI-driven models. Hours saved per forecast cycle × cycles per period × treasury labor rate. | (baseline_treasury_hours - current_treasury_hours) * forecast_cycles_per_period * labor_rate_cents_per_hour | current_treasury_hours — reading of finance.treasury.manual_hours_per_cycle; baseline_treasury_hours — baseline of finance.treasury.manual_hours_per_cycle; labor_rate_cents_per_hour — assumption; forecast_cycles_per_period — reading of operations.forecasting.cycles |
| Working Capital Optimization Value; Estimates the cost-of-capital benefit from better forecast accuracy. More accurate forecasts allow treasury to reduce idle cash, optimize short-term borrowing, and earn more on deployable liquidity. Forecast accuracy improvement × average cash balance × the period's share of the annual cost of capital. | (current_forecast_accuracy_pct - baseline_forecast_accuracy_pct) / 100 * avg_cash_balance_usd * annual_cost_of_capital_decimal * months_in_period / 12 * 100 | months_in_period; avg_cash_balance_usd — reading of finance.treasury.avg_cash_balance; current_forecast_accuracy_pct — reading of operations.forecasting.accuracy; annual_cost_of_capital_decimal — assumption; baseline_forecast_accuracy_pct — baseline of operations.forecasting.accuracy |

## Sales
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| At-Risk Deal Recovery Value; Estimates revenue recovered by identifying and saving at-risk deals earlier. Reduction in the at-risk share × the pipeline that moves through in the period × expected close rate on a saved deal. The open pipeline is a snapshot that turns over once a sales cycle, so the part of it the period accounts for is its value × days in the period ÷ average sales cycle days; the whole of it would be counted again every month a deal stayed open. | (baseline_at_risk_deal_pct - current_at_risk_deal_pct) / 100 * total_pipeline_value_usd * days_in_period / avg_sales_cycle_days * close_rate_decimal * 100 | days_in_period; close_rate_decimal — assumption; avg_sales_cycle_days — reading of crm.deals.avg_sales_cycle_days; current_at_risk_deal_pct — reading of crm.deals.at_risk_share; total_pipeline_value_usd — reading of crm.deals.pipeline_value; baseline_at_risk_deal_pct — baseline of crm.deals.at_risk_share |
| Contract Processing Labor Savings; Monetizes the legal, sales, and ops time saved by automating contract drafting, redlining, and approval routing. Contracts processed × hours saved × blended labor rate. | contracts_processed * (baseline_hours_per_contract - current_hours_per_contract) * labor_rate_cents_per_hour | contracts_processed — reading of crm.contracts.processed_count; labor_rate_cents_per_hour — assumption; current_hours_per_contract — reading of crm.contracts.hours_per_contract; baseline_hours_per_contract — baseline of crm.contracts.hours_per_contract |
| Deal Cycle Acceleration Value; Estimates the revenue brought forward by sending proposals sooner. Each deal won in the period closed that many days earlier, and a day of a deal is its value ÷ 365 — the revenue a contract starts earning a day sooner. Deals won × days saved between demo and proposal × average won deal value ÷ 365. A proposal that did not win brought nothing forward, so proposals sent are not the count. | deals_won * (baseline_days_to_proposal - current_days_to_proposal) * avg_daily_deal_value_cents | deals_won — reading of crm.deals.won_count; crm_deals_won_value_usd — reading of crm.deals.won_value; current_days_to_proposal — reading of crm.deals.avg_demo_to_proposal_days; baseline_days_to_proposal — baseline of crm.deals.avg_demo_to_proposal_days; avg_daily_deal_value_cents |
| Deal Cycle Compression Value; Estimates the revenue value of compressing the sales cycle. Shorter cycles unlock pipeline capacity earlier: deals closed × days reduced × daily average deal value. | deals_closed * (baseline_cycle_days - current_cycle_days) * avg_daily_deal_value_cents | deals_closed — reading of crm.deals.won_count; current_cycle_days — reading of crm.deals.avg_sales_cycle_days; baseline_cycle_days — baseline of crm.deals.avg_sales_cycle_days; crm_deals_won_value_usd — reading of crm.deals.won_value; avg_daily_deal_value_cents |
| Deal Slippage Prevention Value; Estimates the revenue brought forward by removing legal bottlenecks that hold up signature. Each deal won in the period signed that many days earlier, and a day of a deal is its value ÷ 365 — the revenue a contract starts earning a day sooner. Deals won × contract cycle days saved × average won deal value ÷ 365. | deals_per_period * (baseline_contract_cycle_days - current_contract_cycle_days) * avg_daily_deal_value_cents | deals_per_period — reading of crm.deals.won_count; crm_deals_won_value_usd — reading of crm.deals.won_value; avg_daily_deal_value_cents; current_contract_cycle_days — reading of crm.contracts.avg_cycle_days; baseline_contract_cycle_days — baseline of crm.contracts.avg_cycle_days |
| Forecast Accuracy Revenue Value; Estimates the business value of more accurate forecasts. Better accuracy allows resource allocation and incentive planning to be better calibrated — reducing the cost of forecast miss (over-investment in deals that slip, under-resourcing deals that could close). Revenue × improvement in accuracy × capture rate. | (current_forecast_accuracy_pct - baseline_forecast_accuracy_pct) / 100 * quarterly_revenue_cents * forecast_miss_cost_rate_decimal | financials_revenue_usd — reading of finance.pnl.revenue; quarterly_revenue_cents; current_forecast_accuracy_pct — reading of operations.forecasting.accuracy; baseline_forecast_accuracy_pct — baseline of operations.forecasting.accuracy; forecast_miss_cost_rate_decimal — assumption |
| Lead Qualification Time Savings; Monetizes the sales rep time recovered by replacing manual lead review with AI scoring. | leads_scored * (baseline_minutes_per_lead - current_minutes_per_lead) * labor_rate_cents_per_minute | leads_scored — reading of crm.contacts.new_count; current_minutes_per_lead — reading of crm.leads.avg_qualify_minutes; baseline_minutes_per_lead — baseline of crm.leads.avg_qualify_minutes; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute |
| Pipeline Revenue from Higher Conversion; Estimates incremental revenue from converting more new leads into sales-qualified leads. Uplift in the lead-to-SQL rate × new leads × average deal value × close rate. Both rates are SQLs ÷ new leads from the CRM, so the before-value is the account's own months before the initiative. | (current_lead_to_sql_rate_pct - baseline_lead_to_sql_rate_pct) / 100 * leads_scored * avg_deal_value_cents * close_rate_decimal | leads_scored — reading of crm.contacts.new_count; close_rate_decimal; crm_deals_won_count — reading of crm.deals.won_count; avg_deal_value_cents; crm_contacts_sql_count — reading of crm.contacts.sql_count; crm_deals_win_rate_pct — reading of crm.deals.win_rate; crm_deals_won_value_usd — reading of crm.deals.won_value; current_lead_to_sql_rate_pct; baseline_lead_to_sql_rate_pct — baseline of crm.leads.sql_rate |
| Pipeline Revenue Uplift; Estimates incremental pipeline value from routing more qualified leads to reps. Increase in qualified leads × average deal value × close rate. | (qualified_leads - baseline_qualified_leads) * avg_deal_value_cents * close_rate_decimal | qualified_leads — reading of crm.leads.qualified_count; close_rate_decimal; crm_deals_won_count — reading of crm.deals.won_count; avg_deal_value_cents; crm_deals_win_rate_pct — reading of crm.deals.win_rate; crm_deals_won_value_usd — reading of crm.deals.won_value; baseline_qualified_leads — baseline of crm.leads.qualified_count |
| Proposal Assembly Labor Savings; Monetizes the rep and pre-sales time recovered by automating proposal assembly. Proposals sent × hours saved per proposal × labor rate. | proposals_sent * (baseline_hours_per_proposal - current_hours_per_proposal) * labor_rate_cents_per_hour | proposals_sent — reading of crm.proposals.sent_count; labor_rate_cents_per_hour — assumption; current_hours_per_proposal — reading of crm.proposals.hours_per_proposal; baseline_hours_per_proposal — baseline of crm.proposals.hours_per_proposal |
| Sales Rep Screening Time Savings; Monetizes the rep time recovered when AI scoring eliminates manual lead review. Volume of scored leads × time saved per lead × blended rep cost. | leads_scored * (baseline_minutes_per_lead - current_minutes_per_lead) * labor_rate_cents_per_minute | leads_scored — reading of crm.contacts.new_count; current_minutes_per_lead — reading of crm.leads.avg_qualify_minutes; baseline_minutes_per_lead — baseline of crm.leads.avg_qualify_minutes; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute |
| Win Rate Revenue Uplift; Translates the improvement in win rate into the revenue it won in the period: deals decided in the period (won plus lost) × the win-rate gain × average won deal value — the deals won that the old win rate would have lost. The open pipeline is not the base: it is a snapshot of deals that stay open for a whole sales cycle, so it would be counted again every month until they closed. | decided_deals * (current_win_rate_pct - baseline_win_rate_pct) / 100 * avg_deal_value_cents | decided_deals; crm_deals_won_count — reading of crm.deals.won_count; avg_deal_value_cents; crm_deals_lost_count — reading of crm.deals.lost_count; current_win_rate_pct — reading of crm.deals.win_rate; baseline_win_rate_pct — baseline of crm.deals.win_rate; crm_deals_won_value_usd — reading of crm.deals.won_value |

## Legal compliance
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Compliance Monitoring Labor Savings; Monetizes compliance team time recovered by automating regulatory change tracking and control mapping. Hours saved per week × weeks × blended compliance analyst rate. | (baseline_monitoring_hours_per_week - current_monitoring_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_monitoring_hours_per_week — reading of legal.compliance.monitoring_hours_per_week; baseline_monitoring_hours_per_week — baseline of legal.compliance.monitoring_hours_per_week |
| Compliance Penalty Risk Reduction; Estimates the expected regulatory penalty cost avoided by achieving near-100% policy attestation completion — ensuring no regulator can point to employees who were unaware of required policies. Findings avoided a year, spread evenly over its months, × average fine per finding. | non_completion_incidents_prevented_per_year * months_in_period / 12 * avg_regulatory_fine_cents | months_in_period; avg_regulatory_fine_cents — assumption; non_completion_incidents_prevented_per_year — assumption |
| Contract Search Labor Savings; Monetizes legal and business team time recovered when AI search replaces manual hunting through disparate contract storage locations. Hours saved per week × weeks × blended rate of staff searching contracts. | (baseline_search_hours_per_week - current_search_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_search_hours_per_week — reading of legal.contracts.search_hours_per_week; baseline_search_hours_per_week — baseline of legal.contracts.search_hours_per_week |
| Legal Research Labor Savings; Monetizes in-house attorney and paralegal time recovered by AI-accelerated research and document review. Matters closed × hours saved per matter × blended in-house legal rate. | matters_per_period * (baseline_research_hours_per_matter - current_research_hours_per_matter) * labor_rate_cents_per_hour | matters_per_period — reading of legal.matters.closed_count; labor_rate_cents_per_hour — assumption; current_research_hours_per_matter — reading of legal.matters.research_hours_per_matter; baseline_research_hours_per_matter — baseline of legal.matters.research_hours_per_matter |
| Missed Renewal Cost Avoidance; Estimates the cost avoided when automated renewal alerts prevent contracts from auto-renewing on unfavorable terms or lapsing without negotiation. Each missed renewal can lock in below-market pricing or create coverage gaps. Missed renewals avoided a year, spread evenly over its months, × average cost per missed renewal. | (baseline_missed_renewals_per_year - current_missed_renewals_per_year) * months_in_period / 12 * avg_missed_renewal_cost_cents | months_in_period; avg_missed_renewal_cost_cents — assumption; current_missed_renewals_per_year — reading of legal.contracts.missed_renewals_per_year; baseline_missed_renewals_per_year — baseline of legal.contracts.missed_renewals_per_year |
| Outside Counsel Spend Reduction; Estimates direct savings from shifting routine legal research and matter work in-house using AI, reducing reliance on expensive outside counsel billing rates. | (baseline_outside_counsel_monthly_usd - current_outside_counsel_monthly_usd) * months_in_period * 100 | months_in_period; current_outside_counsel_monthly_usd — reading of legal.outside_counsel.monthly_spend; baseline_outside_counsel_monthly_usd — baseline of legal.outside_counsel.monthly_spend |
| Periodic Vendor Review Labor Savings; Monetizes the procurement and legal time recovered by replacing manual periodic reassessments with continuous automated monitoring and risk scoring. Vendors × review hours saved per vendor × the review cycles in the period (a year's cycles spread evenly over its months) × blended rate. | vendors_in_portfolio * (baseline_review_hours_per_vendor - current_review_hours_per_vendor) * review_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; vendors_in_portfolio — reading of legal.vendors.portfolio_count; review_cycles_per_year — assumption; labor_rate_cents_per_hour — assumption; current_review_hours_per_vendor — reading of legal.vendors.review_hours_per_vendor; baseline_review_hours_per_vendor — baseline of legal.vendors.review_hours_per_vendor |
| Regulatory Penalty & Examination Cost Avoidance; Estimates the expected value of penalties avoided by closing compliance gaps before they become findings. Most gaps left open never draw a finding, so each gap closed is worth the chance it would have become one times what a finding costs: gaps prevented × share that would have become a penalized finding × average cost of a finding. | compliance_gaps_prevented_in_period * gap_to_finding_rate_decimal * avg_regulatory_penalty_cents | gap_to_finding_rate_decimal — assumption; avg_regulatory_penalty_cents — assumption; compliance_gaps_prevented_in_period — reading of legal.compliance.gaps_prevented |
| Supplier Review Labor Savings; Monetizes procurement team time recovered by automating the continuous supplier health monitoring that previously required periodic manual assessments. | (baseline_supplier_review_hours_per_month - current_supplier_review_hours_per_month) * months_in_period * labor_rate_cents_per_hour | months_in_period; labor_rate_cents_per_hour — assumption; current_supplier_review_hours_per_month — reading of supply_chain.suppliers.review_hours_per_month; baseline_supplier_review_hours_per_month — baseline of supply_chain.suppliers.review_hours_per_month |
| Supply Disruption Cost Avoidance; Estimates the cost avoided by detecting supplier risk early enough to activate alternatives before a disruption occurs. Each prevented disruption avoids production stoppages, emergency sourcing premiums, and customer impact costs. Disruptions avoided a year, spread evenly over its months, × average cost per disruption. | (baseline_disruption_incidents_per_year - current_disruption_incidents_per_year) * months_in_period / 12 * avg_disruption_cost_cents | months_in_period; avg_disruption_cost_cents — assumption; current_disruption_incidents_per_year — reading of supply_chain.suppliers.disruptions_per_year; baseline_disruption_incidents_per_year — baseline of supply_chain.suppliers.disruptions_per_year |
| Training Administration Labor Savings; Monetizes the HR and compliance team time recovered by automating training enrollment, reminders, completion tracking, and attestation collection. Admin hours saved per cycle × the training cycles in the period (a year's cycles spread evenly over its months) × blended rate. | (baseline_training_admin_hours - current_training_admin_hours) * training_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; training_cycles_per_year — assumption; labor_rate_cents_per_hour — assumption; current_training_admin_hours — reading of legal.training.admin_hours_per_cycle; baseline_training_admin_hours — baseline of legal.training.admin_hours_per_cycle |
| Vendor Onboarding Delay Cost Avoidance; Estimates the business value unlocked by onboarding vendors faster. Each day of onboarding delay is a day the vendor relationship isn't generating value — whether that's a new software deployment, a new supplier, or a new service contract. | vendors_onboarded * (baseline_onboarding_days - current_onboarding_days) * avg_daily_blocked_value_cents | vendors_onboarded — reading of legal.vendors.onboarded_count; current_onboarding_days — reading of legal.vendors.onboarding_days; baseline_onboarding_days — baseline of legal.vendors.onboarding_days; avg_daily_blocked_value_cents — assumption |

## Marketing
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Attribution Reporting Labor Savings; Monetizes the marketing analyst time recovered by replacing manual multi-channel reporting with automated attribution dashboards. | (baseline_reporting_hours_per_week - current_reporting_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_reporting_hours_per_week — reading of marketing.reporting.hours_per_week; baseline_reporting_hours_per_week — baseline of marketing.reporting.hours_per_week |
| Budget Reallocation Revenue Uplift; Estimates incremental revenue from reallocating spend from low-performing to high-ROI channels. A reallocated dollar adds the difference between what it earns where it went and what it earned where it was, not the whole of what it earns now: spend reallocated × (pipeline per dollar in the channels it went to − pipeline per dollar in the channels it left) × close rate. | budget_reallocated_usd * (destination_pipeline_per_dollar - source_pipeline_per_dollar) * close_rate_decimal * 100 | close_rate_decimal — assumption; budget_reallocated_usd — reading of marketing.budget.reallocated; source_pipeline_per_dollar — reading of marketing.budget.source_pipeline_per_dollar; destination_pipeline_per_dollar — reading of marketing.budget.destination_pipeline_per_dollar |
| Campaign Setup Labor Savings; Monetizes the marketing ops time recovered by automating campaign configuration, audience segmentation, and A/B test setup. Campaigns launched × hours saved per campaign × labor rate. | campaigns_launched * (baseline_hours_per_campaign - current_hours_per_campaign) * labor_rate_cents_per_hour | campaigns_launched — reading of marketing.campaigns.launched_count; labor_rate_cents_per_hour — assumption; current_hours_per_campaign — reading of marketing.campaigns.hours_per_launch; baseline_hours_per_campaign — baseline of marketing.campaigns.hours_per_launch |
| Content Production Labor Savings; Monetizes writer and designer time recovered when AI drafting and repurposing accelerates content production. Pieces produced × hours saved per piece × blended content team rate. | content_pieces_produced * (baseline_hours_per_piece - current_hours_per_piece) * labor_rate_cents_per_hour | content_pieces_produced — reading of marketing.content.published_count; current_hours_per_piece — reading of marketing.content.hours_per_piece; baseline_hours_per_piece — baseline of marketing.content.hours_per_piece; labor_rate_cents_per_hour — assumption |
| Content Volume Outsourcing Cost Avoidance; Estimates the agency or freelancer spend avoided by producing more content in-house with AI assistance. Blog posts published in the period above the monthly count before AI × average cost to outsource an equivalent post. Both counts are posts published on the connected CMS, so the before-value is the account's own months before the initiative. | (current_published_count - baseline_published_count) * avg_outsourced_content_cost_cents | current_published_count — reading of marketing.content.published_count; baseline_published_count — baseline of marketing.content.published_count; avg_outsourced_content_cost_cents — assumption |
| Email Performance Revenue Uplift; Estimates incremental revenue driven by improved email open rates from send-time optimization and behavioral personalization. Campaigns × list size × open rate improvement × click-to-conversion rate × average conversion value. | campaigns_launched * emails_per_campaign * (current_open_rate_pct - baseline_open_rate_pct) / 100 * click_to_conversion_rate_decimal * avg_conversion_value_cents | campaigns_launched — reading of marketing.campaigns.launched_count; emails_per_campaign — reading of marketing.campaigns.avg_list_size; current_open_rate_pct — reading of marketing.emails.open_rate; baseline_open_rate_pct — baseline of marketing.emails.open_rate; avg_conversion_value_cents — assumption; click_to_conversion_rate_decimal — assumption |
| MQL-to-SQL Conversion Revenue Uplift; Estimates incremental pipeline revenue from improving the MQL-to-SQL conversion rate. Higher conversion means more opportunities in the pipeline without growing top-of-funnel spend. | mqls_in_period * (current_mql_to_sql_rate_pct - baseline_mql_to_sql_rate_pct) / 100 * avg_deal_value_cents * close_rate_decimal | mqls_in_period — reading of crm.contacts.mql_count; close_rate_decimal; crm_deals_won_count — reading of crm.deals.won_count; avg_deal_value_cents; crm_contacts_sql_count — reading of crm.contacts.sql_count; crm_deals_win_rate_pct — reading of crm.deals.win_rate; crm_deals_won_value_usd — reading of crm.deals.won_value; current_mql_to_sql_rate_pct; baseline_mql_to_sql_rate_pct — baseline of crm.contacts.mql_to_sql_rate |
| Nurture Program Labor Savings; Monetizes the demand gen team time recovered by replacing manual follow-up and segmentation tasks with automated nurture sequences. | (baseline_nurture_hours_per_week - current_nurture_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_nurture_hours_per_week — reading of marketing.nurture.manual_hours_per_week; baseline_nurture_hours_per_week — baseline of marketing.nurture.manual_hours_per_week |
| Paid Acquisition Cost Avoidance; Estimates the paid media spend saved by growing organic traffic. Each additional organic session replaces a paid click that would otherwise be needed to maintain total traffic. Incremental organic sessions × blended cost per equivalent paid click. | (current_organic_sessions_per_month - baseline_organic_sessions_per_month) * months_in_period * cost_per_equivalent_paid_click_cents | months_in_period; current_organic_sessions_per_month — reading of marketing.web.organic_sessions; baseline_organic_sessions_per_month — baseline of marketing.web.organic_sessions; cost_per_equivalent_paid_click_cents — assumption |
| SEO Research & Optimization Labor Savings; Monetizes the content and SEO team time recovered by automating keyword research, gap analysis, and on-page optimization recommendations. | (baseline_seo_research_hours_per_month - current_seo_research_hours_per_month) * months_in_period * labor_rate_cents_per_hour | months_in_period; labor_rate_cents_per_hour — assumption; current_seo_research_hours_per_month — reading of marketing.seo.research_hours_per_month; baseline_seo_research_hours_per_month — baseline of marketing.seo.research_hours_per_month |

## Customer success
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Agent Handle Time Savings; Monetizes the agent time recovered when AI routing and suggested responses reduce the work per handled ticket. Total tickets handled × reduction in agent handle hours per ticket × agent labor rate. Handle time is the time agents work a ticket; a faster first response is the customer waiting less, not agents working less, and is not priced here. | tickets_handled * (baseline_handle_hours - current_handle_hours) * labor_rate_cents_per_hour | tickets_handled — reading of support.tickets.closed_count; current_handle_hours — reading of support.tickets.avg_handle_time; baseline_handle_hours — baseline of support.tickets.avg_handle_time; labor_rate_cents_per_hour — assumption |
| CSM Capacity Expansion Revenue Value; Estimates the incremental recurring revenue accessible when CSMs can manage more accounts with the same headcount. Accounts freed up per CSM × number of CSMs × average account MRR × renewal rate: one month of revenue from the added coverage. | (current_accounts_per_csm - baseline_accounts_per_csm) * csm_count * avg_account_mrr_cents * renewal_rate_decimal | csm_count — reading of customer_success.accounts.csm_count; subscriber_count — reading of billing.subscriptions.active_count; renewal_rate_decimal — assumption; avg_account_mrr_cents; current_accounts_per_csm — reading of customer_success.accounts.per_csm; baseline_accounts_per_csm — baseline of customer_success.accounts.per_csm; billing_subscriptions_mrr_usd — reading of billing.subscriptions.mrr |
| Documentation Maintenance Labor Savings; Monetizes the time recovered when AI keeps knowledge base articles fresh automatically, replacing the manual review and update process. Articles × minutes saved per article per review × the review cycles in the period (a year's cycles spread evenly over its months) × blended rate. | kb_articles_maintained * (baseline_update_minutes_per_article - current_update_minutes_per_article) * update_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_minute | months_in_period; kb_articles_maintained — reading of support.knowledge_base.article_count; update_cycles_per_year — assumption; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute; current_update_minutes_per_article — reading of support.knowledge_base.update_minutes_per_article; baseline_update_minutes_per_article — baseline of support.knowledge_base.update_minutes_per_article |
| MRR Retained from Churn Prevention; Estimates the recurring revenue retained in the period by reducing churn. Reduction in monthly churn × total MRR is the MRR that would have canceled this month and did not; the value is that month of revenue, not a year of it. Net revenue retention on the same template already counts the churn it prevents, so an initiative gets one of the two: this one where billing reports churn. | total_mrr_cents * (baseline_churn_rate_pct - current_churn_rate_pct) / 100 | total_mrr_cents; current_churn_rate_pct — reading of billing.subscriptions.churn_rate; baseline_churn_rate_pct — baseline of billing.subscriptions.churn_rate; billing_subscriptions_mrr_usd — reading of billing.subscriptions.mrr |
| NRR Improvement Revenue Value; Estimates the recurring revenue added by improving net revenue retention. Each percentage point of NRR improvement lifts the revenue kept from existing customers; the value is one month of that lift, the period's MRR × the improvement, not the annual run rate. | billing_subscriptions_mrr_usd * (current_nrr_pct - baseline_nrr_pct) / 100 * 100 | current_nrr_pct — reading of customer_success.revenue.net_retention_rate; baseline_nrr_pct — baseline of customer_success.revenue.net_retention_rate; billing_subscriptions_mrr_usd — reading of billing.subscriptions.mrr |
| QA Program Labor Savings; Monetizes the QA analyst time recovered by replacing manual interaction sampling with automated AI review. With AI QA covering 100% of interactions at a fraction of the cost, the manual sampling program is eliminated or dramatically reduced. | interactions_per_period * baseline_qa_sample_rate_pct / 100 * avg_manual_review_minutes * labor_rate_cents_per_minute | interactions_per_period — reading of support.tickets.volume; avg_manual_review_minutes — reading of support.quality.review_minutes_per_interaction; labor_rate_cents_per_hour — assumption; baseline_qa_sample_rate_pct — baseline of support.quality.sample_rate; labor_rate_cents_per_minute |
| Renewal Rate Revenue Uplift; Estimates the recurring revenue retained by improving renewal rates through automated QBR prep and proactive renewal workflows. ARR up for renewal × improvement in renewal rate is the ARR kept; the value is the period's share of it, one month of revenue for a month's renewals. | arr_up_for_renewal_usd * (current_renewal_rate_pct - baseline_renewal_rate_pct) / 100 * months_in_period / 12 * 100 | months_in_period; arr_up_for_renewal_usd — reading of customer_success.renewals.arr_up_for_renewal; current_renewal_rate_pct — reading of customer_success.renewals.rate; baseline_renewal_rate_pct — baseline of customer_success.renewals.rate |
| Self-Serve Ticket Deflection Savings; Monetizes the support cost avoided when customers resolve issues via the knowledge base rather than submitting tickets. Increase in deflection rate × total ticket volume × cost per handled ticket. | total_tickets_per_period * (current_deflection_rate_pct - baseline_deflection_rate_pct) / 100 * avg_cost_per_handled_ticket_usd * 100 | total_tickets_per_period — reading of support.tickets.volume; current_deflection_rate_pct — reading of support.tickets.self_serve_deflection_rate; baseline_deflection_rate_pct — baseline of support.tickets.self_serve_deflection_rate; avg_cost_per_handled_ticket_usd — reading of support.tickets.cost_per_ticket |

## Supply chain
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Excess Inventory Carrying Cost Reduction; Estimates the carrying cost saved by holding less excess inventory. Better demand forecasts reduce safety stock requirements and overbuying. Avg inventory value × points of forecast accuracy gained × inventory released per point × the period's share of the annual carrying cost rate. | avg_inventory_value_usd * (current_forecast_accuracy_pct - baseline_forecast_accuracy_pct) * target_inventory_reduction_decimal * annual_carrying_cost_rate_decimal * months_in_period / 12 * 100 | months_in_period; avg_inventory_value_usd — reading of supply_chain.inventory.avg_value; current_forecast_accuracy_pct — reading of supply_chain.forecasting.accuracy; baseline_forecast_accuracy_pct — baseline of supply_chain.forecasting.accuracy; annual_carrying_cost_rate_decimal — assumption; target_inventory_reduction_decimal — assumption |
| Invoice Matching Labor Savings; Monetizes the AP team time recovered by automating three-way PO-receipt-invoice matching and exception handling. | (baseline_invoice_matching_hours_per_week - current_invoice_matching_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_invoice_matching_hours_per_week — reading of supply_chain.procurement.invoice_matching_hours_per_week; baseline_invoice_matching_hours_per_week — baseline of supply_chain.procurement.invoice_matching_hours_per_week |
| Late Delivery Penalty & Expedite Cost Avoidance; Estimates the cost avoided when better route planning improves on-time delivery — avoiding SLA penalties, customer credits, and expensive expedite orders. Shipments × on-time delivery improvement × average cost per late delivery. | shipments_per_period * (current_on_time_delivery_rate_pct - baseline_on_time_delivery_rate_pct) / 100 * avg_late_delivery_cost_cents | shipments_per_period — reading of supply_chain.shipments.count; avg_late_delivery_cost_cents — assumption; current_on_time_delivery_rate_pct — reading of supply_chain.deliveries.on_time_rate; baseline_on_time_delivery_rate_pct — baseline of supply_chain.deliveries.on_time_rate |
| Pick Error Correction Cost Savings; Estimates the cost avoided from fewer mispicks. Each pick error triggers a customer return, reshipping, and internal correction cycle — all of which are eliminated with more accurate directed-pick workflows. | total_picks_per_period * (baseline_pick_error_rate_pct - current_pick_error_rate_pct) / 100 * avg_pick_error_cost_cents | total_picks_per_period — reading of supply_chain.warehouse.picks_count; avg_pick_error_cost_cents — assumption; current_pick_error_rate_pct — reading of supply_chain.warehouse.pick_error_rate; baseline_pick_error_rate_pct — baseline of supply_chain.warehouse.pick_error_rate |
| Pick Labor Productivity Savings; Estimates warehouse labor cost saved from improved pick productivity. Directed picking and optimized slotting allow the same number of picks to be completed with fewer labor hours. Total picks × labor hours saved per pick (1 ÷ picks per hour before, less 1 ÷ picks per hour now) × labor rate. | total_picks_per_period * (1 / baseline_picks_per_labor_hour - 1 / current_picks_per_labor_hour) * labor_rate_cents_per_hour | total_picks_per_period — reading of supply_chain.warehouse.picks_count; labor_rate_cents_per_hour — assumption; current_picks_per_labor_hour — reading of supply_chain.warehouse.picks_per_labor_hour; baseline_picks_per_labor_hour — baseline of supply_chain.warehouse.picks_per_labor_hour |
| PO Cycle Labor Savings; Monetizes procurement staff time recovered by automating purchase order creation, approval routing, and invoice matching. POs processed × (hours per PO before − after) × blended procurement labor rate. | pos_processed_per_period * (baseline_hours_per_po - current_hours_per_po) * labor_rate_cents_per_hour | current_hours_per_po — reading of supply_chain.procurement.hours_per_po; baseline_hours_per_po — baseline of supply_chain.procurement.hours_per_po; pos_processed_per_period — reading of supply_chain.procurement.purchase_orders_processed; labor_rate_cents_per_hour — assumption |
| Shipping Cost Reduction; Measures direct shipping cost savings from AI-optimized carrier selection and route planning. Shipments in period × reduction in cost per shipment. | shipments_per_period * (baseline_cost_per_shipment_usd - current_cost_per_shipment_usd) * 100 | shipments_per_period — reading of supply_chain.shipments.count; current_cost_per_shipment_usd — reading of supply_chain.shipments.cost_per_shipment; baseline_cost_per_shipment_usd — baseline of supply_chain.shipments.cost_per_shipment |
| Stockout Lost Revenue Recovery; Estimates revenue recovered by reducing stockouts. Fewer out-of-stock situations means fewer lost sales, fewer expedite orders, and better customer fill rates. | total_revenue_cents * (baseline_stockout_rate_pct - current_stockout_rate_pct) / 100 | total_revenue_cents; financials_revenue_usd — reading of finance.pnl.revenue; current_stockout_rate_pct — reading of supply_chain.inventory.stockout_rate; baseline_stockout_rate_pct — baseline of supply_chain.inventory.stockout_rate |

## HR
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Faster Time-to-Productivity Value; Estimates the value of reducing the ramp period. Shorter ramp × daily productive output value × number of hires. | new_hires_onboarded * (baseline_days_to_productivity - current_days_to_productivity) * avg_daily_output_value_cents | new_hires_onboarded — reading of hr.onboarding.hires_count; avg_daily_output_value_cents — assumption; current_days_to_productivity — reading of hr.onboarding.days_to_productivity; baseline_days_to_productivity — baseline of hr.onboarding.days_to_productivity |
| HR Inquiry Deflection Labor Savings; Monetizes the HR team interrupt cost avoided when the chatbot deflects routine policy, PTO, and benefits questions. Inquiries deflected × average cost for HR to handle one inquiry manually. | tier1_inquiries_per_period * deflection_rate_decimal * avg_hr_handling_cost_cents | deflection_rate_decimal — reading of hr.inquiries.chatbot_deflection_rate; avg_hr_handling_cost_cents — assumption; tier1_inquiries_per_period — reading of hr.inquiries.tier1_per_month |
| Onboarding HR Time Savings; Monetizes the HR admin time recovered by automating onboarding tasks. New hires × hours saved per hire × blended HR rate. | new_hires_onboarded * (baseline_hr_hours_per_hire - current_hr_hours_per_hire) * blended_labor_rate_cents_per_hour | new_hires_onboarded — reading of hr.onboarding.hires_count; current_hr_hours_per_hire — reading of hr.onboarding.admin_hours_per_hire; baseline_hr_hours_per_hire — baseline of hr.onboarding.admin_hours_per_hire; blended_labor_rate_cents_per_hour — assumption |
| Recruiter Screening Time Savings; Monetizes the recruiter time recovered by replacing manual resume review with AI screening. Applicants screened × hours saved per hire × blended recruiter rate. | hires_in_period * (baseline_recruiter_hours_per_hire - current_recruiter_hours_per_hire) * labor_rate_cents_per_hour | hires_in_period — reading of hr.hiring.hires_count; labor_rate_cents_per_hour — assumption; current_recruiter_hours_per_hire — reading of hr.hiring.recruiter_hours_per_hire; baseline_recruiter_hours_per_hire — baseline of hr.hiring.recruiter_hours_per_hire |
| Review Cycle Admin Labor Savings; Monetizes the HR and manager admin time recovered each review cycle. Headcount reviewed × (admin hours per employee before − after) × labor rate × the review cycles in the period (a year's cycles spread evenly over its months). | employees_reviewed * (baseline_admin_hours_per_employee - current_admin_hours_per_employee) * labor_rate_cents_per_hour * review_cycles_per_year * months_in_period / 12 | months_in_period; employees_reviewed — reading of hr.reviews.employees_reviewed; review_cycles_per_year — assumption; labor_rate_cents_per_hour — assumption; current_admin_hours_per_employee — reading of hr.reviews.admin_hours_per_employee; baseline_admin_hours_per_employee — baseline of hr.reviews.admin_hours_per_employee |
| Time-to-Fill Vacancy Cost Reduction; Estimates the business cost avoided by filling roles faster. Every day a role is open carries a productivity cost — either lost output or cost of coverage. Hires × days reduced × daily vacancy cost. | hires_in_period * (baseline_days_to_first_interview - current_days_to_first_interview) * avg_daily_vacancy_cost_cents | hires_in_period — reading of hr.hiring.hires_count; avg_daily_vacancy_cost_cents — assumption; current_days_to_first_interview — reading of hr.hiring.days_to_first_interview; baseline_days_to_first_interview — baseline of hr.hiring.days_to_first_interview |
| Voluntary Attrition Cost Avoidance; Estimates the cost avoided by reducing unwanted voluntary attrition. Each prevented departure avoids recruiting, onboarding, and productivity ramp costs. Headcount × reduction in the annualized attrition rate × the period's share of a year gives the departures avoided in the period, × average replacement cost. | total_headcount * (baseline_attrition_rate_pct - current_attrition_rate_pct) / 100 * months_in_period / 12 * avg_replacement_cost_cents | total_headcount — reading of hr.workforce.headcount; months_in_period; avg_replacement_cost_cents — assumption; current_attrition_rate_pct — reading of hr.workforce.voluntary_attrition_rate; baseline_attrition_rate_pct — baseline of hr.workforce.voluntary_attrition_rate |
| Workforce Reporting Labor Savings; Monetizes the HR analyst time recovered by replacing manual headcount and workforce data reports with automated analytics. Hours saved per month × months × labor rate. | (baseline_reporting_hours_per_month - current_reporting_hours_per_month) * months_in_period * labor_rate_cents_per_hour | months_in_period; labor_rate_cents_per_hour — assumption; current_reporting_hours_per_month — reading of hr.reporting.hours_per_month; baseline_reporting_hours_per_month — baseline of hr.reporting.hours_per_month |

## Operations
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Data Entry Labor Savings; Monetizes the staff time recovered by automating document or form data extraction. Volume × per-item time saved × labor rate. | documents_processed * (baseline_minutes_per_doc - current_minutes_per_doc) * labor_rate_cents_per_minute | documents_processed — reading of operations.documents.processed_count; current_minutes_per_doc — reading of operations.documents.data_entry_minutes_per_document; baseline_minutes_per_doc — baseline of operations.documents.data_entry_minutes_per_document; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute |
| Error Rework Cost Savings; Monetizes the reduction in correction and reconciliation work driven by lower AI error rates vs. manual keying. | documents_processed * (baseline_error_rate_pct - current_error_rate_pct) / 100 * avg_rework_cost_cents | documents_processed — reading of operations.documents.processed_count; avg_rework_cost_cents — assumption; current_error_rate_pct — reading of operations.processing.error_rate; baseline_error_rate_pct — baseline of operations.processing.error_rate |
| Manual Work Replaced by Automated Workflows; Values the manual work automated workflows replace: successful runs in the period, summed across workflows, × the minutes of manual work one run replaces × the ops hourly rate. Runs come from n8n, Make or Zapier; the minutes and the rate are Assumptions an account can override. | successful_runs * minutes_per_run * ops_hourly_rate / 60 | minutes_per_run — assumption; ops_hourly_rate — assumption; successful_runs |
| Meeting Time Saved; Values the time returned to the team by shorter, smaller or skipped meetings: attendee-hours spent in meetings before the assistant, less attendee-hours now, at the blended hourly rate. Both before-values come from the account's own calendar history. Attendee counts are the people invited, which is what a calendar records, on both sides of the comparison. | (baseline_meeting_hours * baseline_attendees_per_meeting - meeting_hours * attendees_per_meeting) * blended_labor_rate_cents_per_hour | meeting_hours — reading of calendar.meetings.hours; attendees_per_meeting — reading of calendar.meetings.avg_attendees; baseline_meeting_hours — baseline of calendar.meetings.hours; baseline_attendees_per_meeting — baseline of calendar.meetings.avg_attendees; blended_labor_rate_cents_per_hour — assumption |
| Reporting Time Savings; Monetizes the staff hours recovered by automating report production. Multiply weekly hours saved by the number of weeks measured and the team's blended hourly rate. | hours_saved_per_week * weeks_in_period * blended_labor_rate_cents_per_hour | weeks_in_period; hours_saved_per_week — reading of workforce.time.hours_saved_per_week; blended_labor_rate_cents_per_hour — assumption |

## Construction engineering
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Cost Overrun Margin Protection; Estimates gross margin protected by catching and correcting budget overruns earlier in the project lifecycle. Real-time EAC forecasting surfaces cost variance weeks before month-end reviews — enabling corrective action while there is still margin to protect. An overrun is incurred as the work is put in place, so the base is the period's contract revenue earned, not the contract value of every active project, which would count a two-year job again every month: revenue × fall in the overrun rate × average overrun depth. | financials_revenue_usd * (baseline_overrun_rate_pct - current_overrun_rate_pct) / 100 * avg_overrun_depth_decimal * 100 | financials_revenue_usd — reading of finance.pnl.revenue; current_overrun_rate_pct — reading of construction.projects.cost_overrun_rate; avg_overrun_depth_decimal — assumption; baseline_overrun_rate_pct — baseline of construction.projects.cost_overrun_rate |
| Document Search Labor Savings; Monetizes field and office staff time recovered by replacing manual document hunting with AI-powered search across the Common Data Environment. Hours saved per week × weeks × blended field/office rate. | (baseline_document_search_hours_per_week - current_document_search_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_document_search_hours_per_week — reading of construction.documents.search_hours_per_week; baseline_document_search_hours_per_week — baseline of construction.documents.search_hours_per_week |
| Estimate Accuracy Margin Protection; Estimates gross margin protected by improving bid accuracy. Inaccurate estimates — whether too low (creating margin loss on awarded projects) or too high (losing winnable bids) — destroy value. Tighter estimates using historical actuals protects margin on won work. | awarded_project_value_usd * (baseline_estimate_error_pct - current_estimate_error_pct) / 100 * avg_project_margin_decimal * 100 | awarded_project_value_usd — reading of construction.bids.awarded_value; avg_project_margin_decimal — assumption; current_estimate_error_pct — reading of construction.estimating.variance_rate; baseline_estimate_error_pct — baseline of construction.estimating.variance_rate |
| Estimating Labor Savings; Monetizes estimator time recovered by automating quantity takeoffs and bid assembly. Each hour saved per bid compounds across the bid volume — allowing the same estimating team to respond to more opportunities without additional headcount. | bids_submitted_per_period * (baseline_estimating_hours_per_bid - current_estimating_hours_per_bid) * labor_rate_cents_per_hour | bids_submitted_per_period — reading of construction.bids.submitted_count; labor_rate_cents_per_hour — assumption; current_estimating_hours_per_bid — reading of construction.estimating.hours_per_bid; baseline_estimating_hours_per_bid — baseline of construction.estimating.hours_per_bid |
| Job Cost Review Labor Savings; Monetizes project manager and controller time recovered by automating job cost data aggregation and variance reporting. With always-current dashboards, the manual effort of pulling data into spreadsheets for periodic reviews is eliminated. | active_projects * (baseline_review_hours_per_project_per_cycle - current_review_hours_per_project_per_cycle) * review_cycles_per_period * labor_rate_cents_per_hour | active_projects — reading of construction.projects.active_count; review_cycles_per_period — reading of construction.job_costing.review_cycles; labor_rate_cents_per_hour — assumption; current_review_hours_per_project_per_cycle — reading of construction.job_costing.review_hours_per_project; baseline_review_hours_per_project_per_cycle — baseline of construction.job_costing.review_hours_per_project |
| Payroll Processing Labor Savings; Monetizes payroll and office staff time recovered by replacing paper timecard collection and manual reconciliation with digital time capture. Hours saved per week × weeks × blended payroll/admin rate. | (baseline_payroll_hours_per_week - current_payroll_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_payroll_hours_per_week — reading of construction.payroll.processing_hours_per_week; baseline_payroll_hours_per_week — baseline of construction.payroll.processing_hours_per_week |
| Recordable Incident Cost Avoidance; Estimates the cost avoided by reducing recordable safety incidents. Each recordable incident carries direct costs (medical, OSHA reporting, investigation) and indirect costs (project disruption, crew morale, insurance premium impact, and potential contract disqualification). | total_labor_hours_worked * (baseline_recordable_incident_rate - current_recordable_incident_rate) / 200000 * avg_recordable_incident_cost_cents | total_labor_hours_worked — reading of construction.labor.hours_worked; current_recordable_incident_rate — reading of construction.safety.recordable_incident_rate; baseline_recordable_incident_rate — baseline of construction.safety.recordable_incident_rate; avg_recordable_incident_cost_cents — assumption |
| RFI Delay Cost Reduction; Estimates the cost avoided from faster RFI resolution. Unanswered RFIs block field crews and idle subcontractors — each day of RFI delay has a direct impact on project schedule and labor cost. Faster routing and automated precedent surfacing compresses response time. | rfis_per_period * (baseline_rfi_response_days - current_rfi_response_days) * avg_daily_rfi_delay_cost_cents | rfis_per_period — reading of construction.rfis.submitted_count; current_rfi_response_days — reading of construction.rfis.response_days; baseline_rfi_response_days — baseline of construction.rfis.response_days; avg_daily_rfi_delay_cost_cents — assumption |
| Safety Administration Labor Savings; Monetizes safety officer and superintendent time recovered by digitizing safety observation reporting, JHA distribution, and inspection checklists. Hours saved per week × weeks × blended safety staff rate. | (baseline_safety_observation_hours_per_week - current_safety_observation_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_safety_observation_hours_per_week — reading of construction.safety.observation_hours_per_week; baseline_safety_observation_hours_per_week — baseline of construction.safety.observation_hours_per_week |
| Timecard Error Correction Cost Savings; Estimates the cost avoided from eliminating payroll errors caused by paper timecard mistakes. Each corrected payroll error triggers manual correction workflows, potential payroll reprocessing fees, and compliance risk. Total timecards × error reduction × cost per error. | timecards_processed_per_period * (baseline_timecard_error_rate_pct - current_timecard_error_rate_pct) / 100 * avg_correction_cost_cents | avg_correction_cost_cents — assumption; timecards_processed_per_period — reading of construction.payroll.timecards_processed; current_timecard_error_rate_pct — reading of construction.payroll.timecard_error_rate; baseline_timecard_error_rate_pct — baseline of construction.payroll.timecard_error_rate |

## Education
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Admissions Staff Time Savings; Monetizes the admissions team time recovered by automating application review, document verification, and routine communications. Applications processed × staff minutes saved per application × staff rate. The minutes are hands-on review time, not the days an application waits for a decision. | applications_processed * (baseline_review_minutes_per_application - current_review_minutes_per_application) * labor_rate_cents_per_minute | applications_processed — reading of operations.applications.processed_count; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute; current_review_minutes_per_application — reading of education.admissions.review_minutes_per_application; baseline_review_minutes_per_application — baseline of education.admissions.review_minutes_per_application |
| Advising & Outreach Labor Savings; Monetizes the advising and student-success staff time recovered by automating routine at-risk identification and initial outreach, freeing staff for high-touch interventions. | (baseline_outreach_hours_per_week - current_outreach_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_outreach_hours_per_week — reading of workforce.outreach.hours_per_week; baseline_outreach_hours_per_week — baseline of workforce.outreach.hours_per_week |
| DFW Rate Reduction — Revenue Protection; Estimates tuition revenue protected by reducing the rate of students who fail, withdraw, or earn a D. A D, F or W does not always end a student's enrollment, so only the departures it adds count: enrolled students × DFW improvement × the extra share of DFW students who leave × the period's share of annual revenue per student. | enrolled_students * (baseline_dfw_rate_pct - current_dfw_rate_pct) / 100 * dfw_attrition_rate_decimal * avg_annual_revenue_per_student_cents * months_in_period / 12 | months_in_period; enrolled_students — reading of education.enrollment.student_count; current_dfw_rate_pct — reading of education.courses.dfw_rate; baseline_dfw_rate_pct — baseline of education.courses.dfw_rate; dfw_attrition_rate_decimal — assumption; avg_annual_revenue_per_student_cents — assumption |
| Faculty Administrative Time Savings; Monetizes faculty hours recovered from course administration — syllabus drafting, assessment creation, objective grading, and content formatting. Faculty × hours saved per week × weeks × faculty hourly cost. | faculty_count * (baseline_admin_hours_per_week - current_admin_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | faculty_count — reading of education.faculty.count; weeks_in_period; labor_rate_cents_per_hour — assumption; current_admin_hours_per_week — reading of education.faculty.admin_hours_per_week; baseline_admin_hours_per_week — baseline of education.faculty.admin_hours_per_week |
| Improved Yield Revenue; Estimates incremental tuition revenue from improving enrollment yield through faster decisions and personalized outreach. Yield is enrolled ÷ admitted, so the gain applies to students admitted, not to every applicant: admitted × yield improvement × the period's share of avg annual revenue per enrolled student — a month's tuition from the added enrollees. | admitted_in_period * (current_yield_rate_pct - baseline_yield_rate_pct) / 100 * avg_annual_revenue_per_enrolled_student_cents * months_in_period / 12 | months_in_period; admitted_in_period — reading of education.admissions.admitted_count; current_yield_rate_pct — reading of education.admissions.yield_rate; baseline_yield_rate_pct — baseline of education.admissions.yield_rate; avg_annual_revenue_per_enrolled_student_cents — assumption |
| Retained Student Revenue Value; Estimates revenue protected by improving retention. Each percentage point of retention improvement keeps more students enrolled and paying tuition. Retention gain × enrolled students × the period's share of annual revenue per student: a month's tuition from the students kept. | (current_retention_rate_pct - baseline_retention_rate_pct) / 100 * students_enrolled * avg_annual_revenue_per_student_cents * months_in_period / 12 | months_in_period; students_enrolled — reading of education.enrollment.student_count; current_retention_rate_pct — reading of education.enrollment.retention_rate; baseline_retention_rate_pct — baseline of education.enrollment.retention_rate; avg_annual_revenue_per_student_cents — assumption |
| Tutoring Cost Reduction; Estimates savings from replacing a portion of labor-intensive human tutoring hours with AI tutoring. Reduction in tutoring cost per student in the period × enrolled students. | (baseline_tutoring_cost_per_student_usd - current_tutoring_cost_per_student_usd) * enrolled_students * 100 | enrolled_students — reading of education.enrollment.student_count; current_tutoring_cost_per_student_usd — reading of education.tutoring.cost_per_student; baseline_tutoring_cost_per_student_usd — baseline of education.tutoring.cost_per_student |

## Energy utilities
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| CapEx Deferral Value; Estimates the financial value of deferring asset replacement by extending healthy asset life through condition-based management. Deferred spend × the period's share of the annual cost of capital: what it costs to finance that spend for the period it stays deferred, booked month by month for as long as the deferral holds. | deferred_capex_usd * annual_cost_of_capital_decimal * months_in_period / 12 * 100 | months_in_period; deferred_capex_usd — reading of energy.assets.deferred_capex; annual_cost_of_capital_decimal — assumption |
| External Counsel Cost Reduction; Estimates external legal spend saved when AI-assisted drafting reduces reliance on outside counsel for regulatory filings and rate case support. Outside counsel hours saved per filing × filings in the period × counsel's billing rate. | (baseline_external_counsel_hours - current_external_counsel_hours) * rate_cases_per_period * external_counsel_rate_cents_per_hour | rate_cases_per_period — reading of energy.rate_cases.count; current_external_counsel_hours — reading of energy.rate_cases.external_counsel_hours; baseline_external_counsel_hours — baseline of energy.rate_cases.external_counsel_hours; external_counsel_rate_cents_per_hour — assumption |
| Forecasting & Planning Labor Savings; Monetizes the planning and trading team time recovered when AI replaces manual load forecasting and scenario modeling. Hours saved per cycle × cycles per period × blended staff rate. | (baseline_forecast_hours_per_cycle - current_forecast_hours_per_cycle) * forecast_cycles_per_period * labor_rate_cents_per_hour | labor_rate_cents_per_hour — assumption; forecast_cycles_per_period — reading of operations.forecasting.cycles; current_forecast_hours_per_cycle — reading of operations.forecasting.hours_per_cycle; baseline_forecast_hours_per_cycle — baseline of operations.forecasting.hours_per_cycle |
| Outage Notification Labor Savings; Monetizes the operations and communications team time recovered by automating customer outage status updates and ETR notifications, replacing manual call trees and email blasts. Staff hours saved per event × the events in the period (a year's events spread evenly over its months) × blended rate. | (baseline_notification_hours_per_event - current_notification_hours_per_event) * storm_events_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; storm_events_per_year — reading of energy.storms.events_per_year; labor_rate_cents_per_hour — assumption; current_notification_hours_per_event — reading of energy.storms.notification_hours_per_event; baseline_notification_hours_per_event — baseline of energy.storms.notification_hours_per_event |
| Rate Case Internal Labor Savings; Monetizes the internal legal, regulatory, and engineering hours recovered by AI-assisted drafting of rate case filings and regulatory responses. Hours saved per filing × filings per period × blended rate. | (baseline_rate_case_prep_hours - current_rate_case_prep_hours) * rate_cases_per_period * labor_rate_cents_per_hour | rate_cases_per_period — reading of energy.rate_cases.count; labor_rate_cents_per_hour — assumption; current_rate_case_prep_hours — reading of energy.rate_cases.prep_hours; baseline_rate_case_prep_hours — baseline of energy.rate_cases.prep_hours |
| Reserve Requirement Reduction Savings; Estimates capacity cost savings from reducing required reserves due to tighter AI load forecasts. More accurate forecasts mean less buffer capacity must be procured or held idle. MW reduced × capacity cost per MW per period. | (baseline_reserve_capacity_mw - current_reserve_capacity_mw) * capacity_cost_cents_per_mw_per_period | current_reserve_capacity_mw — reading of energy.grid.reserve_capacity_mw; baseline_reserve_capacity_mw — baseline of energy.grid.reserve_capacity_mw; capacity_cost_cents_per_mw_per_period — assumption |
| Storm Call Center Cost Avoidance; Estimates call-center cost avoided when proactive ETR communications deflect inbound customer calls during storm events. Calls deflected per event × the events in the period (a year's events spread evenly over its months) × cost per call. | (baseline_storm_calls_per_event - current_storm_calls_per_event) * storm_events_per_year * months_in_period / 12 * avg_cost_per_call_cents | months_in_period; storm_events_per_year — reading of energy.storms.events_per_year; avg_cost_per_call_cents — assumption; current_storm_calls_per_event — reading of energy.storms.call_volume_index; baseline_storm_calls_per_event — baseline of energy.storms.call_volume_index |
| Unplanned Outage Cost Avoidance; Estimates the cost avoided by preventing unplanned critical-asset failures. Each avoided outage eliminates emergency restoration labor, contractor spend, regulatory penalties, and customer compensation. Events avoided a year, spread evenly over its months, × average all-in cost per event. | (baseline_unplanned_outage_events_per_year - current_unplanned_outage_events_per_year) * months_in_period / 12 * avg_outage_event_cost_cents | months_in_period; avg_outage_event_cost_cents — assumption; current_unplanned_outage_events_per_year — reading of energy.assets.unplanned_outages_per_year; baseline_unplanned_outage_events_per_year — baseline of energy.assets.unplanned_outages_per_year |

## Financial services
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Advisor Capacity Revenue Value; Estimates revenue from enabling advisors to serve more accounts at the same service quality. AI reduces prep time per client interaction, expanding effective capacity. Additional accounts per advisor × advisor count × the period's share of average annual revenue per account. | (current_accounts_per_advisor - baseline_accounts_per_advisor) * advisor_count * avg_annual_revenue_per_account_cents * months_in_period / 12 | advisor_count — reading of financial_services.advisors.count; months_in_period; current_accounts_per_advisor — reading of financial_services.advisors.accounts_per_advisor; baseline_accounts_per_advisor — baseline of financial_services.advisors.accounts_per_advisor; avg_annual_revenue_per_account_cents — assumption |
| Data Error Remediation Cost Avoidance; Estimates the cost avoided from eliminating manual data errors that require correction, resubmission, regulatory inquiries, or internal investigation. The error rate is a share of line items, so it applies to the line items filed: report cycles in the period (a year's cycles spread evenly over its months) × line items in a cycle's reports × reduction in error rate × average cost to correct an erroneous line item. | report_cycles_per_year * months_in_period / 12 * line_items_per_cycle * (baseline_error_rate_pct - current_error_rate_pct) / 100 * avg_error_remediation_cost_cents | months_in_period; line_items_per_cycle — reading of financial_services.regulatory_reporting.line_items_per_cycle; current_error_rate_pct — reading of financial_services.regulatory_reporting.error_rate; report_cycles_per_year — assumption; baseline_error_rate_pct — baseline of financial_services.regulatory_reporting.error_rate; avg_error_remediation_cost_cents — assumption |
| False Positive Investigation Savings; Monetizes the fraud operations team time and customer friction avoided by reducing false positive alerts. Fewer false positives means fewer unnecessary investigations, fewer blocked legitimate customers, and lower operational burden. | transactions_flagged * (baseline_false_positive_rate_pct - current_false_positive_rate_pct) / 100 * avg_investigation_cost_cents | transactions_flagged — reading of financial_services.fraud.flagged_count; avg_investigation_cost_cents — assumption; current_false_positive_rate_pct — reading of financial_services.fraud.false_positive_rate; baseline_false_positive_rate_pct — baseline of financial_services.fraud.false_positive_rate |
| Fraud Loss Reduction; Estimates the direct reduction in fraud losses from improved detection accuracy. Better ML models catch more actual fraud before payment clears. Total transaction volume × reduction in fraud loss rate (in basis points). | total_transaction_volume_usd * (baseline_fraud_loss_rate_bps - current_fraud_loss_rate_bps) / 10000 * 100 | current_fraud_loss_rate_bps — reading of financial_services.fraud.loss_rate_bps; baseline_fraud_loss_rate_bps — baseline of financial_services.fraud.loss_rate_bps; total_transaction_volume_usd — reading of financial_services.transactions.volume |
| Incremental Revenue from Expanded Approvals; Estimates the revenue earned in the period from approving creditworthy applicants that rules-based systems would decline. A loan earns its revenue over its term, not the day it is approved, and these are the marginal applicants, so what one is worth is its lifetime revenue less its expected credit loss, spread over its term: applications × reduction in manual review rate × incremental approval rate × (lifetime revenue − expected loss) × months in the period ÷ loan term in months. Once the added loans are a steady book, that is what they earn each month; while the book is still building it errs low. | applications_processed * (baseline_manual_review_rate_pct - current_manual_review_rate_pct) / 100 * incremental_approval_rate_decimal * (avg_loan_revenue_cents - avg_expected_loss_cents) * months_in_period / avg_loan_term_months | months_in_period; avg_loan_term_months — assumption; applications_processed — reading of operations.applications.processed_count; avg_loan_revenue_cents — assumption; avg_expected_loss_cents — assumption; current_manual_review_rate_pct — reading of financial_services.underwriting.manual_review_rate; baseline_manual_review_rate_pct — baseline of financial_services.underwriting.manual_review_rate; incremental_approval_rate_decimal — assumption |
| KYC Review Labor Savings; Monetizes the compliance team time recovered by automating document review, sanctions screening, and identity verification. Accounts onboarded × hours saved per account × blended compliance analyst rate. | accounts_onboarded * (baseline_review_hours_per_account - current_review_hours_per_account) * labor_rate_cents_per_hour | accounts_onboarded — reading of financial_services.onboarding.accounts_count; labor_rate_cents_per_hour — assumption; current_review_hours_per_account — reading of financial_services.kyc.review_hours_per_account; baseline_review_hours_per_account — baseline of financial_services.kyc.review_hours_per_account |
| Onboarding Revenue Acceleration; Estimates revenue unlocked by onboarding customers faster. Accounts that are approved in hours rather than days begin generating revenue sooner. Accounts onboarded × days saved × average daily revenue per account. | accounts_onboarded * (baseline_onboarding_days - current_onboarding_days) * avg_daily_revenue_per_account_cents | accounts_onboarded — reading of financial_services.onboarding.accounts_count; current_onboarding_days — reading of financial_services.onboarding.days_to_active; baseline_onboarding_days — baseline of financial_services.onboarding.days_to_active; avg_daily_revenue_per_account_cents — assumption |
| Proactive Outreach Revenue Value; Estimates the fee revenue from AUM gained by advisors reaching out at higher-value moments. More proactive outreach means catching life events (inheritances, rollovers, rebalancing needs) that drive AUM inflows and referrals. AUM × rise in outreach rate × AUM gained per unit of outreach gives the added assets; the value is the advisory fee they earn in the period, not the assets themselves. | advisor_count * avg_aum_per_advisor_usd * (current_proactive_outreach_rate_pct - baseline_proactive_outreach_rate_pct) / 100 * outreach_to_aum_growth_rate_decimal * advisory_fee_rate_decimal * months_in_period / 12 * 100 | advisor_count — reading of financial_services.advisors.count; months_in_period; avg_aum_per_advisor_usd — baseline of financial_services.advisors.aum_per_advisor; advisory_fee_rate_decimal — assumption; current_proactive_outreach_rate_pct — reading of financial_services.advisors.proactive_outreach_rate; outreach_to_aum_growth_rate_decimal — assumption; baseline_proactive_outreach_rate_pct — baseline of financial_services.advisors.proactive_outreach_rate |
| Report Preparation Labor Savings; Monetizes the compliance and finance team time recovered by automating regulatory data aggregation and submission workflows. Hours saved per reporting cycle × the reporting cycles in the period (a year's cycles spread evenly over its months) × blended compliance rate. | (baseline_report_prep_hours - current_report_prep_hours) * report_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; report_cycles_per_year — assumption; current_report_prep_hours — reading of financial_services.regulatory_reporting.prep_hours_per_report; labor_rate_cents_per_hour — assumption; baseline_report_prep_hours — baseline of financial_services.regulatory_reporting.prep_hours_per_report |
| Underwriting Labor Savings; Monetizes the underwriter time recovered when AI handles automated approvals, reducing the volume of applications requiring manual review. Applications processed × underwriter hours saved per application × analyst rate. The hours are hands-on review time averaged over every application, an automatically decided one counting as none — not the elapsed time to a decision, which is mostly waiting. | applications_processed * (baseline_review_hours_per_application - current_review_hours_per_application) * labor_rate_cents_per_hour | applications_processed — reading of operations.applications.processed_count; labor_rate_cents_per_hour — assumption; current_review_hours_per_application — reading of financial_services.underwriting.review_hours_per_application; baseline_review_hours_per_application — baseline of financial_services.underwriting.review_hours_per_application |

## Government public sector
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Backlog Reduction Economic Value; Estimates the economic and social value of faster eligibility decisions — reducing the period applicants wait for housing, food, or unemployment assistance. Cases × days faster × estimated daily value of timely benefits delivery. | cases_per_period * (baseline_decision_days - current_decision_days) * estimated_daily_value_per_case_cents | cases_per_period — reading of government.cases.processed_count; current_decision_days — reading of government.cases.decision_days; baseline_decision_days — baseline of government.cases.decision_days; estimated_daily_value_per_case_cents — assumption |
| Call Deflection Cost Savings; Estimates contact-center cost saved by deflecting routine 311 and constituent service requests into AI self-service. Calls deflected per month × months × cost per call. | (baseline_calls_per_month - current_calls_per_month) * months_in_period * avg_cost_per_call_cents | months_in_period; avg_cost_per_call_cents — assumption; current_calls_per_month — reading of government.constituent_services.calls_per_month; baseline_calls_per_month — baseline of government.constituent_services.calls_per_month |
| Case Processing Labor Savings; Monetizes caseworker time recovered when AI handles document extraction, eligibility rules checking, and initial routing — freeing staff for complex cases and applicant support. Cases × caseworker hours saved per case × blended rate. The hours are hands-on work, not the days a case waits for a decision. | cases_processed_per_period * (baseline_processing_hours_per_case - current_processing_hours_per_case) * labor_rate_cents_per_hour | labor_rate_cents_per_hour — assumption; cases_processed_per_period — reading of government.cases.processed_count; current_processing_hours_per_case — reading of government.cases.staff_hours_per_case; baseline_processing_hours_per_case — baseline of government.cases.staff_hours_per_case |
| Economic Activity Value Unlocked; Estimates the broader economic value of faster permit decisions — construction, business openings, and development projects that are unblocked by reduced cycle times. Permits × days faster × estimated daily economic value per permit project. | permits_per_period * (baseline_decision_days - current_decision_days) * avg_daily_economic_value_per_permit_cents | permits_per_period — reading of government.permits.processed_count; current_decision_days — reading of government.permits.decision_days; baseline_decision_days — baseline of government.permits.decision_days; avg_daily_economic_value_per_permit_cents — assumption |
| Permit Processing Labor Savings; Monetizes staff time recovered when AI automates permit intake, document review, and inspection scheduling — reducing manual processing hours per permit. Permits × staff hours saved per permit × blended rate. The hours are hands-on work, not the days a permit waits for a decision. | permits_processed_per_period * (baseline_processing_hours_per_permit - current_processing_hours_per_permit) * labor_rate_cents_per_hour | labor_rate_cents_per_hour — assumption; permits_processed_per_period — reading of government.permits.processed_count; current_processing_hours_per_permit — reading of government.permits.staff_hours_per_permit; baseline_processing_hours_per_permit — baseline of government.permits.staff_hours_per_permit |
| Staff Redeployment Value; Estimates the value of constituent-services staff capacity freed by self-service deflection, redeployable to complex cases and high-touch constituent support that requires human judgment. The same deflected contacts are what call deflection cost savings prices at their full cost, staff time included, so an initiative gets one of the two. | (baseline_calls_per_month - current_calls_per_month) * months_in_period * avg_handle_time_minutes * labor_rate_cents_per_minute | months_in_period; avg_handle_time_minutes — reading of support.calls.avg_handle_time; current_calls_per_month — reading of government.constituent_services.calls_per_month; baseline_calls_per_month — baseline of government.constituent_services.calls_per_month; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute |

## Healthcare life sciences
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| AR Day Reduction Cash Value; Estimates the working capital benefit of collecting cash faster. Each day of AR reduction frees cash that was previously tied up in receivables. Monthly revenue × days reduced × daily cost of capital. | average_monthly_revenue_cents * (baseline_days_in_ar - current_days_in_ar) * daily_cost_of_capital_decimal | current_days_in_ar — reading of healthcare.revenue_cycle.days_in_ar; baseline_days_in_ar — baseline of healthcare.revenue_cycle.days_in_ar; financials_revenue_usd — reading of finance.pnl.revenue; average_monthly_revenue_cents; daily_cost_of_capital_decimal — assumption |
| Audit Preparation Labor Savings; Monetizes the compliance team time recovered by maintaining always-ready documentation rather than scrambling to prepare evidence for audits. Hours saved per cycle × the audit cycles in the period (a year's cycles spread evenly over its months) × blended compliance rate. | (baseline_audit_prep_hours - current_audit_prep_hours) * audit_cycles_per_year * months_in_period / 12 * labor_rate_cents_per_hour | months_in_period; audit_cycles_per_year — assumption; current_audit_prep_hours — reading of healthcare.audits.prep_hours_per_cycle; baseline_audit_prep_hours — baseline of healthcare.audits.prep_hours_per_cycle; labor_rate_cents_per_hour — assumption |
| Clinician Documentation Time Savings; Monetizes the clinical time recovered when ambient AI eliminates after-hours charting. Fewer documentation hours means reduced burnout, lower locum/overtime costs, and time that can be redeployed to patients. Clinicians × weekly hours saved × weeks × hourly clinical rate. | clinician_count * (baseline_after_hours_charting_hours - current_after_hours_charting_hours) * weeks_in_period * labor_rate_cents_per_hour | clinician_count — reading of healthcare.documentation.clinician_count; weeks_in_period; labor_rate_cents_per_hour — assumption; current_after_hours_charting_hours — reading of healthcare.documentation.after_hours_charting_hours_per_week; baseline_after_hours_charting_hours — baseline of healthcare.documentation.after_hours_charting_hours_per_week |
| Coder Productivity Value; Estimates revenue from increased coder throughput. AI-suggested codes allow coders to process more charts per day, reducing backlogs and accelerating the chart-to-claim timeline. Coders × incremental charts per day × avg revenue per chart × working days. | coder_count * (current_charts_per_coder_per_day - baseline_charts_per_coder_per_day) * working_days_in_period * avg_revenue_per_chart_cents | coder_count — reading of healthcare.coding.coder_count; working_days_in_period; avg_revenue_per_chart_cents — assumption; current_charts_per_coder_per_day — reading of healthcare.coding.charts_per_coder_per_day; baseline_charts_per_coder_per_day — baseline of healthcare.coding.charts_per_coder_per_day |
| Coding Error Revenue Recovery; Estimates revenue recovered by reducing coding errors that cause denials, downcoding, or missed HCC captures. Fewer coding errors means more accurate, complete reimbursement. | charts_coded_per_period * (baseline_coding_error_rate_pct - current_coding_error_rate_pct) / 100 * avg_revenue_impact_per_coding_error_cents | charts_coded_per_period — reading of healthcare.coding.charts_coded; current_coding_error_rate_pct — reading of healthcare.coding.error_rate; baseline_coding_error_rate_pct — baseline of healthcare.coding.error_rate; avg_revenue_impact_per_coding_error_cents — assumption |
| Denial Recovery Revenue; Estimates incremental revenue recovered by improving denial appeal success rates. Faster routing and automated appeals documentation means more denied claims are overturned before the appeal deadline. | claims_denied_per_period * avg_claim_value_cents * (current_appeal_success_rate_decimal - baseline_appeal_success_rate_decimal) | avg_claim_value_cents — assumption; claims_denied_per_period — reading of healthcare.claims.denied_count; current_appeal_success_rate_decimal — reading of healthcare.denials.appeal_success_rate; baseline_appeal_success_rate_decimal — baseline of healthcare.denials.appeal_success_rate |
| No-Show Reduction Revenue; Estimates revenue recovered by reducing appointment no-shows. Each prevented no-show allows the slot to be filled by a waiting patient or avoided as wasted clinical capacity. | scheduled_appointments_per_period * (baseline_no_show_rate_pct - current_no_show_rate_pct) / 100 * avg_visit_revenue_cents | avg_visit_revenue_cents — assumption; current_no_show_rate_pct — reading of healthcare.appointments.no_show_rate; baseline_no_show_rate_pct — baseline of healthcare.appointments.no_show_rate; scheduled_appointments_per_period — reading of healthcare.appointments.scheduled_count |
| Outreach Program Labor Savings; Monetizes the care team and administrative time recovered by automating appointment reminders, care gap calls, and chronic disease management touchpoints. | (baseline_outreach_hours_per_week - current_outreach_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_outreach_hours_per_week — reading of workforce.outreach.hours_per_week; baseline_outreach_hours_per_week — baseline of workforce.outreach.hours_per_week |
| Reclaimed Documentation Time — Additional Visit Revenue; Estimates incremental revenue from converting recovered documentation time into additional patient visits. Time recovered becomes visits at the length of a visit: minutes saved per encounter × encounters per day × clinicians × working days ÷ minutes per visit × fraction recaptured as visits × avg visit revenue. | clinician_count * (baseline_documentation_minutes_per_encounter - current_documentation_minutes_per_encounter) * encounters_per_clinician_per_day * working_days_in_period / avg_visit_minutes * time_to_visit_capture_rate_decimal * avg_visit_revenue_cents | clinician_count — reading of healthcare.documentation.clinician_count; avg_visit_minutes — reading of healthcare.appointments.avg_length; working_days_in_period; avg_visit_revenue_cents — assumption; encounters_per_clinician_per_day — reading of healthcare.documentation.encounters_per_clinician_per_day; time_to_visit_capture_rate_decimal — assumption; current_documentation_minutes_per_encounter — reading of healthcare.documentation.minutes_per_encounter; baseline_documentation_minutes_per_encounter — baseline of healthcare.documentation.minutes_per_encounter |
| Value-Based Contract Performance Value; Estimates revenue protected or earned by improving quality measure performance in value-based care contracts. Better HEDIS/MIPS scores translate directly to shared savings, quality bonuses, and avoided penalties. A performance year's revenue at risk × closure-rate improvement × payout sensitivity, × the period's share of the year. | value_based_contract_revenue_at_risk_usd * (current_care_gap_closure_rate_pct - baseline_care_gap_closure_rate_pct) / 100 * quality_score_to_payout_multiplier_decimal * months_in_period / 12 * 100 | months_in_period; current_care_gap_closure_rate_pct — reading of healthcare.quality.care_gap_closure_rate; baseline_care_gap_closure_rate_pct — baseline of healthcare.quality.care_gap_closure_rate; value_based_contract_revenue_at_risk_usd — reading of healthcare.quality.value_based_revenue_at_risk; quality_score_to_payout_multiplier_decimal — assumption |

## Hospitality
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Ancillary Spend Uplift; Estimates incremental ancillary revenue (F&B, spa, upgrades, experiences) from personalized pre-arrival and in-stay offers. Guests × incremental ancillary spend per guest. | guests_in_period * (current_ancillary_spend_per_guest_usd - baseline_ancillary_spend_per_guest_usd) * 100 | guests_in_period — reading of hospitality.guests.count; current_ancillary_spend_per_guest_usd — reading of hospitality.guests.ancillary_spend_per_guest; baseline_ancillary_spend_per_guest_usd — baseline of hospitality.guests.ancillary_spend_per_guest |
| F&B Prep Labor Savings; Monetizes kitchen and prep labor saved when accurate forecasts eliminate over-preparation shifts and reduce last-minute scrambles. Reduction in prep hours × periods × labor rate. | (baseline_prep_hours_per_period - current_prep_hours_per_period) * months_in_period * labor_rate_cents_per_hour | months_in_period; labor_rate_cents_per_hour — assumption; current_prep_hours_per_period — reading of hospitality.food_beverage.prep_hours; baseline_prep_hours_per_period — baseline of hospitality.food_beverage.prep_hours |
| Food Waste Cost Avoidance; Estimates food cost recovered by reducing waste and spoilage through AI-driven demand forecasting and par-level optimization. F&B cost of goods × waste rate reduction. | fb_cost_of_goods_usd * (baseline_waste_pct - current_waste_pct) / 100 * 100 | current_waste_pct — reading of hospitality.food_beverage.waste_share; baseline_waste_pct — baseline of hospitality.food_beverage.waste_share; fb_cost_of_goods_usd — reading of hospitality.food_beverage.cost_of_goods |
| Housekeeping Labor Savings; Estimates labor cost saved by reducing minutes per room turn through AI-optimized scheduling and task dispatch. Rooms turned × minutes saved per turn × labor rate per minute. | rooms_turned_per_period * (baseline_minutes_per_room_turn - current_minutes_per_room_turn) * labor_rate_cents_per_minute | rooms_turned_per_period — reading of hospitality.housekeeping.rooms_turned; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute; current_minutes_per_room_turn — reading of hospitality.housekeeping.minutes_per_room_turn; baseline_minutes_per_room_turn — baseline of hospitality.housekeeping.minutes_per_room_turn |
| OTA Commission Savings; Estimates commissions saved by shifting bookings from OTAs to direct channels through AI-powered conversational booking and personalized offers. Total room revenue × channel shift × OTA commission rate. | total_room_revenue_usd * (current_direct_booking_share_pct - baseline_direct_booking_share_pct) / 100 * avg_ota_commission_rate_decimal * 100 | total_room_revenue_usd — reading of hospitality.rooms.revenue; avg_ota_commission_rate_decimal — assumption; current_direct_booking_share_pct — reading of hospitality.bookings.direct_share; baseline_direct_booking_share_pct — baseline of hospitality.bookings.direct_share |
| Revenue Manager Labor Savings; Monetizes revenue management staff time recovered when AI automates daily rate decisions, comp set monitoring, and reporting — freeing managers for strategy and multi-property oversight. | (baseline_rate_management_hours_per_week - current_rate_management_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_rate_management_hours_per_week — reading of hospitality.revenue_management.hours_per_week; baseline_rate_management_hours_per_week — baseline of hospitality.revenue_management.hours_per_week |
| RevPAR Uplift Revenue; Estimates incremental room revenue from AI-driven rate optimization. Improvement in revenue per available room × total available room nights in the period. | (current_revpar_usd - baseline_revpar_usd) * available_room_nights_per_period * 100 | current_revpar_usd — reading of hospitality.rooms.revpar; baseline_revpar_usd — baseline of hospitality.rooms.revpar; available_room_nights_per_period — reading of hospitality.rooms.available_room_nights |

## Insurance
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Adjudicator Labor Savings; Monetizes adjuster time recovered on auto-adjudicated routine claims — time that can be redeployed to complex, high-severity claims. Claims × adjuster hours saved per claim × hourly rate. The hours are hands-on adjuster work averaged over every claim, a straight-through claim counting as none, so the drop already reflects the share auto-adjudicated; it is not the days from first notice of loss to close. | claims_processed_per_period * (baseline_adjuster_hours_per_claim - current_adjuster_hours_per_claim) * labor_rate_cents_per_hour | labor_rate_cents_per_hour — assumption; claims_processed_per_period — reading of insurance.claims.processed_count; current_adjuster_hours_per_claim — reading of insurance.claims.adjuster_hours_per_claim; baseline_adjuster_hours_per_claim — baseline of insurance.claims.adjuster_hours_per_claim |
| Call Deflection Cost Savings; Estimates call-center cost saved by deflecting routine policy-servicing requests (endorsements, certificates, billing questions) into AI-assisted self-service. Calls deflected per month × months × cost per call. | (baseline_calls_per_month - current_calls_per_month) * months_in_period * avg_cost_per_call_cents | months_in_period; avg_cost_per_call_cents — assumption; current_calls_per_month — reading of insurance.policy_servicing.calls_per_month; baseline_calls_per_month — baseline of insurance.policy_servicing.calls_per_month |
| Claims Fraud Leakage Recovery; Estimates paid-loss reduction from detecting and denying or settling fraudulent claims earlier. Improvement in claims leakage rate × incurred losses. | incurred_losses_usd * (baseline_leakage_pct - current_leakage_pct) / 100 * 100 | current_leakage_pct — reading of insurance.claims.leakage_rate; incurred_losses_usd — reading of insurance.claims.incurred_losses; baseline_leakage_pct — baseline of insurance.claims.leakage_rate |
| Loss Adjustment Expense (LAE) Reduction; Estimates savings from reducing the loss adjustment expense ratio through automated triage and straight-through adjudication. Lower LAE means less adjuster time per claim without increasing error rates. Incurred losses × LAE ratio improvement. | incurred_losses_usd * (baseline_lae_ratio_pct - current_lae_ratio_pct) / 100 * 100 | incurred_losses_usd — reading of insurance.claims.incurred_losses; current_lae_ratio_pct — reading of insurance.claims.lae_ratio; baseline_lae_ratio_pct — baseline of insurance.claims.lae_ratio |
| Loss Ratio Improvement Value; Estimates profit improvement from improving loss ratio precision through better risk selection. Each percentage point improvement in loss ratio flows directly to underwriting profit. Earned premium × loss ratio reduction. | earned_premium_usd * (baseline_loss_ratio_pct - current_loss_ratio_pct) / 100 * 100 | earned_premium_usd — reading of insurance.underwriting.earned_premium; current_loss_ratio_pct — reading of insurance.underwriting.loss_ratio; baseline_loss_ratio_pct — baseline of insurance.underwriting.loss_ratio |
| Service Agent Redeployment Value; Estimates the value of service agent capacity freed by self-service deflection, redeployed to retention, cross-sell, or complex high-value servicing instead of routine transactions. The same deflected calls are what call deflection cost savings prices at their full cost, agent time included, so an initiative gets one of the two. | (baseline_calls_per_month - current_calls_per_month) * months_in_period * avg_handle_time_minutes * labor_rate_cents_per_minute | months_in_period; avg_handle_time_minutes — reading of support.calls.avg_handle_time; current_calls_per_month — reading of insurance.policy_servicing.calls_per_month; baseline_calls_per_month — baseline of insurance.policy_servicing.calls_per_month; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute |
| SIU Capacity Efficiency Value; Estimates the value of SIU investigator time redirected from false-positive referrals to high-quality fraud leads. Fewer wasted investigations × cost per investigation. | (baseline_false_positive_referrals_per_month - current_false_positive_referrals_per_month) * months_in_period * avg_investigation_cost_cents | months_in_period; avg_investigation_cost_cents — assumption; current_false_positive_referrals_per_month — reading of insurance.fraud.false_positive_siu_referrals; baseline_false_positive_referrals_per_month — baseline of insurance.fraud.false_positive_siu_referrals |
| Underwriter Labor Savings; Monetizes underwriter time recovered when AI pre-screens submissions and surfaces data, compressing manual review from days to hours on standard risks. Submissions × hours saved per submission × underwriter rate. | submissions_per_period * (baseline_quote_hours - current_quote_hours) * labor_rate_cents_per_hour | current_quote_hours — reading of insurance.underwriting.quote_hours; baseline_quote_hours — baseline of insurance.underwriting.quote_hours; submissions_per_period — reading of insurance.underwriting.submission_count; labor_rate_cents_per_hour — assumption |

## Manufacturing
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Condition-Based Maintenance Cost Reduction; Estimates maintenance spend saved by servicing equipment based on actual condition rather than fixed time schedules. Condition-based maintenance eliminates unnecessary preventive services and reduces catastrophic failure repair costs. Annual maintenance spend × reduction × the period's share of the year. | total_maintenance_spend_usd * maintenance_cost_reduction_pct / 100 * months_in_period / 12 * 100 | months_in_period; total_maintenance_spend_usd — reading of manufacturing.maintenance.annual_spend; maintenance_cost_reduction_pct — reading of manufacturing.maintenance.cost_reduction_rate |
| Cross-System Data Reconciliation Labor Savings; Monetizes the time recovered by eliminating manual data reconciliation between MES, SCADA, CMMS, and ERP systems. A unified digital thread means no more manual exports, spreadsheet reconciliations, and data-quality firefighting. | (baseline_reconciliation_hours_per_week - current_reconciliation_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_reconciliation_hours_per_week — reading of manufacturing.data.reconciliation_hours_per_week; baseline_reconciliation_hours_per_week — baseline of manufacturing.data.reconciliation_hours_per_week |
| Defect Escape & Warranty Cost Reduction; Estimates the warranty liability and field service cost avoided by catching more defects before they escape to the customer. Fewer escaped defects means fewer warranty claims, field repairs, and recall risks. | units_produced_per_period * (baseline_defect_escape_rate_pct - current_defect_escape_rate_pct) / 100 * avg_warranty_cost_per_escaped_defect_cents | units_produced_per_period — reading of manufacturing.production.units_produced; current_defect_escape_rate_pct — reading of manufacturing.quality.defect_escape_rate; baseline_defect_escape_rate_pct — baseline of manufacturing.quality.defect_escape_rate; avg_warranty_cost_per_escaped_defect_cents — assumption |
| Material Shortage Stoppage Cost Avoidance; Estimates production cost avoided by eliminating material shortage-driven line stoppages. Real-time visibility into inventory and supply status allows planners to act before a shortage becomes a stoppage. | (baseline_shortage_incidents_per_month - current_shortage_incidents_per_month) * months_in_period * avg_stoppage_cost_cents | months_in_period; avg_stoppage_cost_cents — assumption; current_shortage_incidents_per_month — reading of manufacturing.materials.shortage_incidents_per_month; baseline_shortage_incidents_per_month — baseline of manufacturing.materials.shortage_incidents_per_month |
| OEE Data Collection Labor Savings; Monetizes the operations team time recovered by replacing manual OEE data collection, loss categorization, and report building with automated real-time dashboards. | (baseline_oee_reporting_hours_per_week - current_oee_reporting_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_oee_reporting_hours_per_week — reading of manufacturing.equipment.oee_reporting_hours_per_week; baseline_oee_reporting_hours_per_week — baseline of manufacturing.equipment.oee_reporting_hours_per_week |
| OEE Improvement Production Value; Translates OEE percentage point improvements directly into production throughput value. Each point of OEE improvement means more good units produced from the same assets. OEE gain × theoretical capacity × revenue per unit. | (current_oee_pct - baseline_oee_pct) / 100 * theoretical_capacity_units_per_period * avg_revenue_per_unit_cents | current_oee_pct — reading of manufacturing.equipment.oee; baseline_oee_pct — baseline of manufacturing.equipment.oee; avg_revenue_per_unit_cents — assumption; theoretical_capacity_units_per_period — reading of manufacturing.production.theoretical_capacity_units |
| On-Time Delivery Improvement Revenue Value; Estimates revenue protected or unlocked by improving on-time delivery rates. Late deliveries create customer penalties, order cancellations, and lost future business. OTD improvement × revenue at risk. | (current_otd_rate_pct - baseline_otd_rate_pct) / 100 * revenue_at_risk_from_late_delivery_usd * 100 | current_otd_rate_pct — reading of supply_chain.deliveries.on_time_rate; baseline_otd_rate_pct — baseline of supply_chain.deliveries.on_time_rate; revenue_at_risk_from_late_delivery_usd — reading of supply_chain.deliveries.revenue_at_risk |
| Production Scheduling Labor Savings; Monetizes the planning and operations team time recovered by replacing manual schedule building with AI optimization. Hours saved per week × weeks × planner rate. | (baseline_scheduling_hours_per_week - current_scheduling_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_scheduling_hours_per_week — reading of manufacturing.scheduling.hours_per_week; baseline_scheduling_hours_per_week — baseline of manufacturing.scheduling.hours_per_week |
| Scrap & Rework Cost Reduction; Measures the direct reduction in material scrap and rework labor costs from catching defects earlier in the production process. Real-time detection prevents defective work-in-progress from accumulating before the defect is caught. | (baseline_scrap_cost_per_month_usd - current_scrap_cost_per_month_usd) * months_in_period * 100 | months_in_period; current_scrap_cost_per_month_usd — reading of manufacturing.quality.scrap_cost_per_month; baseline_scrap_cost_per_month_usd — baseline of manufacturing.quality.scrap_cost_per_month |
| Unplanned Downtime Cost Avoidance; Estimates the production value recovered by preventing unplanned equipment failures. Each avoided downtime hour represents lost throughput, labor idled, and expedite costs to recover. Hours avoided × hourly production loss value. | (baseline_downtime_hours_per_month - current_downtime_hours_per_month) * months_in_period * hourly_production_loss_cents | months_in_period; hourly_production_loss_cents — assumption; current_downtime_hours_per_month — reading of manufacturing.equipment.unplanned_downtime_hours_per_month; baseline_downtime_hours_per_month — baseline of manufacturing.equipment.unplanned_downtime_hours_per_month |

## Media entertainment
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Churn Reduction Revenue; Estimates subscription revenue protected by reducing monthly churn through better recommendations and engagement. More relevant content keeps subscribers from canceling. Subscribers × churn reduction × months × ARPU. | subscriber_count * (baseline_monthly_churn_rate_pct - current_monthly_churn_rate_pct) / 100 * months_in_period * avg_revenue_per_subscriber_per_month_cents | months_in_period; subscriber_count — reading of billing.subscriptions.active_count; billing_subscriptions_mrr_usd — reading of billing.subscriptions.mrr; current_monthly_churn_rate_pct — reading of billing.subscriptions.churn_rate; baseline_monthly_churn_rate_pct — baseline of billing.subscriptions.churn_rate; avg_revenue_per_subscriber_per_month_cents |
| Engagement-Driven Advertising Revenue Uplift; Estimates incremental advertising revenue from increased watch time through better recommendations. More engaged users watch more ad-supported content and command higher CPMs. | avg_active_users * (current_watch_time_hours_per_user - baseline_watch_time_hours_per_user) * ad_revenue_per_hour_watched_cents | avg_active_users — reading of media.audience.avg_active_users; ad_revenue_per_hour_watched_cents — assumption; current_watch_time_hours_per_user — reading of media.audience.monthly_watch_hours_per_user; baseline_watch_time_hours_per_user — baseline of media.audience.monthly_watch_hours_per_user |
| Localization Cost Reduction; Estimates savings from AI-assisted dubbing, subtitling, and localization quality review, reducing vendor cost per episode per language while unlocking additional markets. | episodes_localised_per_period * (baseline_localisation_cost_per_episode_usd - current_localisation_cost_per_episode_usd) * 100 | episodes_localised_per_period — reading of media.localization.episodes_localized; current_localisation_cost_per_episode_usd — reading of media.localization.cost_per_episode; baseline_localisation_cost_per_episode_usd — baseline of media.localization.cost_per_episode |
| Post-Production Cost Savings; Estimates post-production cost saved by compressing timelines through AI-assisted editing, VFX clean-up, and transcription. Episodes × days saved per episode × daily crew and facility cost. | episodes_per_period * (baseline_post_days_per_episode - current_post_days_per_episode) * avg_daily_post_cost_cents | episodes_per_period — reading of media.post_production.episode_count; avg_daily_post_cost_cents — assumption; current_post_days_per_episode — reading of media.post_production.days_per_episode; baseline_post_days_per_episode — baseline of media.post_production.days_per_episode |
| Rights Administration Labor Savings; Monetizes the rights, legal, and business-affairs team time recovered by automating metadata enrichment, rights ingestion, and catalog reconciliation. | (baseline_rights_admin_hours_per_week - current_rights_admin_hours_per_week) * weeks_in_period * labor_rate_cents_per_hour | weeks_in_period; labor_rate_cents_per_hour — assumption; current_rights_admin_hours_per_week — reading of media.rights.admin_hours_per_week; baseline_rights_admin_hours_per_week — baseline of media.rights.admin_hours_per_week |
| Royalty Dispute Cost Reduction; Estimates cost saved by reducing royalty statement disputes through AI-verified accurate rights data and automated calculations. Fewer disputes mean less legal and finance time spent on corrections. | royalty_statements_per_period * (baseline_dispute_rate_pct - current_dispute_rate_pct) / 100 * avg_dispute_resolution_cost_cents | current_dispute_rate_pct — reading of media.royalties.dispute_rate; baseline_dispute_rate_pct — baseline of media.royalties.dispute_rate; royalty_statements_per_period — reading of media.royalties.statement_count; avg_dispute_resolution_cost_cents — assumption |

## Professional services
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Bench Cost Reduction; Estimates the cost saved by reducing unassigned bench time. Practitioners on bench represent fully-loaded salary cost with no revenue offset. Each percentage point of bench reduction converts idle cost into deployable capacity. | practitioner_count * available_hours_per_practitioner_per_period * (baseline_bench_time_pct - current_bench_time_pct) / 100 * fully_loaded_cost_cents_per_hour | practitioner_count — reading of professional_services.practitioners.billable_count; current_bench_time_pct — reading of professional_services.utilization.bench_share; baseline_bench_time_pct — baseline of professional_services.utilization.bench_share; fully_loaded_cost_cents_per_hour — assumption; available_hours_per_practitioner_per_period — reading of professional_services.practitioners.available_hours |
| Billable Utilization Revenue Uplift; Estimates incremental revenue from improving billable utilization rates across the practitioner base. Each percentage point of utilization improvement converts previously undeployed capacity into billed hours. Total practitioners × available hours × utilization improvement × blended bill rate. | practitioner_count * available_hours_per_practitioner_per_period * (current_billable_utilization_rate_pct - baseline_billable_utilization_rate_pct) / 100 * avg_bill_rate_cents_per_hour | practitioner_count — reading of professional_services.practitioners.billable_count; avg_bill_rate_cents_per_hour — assumption; current_billable_utilization_rate_pct — reading of professional_services.utilization.billable_rate; baseline_billable_utilization_rate_pct — baseline of professional_services.utilization.billable_rate; available_hours_per_practitioner_per_period — reading of professional_services.practitioners.available_hours |
| Budget Overrun Cost Avoidance; Estimates the margin value recovered by catching at-risk projects earlier and avoiding budget overruns. Each percentage point reduction in overrun rate represents real margin that would otherwise be absorbed as write-offs or discounts. An overrun is incurred as the work is done, so the base is the period's revenue, not the contract value of every active project, which would count a long engagement again each month it ran: revenue × fall in the overrun rate × average overrun depth. | financials_revenue_usd * (baseline_overrun_rate_pct - current_overrun_rate_pct) / 100 * avg_overrun_depth_decimal * 100 | financials_revenue_usd — reading of finance.pnl.revenue; current_overrun_rate_pct — reading of professional_services.projects.budget_overrun_rate; avg_overrun_depth_decimal — assumption; baseline_overrun_rate_pct — baseline of professional_services.projects.budget_overrun_rate |
| Early Intervention — PM Escalation Labor Savings; Estimates the project management time saved when AI flags at-risk engagements weeks earlier, allowing lightweight interventions instead of costly late-stage crisis management. Earlier detection means fewer all-hands escalation events and less senior time spent on recovery. | at_risk_projects_per_period * (baseline_pm_recovery_hours - current_pm_recovery_hours) * labor_rate_cents_per_hour | current_pm_recovery_hours — reading of professional_services.projects.recovery_hours_per_project; labor_rate_cents_per_hour — assumption; baseline_pm_recovery_hours — baseline of professional_services.projects.recovery_hours_per_project; at_risk_projects_per_period — reading of professional_services.projects.at_risk_identified_count |
| Knowledge Search Labor Savings; Monetizes practitioner time recovered by replacing manual search across shared drives, email, and intranets with AI-powered knowledge retrieval. Hours saved per task × tasks per period × blended practitioner rate. | practitioners_using_system * knowledge_search_tasks_per_practitioner_per_period * (baseline_search_minutes - current_search_minutes) / 60 * labor_rate_cents_per_hour | current_search_minutes — reading of professional_services.knowledge.search_minutes; baseline_search_minutes — baseline of professional_services.knowledge.search_minutes; labor_rate_cents_per_hour — assumption; practitioners_using_system — reading of professional_services.knowledge.active_users; knowledge_search_tasks_per_practitioner_per_period — reading of professional_services.knowledge.searches_per_practitioner |
| PM Capacity Expansion Value; Estimates the revenue value of PM capacity freed by automating client reporting. Each PM hour recovered from report assembly can be redeployed to client relationship work or managing additional project scope — increasing the portfolio each PM can carry without adding headcount. | pm_count * (baseline_hours_per_report - current_hours_per_report) * reports_per_pm_per_period * avg_pm_billing_rate_cents_per_hour * redeployment_rate_decimal | pm_count — reading of professional_services.projects.pm_count_using_automation; current_hours_per_report — reading of professional_services.reporting.hours_per_client_report; baseline_hours_per_report — baseline of professional_services.reporting.hours_per_client_report; redeployment_rate_decimal — assumption; reports_per_pm_per_period — reading of professional_services.reporting.reports_per_pm; avg_pm_billing_rate_cents_per_hour — assumption |
| Proposal Assembly Labor Savings; Estimates the practitioner time saved on proposal and SOW development by surfacing relevant prior proposals, case studies, and rate cards instantly. Fewer hours per proposal × proposals per period × blended staff rate. | proposals_per_period * (baseline_proposal_prep_hours - current_proposal_prep_hours) * labor_rate_cents_per_hour | proposals_per_period — reading of professional_services.proposals.submitted_count; labor_rate_cents_per_hour — assumption; current_proposal_prep_hours — reading of professional_services.proposals.prep_hours; baseline_proposal_prep_hours — baseline of professional_services.proposals.prep_hours |
| Realization Rate Improvement Revenue; Estimates revenue from improving the realization rate — the fraction of worked hours actually billed and collected. Better time capture and tighter write-off controls convert more worked hours into invoiced and collected revenue. Total worked hours × realization improvement × blended bill rate. | total_practitioner_hours_worked * (current_realization_rate_pct - baseline_realization_rate_pct) / 100 * avg_bill_rate_cents_per_hour | avg_bill_rate_cents_per_hour — assumption; current_realization_rate_pct — reading of professional_services.billing.realization_rate; baseline_realization_rate_pct — baseline of professional_services.billing.realization_rate; total_practitioner_hours_worked — reading of professional_services.practitioners.billable_hours_worked |
| Recovered Billable Hours Revenue; Estimates revenue from billable hours previously lost to under-capture. AI time suggestions recover hours that practitioners forgot to log or intentionally wrote off due to friction in the timekeeping system. Practitioners × weekly hours recovered × weeks × bill rate. | practitioner_count * recovered_billable_hours_per_practitioner_per_week * weeks_in_period * avg_bill_rate_cents_per_hour | weeks_in_period; practitioner_count — reading of professional_services.practitioners.billable_count; avg_bill_rate_cents_per_hour — assumption; recovered_billable_hours_per_practitioner_per_week — reading of professional_services.billing.unbilled_hours_per_practitioner_weekly |
| Report Assembly Labor Savings; Monetizes project manager and analyst time recovered by automating the data gathering, formatting, and delivery of client status reports. Hours saved per report × reports per period × blended PM rate. | reports_per_period * (baseline_hours_per_report - current_hours_per_report) * labor_rate_cents_per_hour | reports_per_period — reading of professional_services.reporting.client_reports_delivered; current_hours_per_report — reading of professional_services.reporting.hours_per_client_report; baseline_hours_per_report — baseline of professional_services.reporting.hours_per_client_report; labor_rate_cents_per_hour — assumption |

## Real estate property
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Capital Allocation Improvement Value; Estimates the return value of making better capital allocation decisions with real-time portfolio intelligence. Portfolio-wide scenario modeling enables earlier identification of underperforming assets and more optimal CapEx sequencing — improving overall portfolio yield. AUM × the improvement in annual yield × the period's share of the year. | total_aum_usd * capital_allocation_improvement_bps / 10000 * months_in_period / 12 * 100 | total_aum_usd — reading of real_estate.portfolio.aum; months_in_period; capital_allocation_improvement_bps — assumption |
| Emergency Repair Cost Avoidance; Estimates the repair cost avoided by catching equipment failures early. Emergency repairs from deferred or undetected building system failures cost 2–5× more than planned interventions — and create tenant satisfaction impacts that drive turnover. | (baseline_emergency_incidents_per_month - current_emergency_incidents_per_month) * months_in_period * avg_emergency_repair_cost_cents | months_in_period; avg_emergency_repair_cost_cents — assumption; current_emergency_incidents_per_month — reading of real_estate.maintenance.emergency_repairs_per_month; baseline_emergency_incidents_per_month — baseline of real_estate.maintenance.emergency_repairs_per_month |
| Financial Reporting Labor Savings; Monetizes accounting and asset management staff time recovered by replacing manual spreadsheet assembly with automated property-level financial reporting. Hours saved per property per month × property count × months × blended accounting rate. | property_count * (baseline_reporting_hours_per_property - current_reporting_hours_per_property) * months_in_period * labor_rate_cents_per_hour | property_count — reading of real_estate.portfolio.property_count; months_in_period; labor_rate_cents_per_hour — assumption; current_reporting_hours_per_property — reading of real_estate.reporting.hours_per_property_monthly; baseline_reporting_hours_per_property — baseline of real_estate.reporting.hours_per_property_monthly |
| Investor Reporting Cycle Compression Value; Estimates the capital cost value of delivering investor reports faster. Faster reporting shortens the investor decision cycle for capital calls and distributions, and reduces management time on ad-hoc data requests from investors who are waiting on delayed reports. The reporting cycles in the period (a year's cycles spread evenly over its months) × lag days saved × cost per lag day. | investor_report_cycles_per_year * months_in_period / 12 * (baseline_delivery_lag_days - current_delivery_lag_days) * daily_investor_inquiry_cost_cents | months_in_period; current_delivery_lag_days — reading of operations.reporting.delivery_lag_days; baseline_delivery_lag_days — baseline of operations.reporting.delivery_lag_days; investor_report_cycles_per_year — assumption; daily_investor_inquiry_cost_cents — assumption |
| Leasing Staff Labor Savings; Monetizes leasing staff time recovered when AI handles initial lead response, qualification screening, and tour scheduling. Fewer manual touchpoints per lease allows the same team to manage higher lead volume or handle more units. | units_leased_per_period * (baseline_staff_hours_per_lease - current_staff_hours_per_lease) * labor_rate_cents_per_hour | units_leased_per_period — reading of real_estate.leasing.units_leased; labor_rate_cents_per_hour — assumption; current_staff_hours_per_lease — reading of real_estate.leasing.staff_hours_per_lease; baseline_staff_hours_per_lease — baseline of real_estate.leasing.staff_hours_per_lease |
| Maintenance Request Self-Serve Deflection Savings; Estimates staff time saved when tenants self-serve common maintenance requests and status checks through the tenant app rather than calling the management office. Fewer inbound calls and emails means less property management staff time per unit. | units_in_portfolio * (baseline_staff_contacts_per_unit_per_month - current_staff_contacts_per_unit_per_month) * months_in_period * avg_contact_handling_minutes / 60 * labor_rate_cents_per_hour | months_in_period; units_in_portfolio — reading of real_estate.portfolio.unit_count; labor_rate_cents_per_hour — assumption; avg_contact_handling_minutes — reading of real_estate.tenants.minutes_per_contact; current_staff_contacts_per_unit_per_month — reading of real_estate.tenants.contacts_per_unit_monthly; baseline_staff_contacts_per_unit_per_month — baseline of real_estate.tenants.contacts_per_unit_monthly |
| Portfolio Reporting Labor Savings; Monetizes asset management and finance team time recovered by replacing manual cross-property data aggregation with automated portfolio dashboards. Hours saved per property per month × portfolio size × months × blended analyst rate. | properties_in_portfolio * (baseline_reporting_hours_per_property_per_month - current_reporting_hours_per_property_per_month) * months_in_period * labor_rate_cents_per_hour | months_in_period; properties_in_portfolio — reading of real_estate.portfolio.property_count; labor_rate_cents_per_hour — assumption; current_reporting_hours_per_property_per_month — reading of real_estate.reporting.hours_per_property_monthly; baseline_reporting_hours_per_property_per_month — baseline of real_estate.reporting.hours_per_property_monthly |
| Tenant Turnover Cost Avoidance; Estimates the cost avoided by improving tenant renewal rates. Each prevented turnover avoids vacancy loss, re-leasing commissions, unit make-ready costs, and the time-to-lease gap. Improved renewal rate × units up for renewal × total turnover cost per unit. | units_up_for_renewal * (current_renewal_rate_pct - baseline_renewal_rate_pct) / 100 * avg_turnover_cost_cents | units_up_for_renewal — reading of real_estate.leasing.units_up_for_renewal; avg_turnover_cost_cents — assumption; current_renewal_rate_pct — reading of real_estate.leasing.annual_renewal_rate; baseline_renewal_rate_pct — baseline of real_estate.leasing.annual_renewal_rate |
| Vacancy Day Reduction Revenue; Estimates incremental revenue from compressing time-to-lease. Every day a unit sits vacant is lost rent. Faster lead response and self-tour scheduling convert more leads before they lease elsewhere — directly reducing average vacancy days per unit. | units_leased_per_period * (baseline_time_to_lease_days - current_time_to_lease_days) * avg_daily_rent_cents | avg_daily_rent_cents — assumption; units_leased_per_period — reading of real_estate.leasing.units_leased; current_time_to_lease_days — reading of real_estate.leasing.days_to_lease; baseline_time_to_lease_days — baseline of real_estate.leasing.days_to_lease |
| Work Order Administration Labor Savings; Monetizes property management staff time recovered by automating work order creation, vendor dispatch, and status tracking. Fewer manual coordination touchpoints per work order × volume × blended property management rate. | work_orders_per_period * (baseline_admin_minutes_per_work_order - current_admin_minutes_per_work_order) / 60 * labor_rate_cents_per_hour | work_orders_per_period — reading of real_estate.work_orders.created_count; labor_rate_cents_per_hour — assumption; current_admin_minutes_per_work_order — reading of real_estate.work_orders.admin_minutes_per_order; baseline_admin_minutes_per_work_order — baseline of real_estate.work_orders.admin_minutes_per_order |

## Retail ecommerce
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Average Order Value Uplift Revenue; Estimates incremental revenue from increasing average order value through AI-driven product recommendations (cross-sell, upsell, complementary items). Orders in period × AOV increase. | orders_per_period * (current_avg_order_value_usd - baseline_avg_order_value_usd) * 100 | orders_per_period — reading of retail.orders.count; current_avg_order_value_usd — reading of retail.orders.avg_value; baseline_avg_order_value_usd — baseline of retail.orders.avg_value |
| Conversion Rate Uplift Revenue; Estimates incremental revenue from improving on-site conversion through personalized experiences. More relevant product presentation means more sessions convert to purchases. Sessions × conversion rate improvement × average order value. | sessions_per_period * (current_conversion_rate_pct - baseline_conversion_rate_pct) / 100 * avg_order_value_usd * 100 | avg_order_value_usd — reading of retail.orders.avg_value; sessions_per_period — reading of retail.web.session_count; current_conversion_rate_pct — reading of retail.web.conversion_rate; baseline_conversion_rate_pct — baseline of retail.web.conversion_rate |
| Customer Retention Revenue Value; Estimates the revenue impact of improving customer retention. Each percentage point of retention improvement keeps more customers buying; the value is the period's share of a retained customer's annual revenue, one month of it for a month. | customer_base * (current_retention_rate_pct - baseline_retention_rate_pct) / 100 * avg_annual_revenue_per_customer_cents * months_in_period / 12 | customer_base — reading of retail.customers.active_count; months_in_period; current_retention_rate_pct — reading of retail.customers.retention_rate; baseline_retention_rate_pct — baseline of retail.customers.retention_rate; avg_annual_revenue_per_customer_cents — assumption |
| Gross Margin Improvement Value; Estimates the revenue value of gross margin percentage improvement from AI-optimized pricing and markdown timing. Demand-responsive pricing captures more value at full price and limits margin sacrifice on clearance. | total_revenue_cents * (current_gross_margin_pct - baseline_gross_margin_pct) / 100 | total_revenue_cents; financials_revenue_usd — reading of finance.pnl.revenue; current_gross_margin_pct — reading of retail.pricing.gross_margin; baseline_gross_margin_pct — baseline of retail.pricing.gross_margin |
| Inventory Accuracy & Shrink Reduction; Estimates the cost avoided by improving inventory record accuracy. Higher accuracy means fewer phantom inventory situations, less shrink write-off, and better purchasing decisions based on true on-hand levels. The shrink rate is a share of sales, so it is applied to the period's sales: revenue × accuracy improvement × shrink rate. | financials_revenue_usd * (current_inventory_accuracy_pct - baseline_inventory_accuracy_pct) / 100 * shrink_cost_rate_decimal * 100 | financials_revenue_usd — reading of finance.pnl.revenue; shrink_cost_rate_decimal — assumption; current_inventory_accuracy_pct — reading of retail.inventory.accuracy; baseline_inventory_accuracy_pct — baseline of retail.inventory.accuracy |
| Markdown & Overstock Cost Reduction; Estimates gross margin saved by reducing the volume and depth of markdowns. Better demand forecasts mean less overbuying, fewer full-clearance markdowns, and a higher share of sales made at full price. The markdown rate is a share of what sold, so it is applied to the period's sales, not to the stock on hand: revenue × fall in the markdown rate × average markdown depth. Revenue is already net of markdowns, so the figure errs low. | financials_revenue_usd * (baseline_markdown_rate_pct - current_markdown_rate_pct) / 100 * avg_markdown_depth_decimal * 100 | financials_revenue_usd — reading of finance.pnl.revenue; current_markdown_rate_pct — reading of retail.pricing.markdown_rate; avg_markdown_depth_decimal — assumption; baseline_markdown_rate_pct — baseline of retail.pricing.markdown_rate |
| Omnichannel Stockout Revenue Recovery; Estimates revenue recovered by eliminating stockouts across all channels. Unified inventory visibility allows ship-from-store and accurate ATP promises — converting what was previously a lost sale into a fulfilled order. | total_revenue_cents * (baseline_stockout_rate_pct - current_stockout_rate_pct) / 100 | total_revenue_cents; financials_revenue_usd — reading of finance.pnl.revenue; current_stockout_rate_pct — reading of supply_chain.inventory.stockout_rate; baseline_stockout_rate_pct — baseline of supply_chain.inventory.stockout_rate |
| Repeat Purchase Revenue Uplift; Estimates incremental revenue from improving the repeat purchase rate through CLV-optimized lifecycle marketing. More customers returning for a second purchase directly increases revenue without additional acquisition spend. The repeat rate is over twelve months, so the added second purchases are a year's, spread evenly over its months. | customer_base * (current_repeat_purchase_rate_pct - baseline_repeat_purchase_rate_pct) / 100 * months_in_period / 12 * avg_second_purchase_value_cents | customer_base — reading of retail.customers.active_count; months_in_period; avg_second_purchase_value_cents — assumption; current_repeat_purchase_rate_pct — reading of retail.customers.repeat_purchase_rate; baseline_repeat_purchase_rate_pct — baseline of retail.customers.repeat_purchase_rate |
| Sell-Through Improvement Value; Estimates the revenue recovered by selling more seasonal stock at its target price instead of clearing it at a markdown. Of the markdown-eligible stock that sold or cleared in the period, the share that sold at target rose; each of those sales kept the markdown it would otherwise have given up. Retail value sold or cleared × sell-through gain × average markdown depth. Gross margin improvement on the same template counts this too, so an initiative gets one of the two. | markdown_eligible_sold_value_usd * (current_sell_through_rate_pct - baseline_sell_through_rate_pct) / 100 * avg_markdown_depth_decimal * 100 | avg_markdown_depth_decimal — assumption; current_sell_through_rate_pct — reading of retail.inventory.sell_through_rate; baseline_sell_through_rate_pct — baseline of retail.inventory.sell_through_rate; markdown_eligible_sold_value_usd — reading of retail.inventory.markdown_eligible_sold_value |
| Stockout Lost Revenue Recovery; Estimates revenue recovered by reducing in-season stockouts. Each percentage point of stockout reduction means fewer lost sales from customers who found the product unavailable. | total_revenue_cents * (baseline_stockout_rate_pct - current_stockout_rate_pct) / 100 | total_revenue_cents; financials_revenue_usd — reading of finance.pnl.revenue; current_stockout_rate_pct — reading of supply_chain.inventory.stockout_rate; baseline_stockout_rate_pct — baseline of supply_chain.inventory.stockout_rate |

## Telecom
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Call Deflection Labor Savings; Estimates agent labor saved by deflecting routine billing, account, and technical inquiries into self-service. Calls deflected per month × months × average handle time × agent rate. Care cost per subscriber on the same template already falls by the agent time deflection saves, so an initiative gets one of the two. | (baseline_calls_per_month - current_calls_per_month) * months_in_period * avg_handle_time_minutes * labor_rate_cents_per_minute | months_in_period; avg_handle_time_minutes — reading of support.calls.avg_handle_time; current_calls_per_month — reading of telecom.customer_care.calls_per_month; baseline_calls_per_month — baseline of telecom.customer_care.calls_per_month; labor_rate_cents_per_hour — assumption; labor_rate_cents_per_minute |
| Care Cost Per Subscriber Reduction; Estimates the total care cost saved by reducing per-subscriber care cost through deflection bots and AI agent assist. Subscriber base × cost reduction per sub × measurement periods. | subscriber_count * (baseline_care_cost_per_sub_usd - current_care_cost_per_sub_usd) * months_in_period * 100 | months_in_period; subscriber_count — reading of billing.subscriptions.active_count; current_care_cost_per_sub_usd — reading of telecom.customer_care.cost_per_subscriber; baseline_care_cost_per_sub_usd — baseline of telecom.customer_care.cost_per_subscriber |
| Churn Prevention Revenue; Estimates ARPU revenue protected by reducing monthly churn through predictive save plays. Subscribers retained × months × ARPU. | subscriber_count * (baseline_monthly_churn_rate_pct - current_monthly_churn_rate_pct) / 100 * months_in_period * avg_arpu_cents_per_month | months_in_period; subscriber_count — reading of billing.subscriptions.active_count; avg_arpu_cents_per_month; billing_subscriptions_mrr_usd — reading of billing.subscriptions.mrr; current_monthly_churn_rate_pct — reading of billing.subscriptions.churn_rate; baseline_monthly_churn_rate_pct — baseline of billing.subscriptions.churn_rate |
| MTTR Reduction Cost Avoidance; Estimates the operational cost avoided by resolving network incidents faster. Shorter MTTR means fewer NOC engineer hours per incident, lower overtime, and reduced customer SLA credit exposure. Incidents × MTTR reduction × cost per hour of incident. | incidents_per_period * (baseline_mttr_minutes - current_mttr_minutes) / 60 * cost_per_incident_hour_cents | current_mttr_minutes — reading of telecom.network.mttr_minutes; incidents_per_period — reading of telecom.network.incident_count; baseline_mttr_minutes — baseline of telecom.network.mttr_minutes; cost_per_incident_hour_cents — assumption |
| Save Play Cost Efficiency; Estimates the net value of targeted save offers versus blanket discounting — fewer wasted retention offers on low-risk subscribers reduces discount cost while protecting revenue on high-risk, high-ARPU targets. | (baseline_save_offer_cost_per_retained_subscriber_usd - current_save_offer_cost_per_retained_subscriber_usd) * retained_subscribers_per_period * 100 | retained_subscribers_per_period — reading of telecom.retention.retained_subscribers; current_save_offer_cost_per_retained_subscriber_usd — reading of telecom.retention.cost_per_retained_subscriber; baseline_save_offer_cost_per_retained_subscriber_usd — baseline of telecom.retention.cost_per_retained_subscriber |
| Truck Roll Reduction Savings; Estimates field dispatch cost avoided when predictive network maintenance and remote diagnostics eliminate unnecessary truck rolls. Fewer truck rolls × cost per dispatch. | (baseline_truck_rolls_per_month - current_truck_rolls_per_month) * months_in_period * avg_truck_roll_cost_cents | months_in_period; avg_truck_roll_cost_cents — assumption; current_truck_rolls_per_month — reading of telecom.field_service.truck_rolls_per_month; baseline_truck_rolls_per_month — baseline of telecom.field_service.truck_rolls_per_month |

## Transportation logistics
| Formula | Arithmetic | Inputs |
| --- | --- | --- |
| Breakdown Cost Avoidance; Estimates cost avoided by preventing unplanned roadside breakdowns. Each avoided event eliminates towing, roadside labor, missed-load penalties, and expedite recovery costs. Events avoided a year, spread evenly over its months, × average all-in breakdown cost. | (baseline_roadside_events_per_year - current_roadside_events_per_year) * months_in_period / 12 * avg_breakdown_cost_cents | months_in_period; avg_breakdown_cost_cents — assumption; current_roadside_events_per_year — reading of logistics.fleet.roadside_events_per_year; baseline_roadside_events_per_year — baseline of logistics.fleet.roadside_events_per_year |
| Driver Productivity Value; Estimates incremental revenue or cost capacity unlocked by improving stops per driver hour. More stops per hour means the same drivers can handle more volume without adding headcount. Added stops per driver hour × the fleet's driver hours in the period × revenue per stop. | (current_stops_per_driver_hour - baseline_stops_per_driver_hour) * operating_hours_per_period * revenue_per_stop_cents | revenue_per_stop_cents — assumption; operating_hours_per_period — reading of logistics.drivers.operating_hours; current_stops_per_driver_hour — reading of logistics.drivers.stops_per_hour; baseline_stops_per_driver_hour — baseline of logistics.drivers.stops_per_hour |
| Fuel & Mile Reduction Savings; Estimates cost saved by reducing total fleet miles through AI route optimization. Fewer miles directly lower fuel, tire, and per-mile maintenance spend. Miles eliminated × blended cost per mile. | (baseline_fleet_miles - current_fleet_miles) * avg_cost_cents_per_mile | current_fleet_miles — reading of logistics.fleet.miles; baseline_fleet_miles — baseline of logistics.fleet.miles; avg_cost_cents_per_mile — assumption |
| Insurance Premium Reduction; Estimates the reduction in commercial auto and cargo insurance premiums driven by a demonstrated improvement in loss history and safety program maturity. Annual premium × expected reduction × the period's share of the year. | annual_insurance_premium_usd * premium_reduction_pct / 100 * months_in_period / 12 * 100 | months_in_period; premium_reduction_pct — assumption; annual_insurance_premium_usd — reading of logistics.fleet.annual_insurance_premium |
| Maintenance Labor Savings; Monetizes shop and maintenance team time recovered when condition-based scheduling replaces reactive and over-scheduled preventive services, reducing unnecessary labor hours. | (baseline_maintenance_hours_per_month - current_maintenance_hours_per_month) * months_in_period * labor_rate_cents_per_hour | months_in_period; labor_rate_cents_per_hour — assumption; current_maintenance_hours_per_month — reading of logistics.fleet.maintenance_hours_per_month; baseline_maintenance_hours_per_month — baseline of logistics.fleet.maintenance_hours_per_month |
| On-Time Delivery Revenue Protection; Estimates revenue protected or penalties avoided by improving on-time delivery rates. Late deliveries cause customer chargebacks, contract penalties, and order cancellations. OTD improvement × revenue at risk. | (current_on_time_delivery_pct - baseline_on_time_delivery_pct) / 100 * revenue_at_risk_from_late_delivery_usd * 100 | current_on_time_delivery_pct — reading of logistics.deliveries.on_time_rate; baseline_on_time_delivery_pct — baseline of logistics.deliveries.on_time_rate; revenue_at_risk_from_late_delivery_usd — reading of supply_chain.deliveries.revenue_at_risk |
| Preventable Accident Cost Avoidance; Estimates the all-in cost avoided by reducing preventable accident frequency. Each avoided accident eliminates vehicle repair, cargo loss, injury liability, downtime, and insurance claim impact. Accidents avoided × average cost per accident. | miles_driven_per_period / 1000000 * (baseline_accidents_per_million_miles - current_accidents_per_million_miles) * avg_preventable_accident_cost_cents | miles_driven_per_period — reading of logistics.fleet.miles; avg_preventable_accident_cost_cents — assumption; current_accidents_per_million_miles — reading of logistics.safety.preventable_accidents_per_million_miles; baseline_accidents_per_million_miles — baseline of logistics.safety.preventable_accidents_per_million_miles |
| WISMO Contact Cost Reduction; Estimates customer service cost saved by reducing 'Where Is My Order' contacts through accurate real-time ETAs and proactive delivery status notifications. | (baseline_wismo_contacts_per_month - current_wismo_contacts_per_month) * months_in_period * avg_cost_per_contact_cents | months_in_period; avg_cost_per_contact_cents — assumption; current_wismo_contacts_per_month — reading of logistics.customer_service.wismo_contacts_per_month; baseline_wismo_contacts_per_month — baseline of logistics.customer_service.wismo_contacts_per_month |