Build an Executive Business Case

Steps

1. 1

   Define the decision, alternatives and investment boundary

   Translate the vague ask into a decision with named sponsor, time horizon, required return, hard constraints and alternatives including status quo. Frame 3-5 assertions that must hold for investment to make sense: demand, outcomes, cost, adoption, feasibility and timing. Distinguish who recommends from who can authorize spending. Pull applicable baseline and commercial context from authorized data; if unavailable mark estimates rather than manufacturing figures. Gate: a decision maker can see the actual choice and assumptions that would change it.

   V2 practical protocol:
   Decision protocol: Define the decision in the form "Should we invest [amount/range] in [intervention] for [population] over [horizon]?" List status quo and at least one feasible alternative. Record decision maker, currency, evaluation horizon, discount rate policy (if NPV will be used), spending authority, hard constraints and what evidence would reverse the recommendation. Distinguish a financial target imposed by the sponsor from a threshold suggested by the analyst. Minimum output: decision table with option, decision owner, success metric, horizon, constraint, evidence needed, confidence and unresolved approval.

   Expected outputs
2. 2

   Validate outcome drivers and the counterfactual

   Define the operational mechanism connecting the proposal to revenue, cost, risk reduction or customer impact. Establish current baseline, affected population, usage/adoption rates, time to benefit and counterfactual outcome if nothing changes. Separate observed effects from forecasts and double-counted benefits. Use interviews or real examples to challenge the optimistic case, noting implementation and organizational obstacles. Gate: each claimed benefit has a measurable driver, assumption and evidence grade.

   V2 practical protocol:
   Benefit model: Calculate volumes and rates with consistent time units. Example: capacity hours released/year = eligible transactions/year × minutes saved/transaction ÷ 60 × realistic adoption × realized-efficiency factor. Only count those hours as cash savings if budgeted hours or spend can actually be removed; otherwise label as capacity released. For incremental revenue, calculate contribution after variable cost, not gross booked revenue. For risk reduction, expected avoided loss = exposure × baseline probability × plausible change in probability, with uncertainty explicit. Separate one-time and recurring benefits, gross versus incremental, and overlap between benefits. Use a driver ledger with baseline, counterfactual, numerator/denominator, formula, unit, period, source date, estimate owner and confidence. If no empirical baseline exists, show what to measure before quoting a benefit.

   Expected outputs
3. 3

   Model full economics and downside exposure

   Build an auditable scenario model with setup, labor, integration, licensing, change management, support and ongoing operating costs. Include timing of cash flows and appropriate financial measures (payback, NPV or ROI) only when inputs and discount assumptions permit. Show base, upside and downside scenarios; sensitivity-test 2-3 dominant inputs. Explain uncertainty ranges, opportunity cost and any non-financial constraints. Do not treat a spreadsheet-generated number as certainty. Gate: financial claims reconcile to input assumptions and downside remains visible.

   V2 practical protocol:
   Financial-model method: Produce period-by-period incremental cash flows, typically monthly or annually: net cash flow_t = incremental attributable cash benefits_t − incremental costs_t. Include one-time implementation/enablement, staff time at fully loaded cost, licenses, integration, support, maintenance, adoption ramp and eventual retirement cost if material. Calculate NPV = initial net cash flow + Σ(net cash flow_t/(1+r)^t) using a period-matched rate; if the discount rate is not approved, show undiscounted cash flow and a rate sensitivity rather than inventing a corporate WACC. Simple ROI = (cumulative attributable benefits − total incremental costs)/total incremental costs; mark undefined if denominator zero. Payback is the first point cumulative cash flow recovers upfront investment; "not reached" is a valid result. Base/downside/upside must have driver-level assumptions and 1- or 2-way sensitivity on the highest-impact uncertain variables. Show periods, units, formula trace, source, confidence and reconciliation checks, not decorative precision.

   Expected outputs
4. 4

   Stress-test delivery and strategic fit

   Evaluate technical and organizational feasibility, access to data, procurement or security barriers, customer/employee adoption, sustainability and operational ownership. Compare alternatives using an explicit decision matrix; flag blockers that override an attractive ROI. Seek relevant senior operator challenge and preserve credible dissent. Differentiate assumptions that can be tested by a small pilot from irreversible commitments. Gate: the recommended option survives the strongest realistic counterargument or identifies a decisive uncertainty.

   V2 practical protocol:
   Decision challenge: Test demand, observed impact, ability to implement, change adoption, time-to-value, security/regulatory and vendor dependence. Use a red-flag register with condition, evidence, severity, mitigation owner and whether it is a hard stop irrespective of financial return. Compare status quo, staged pilot and full investment on consistent dimensions; explain why numerical totals should not override prohibitive constraints. Ask a skeptical operator to challenge adoption ramp and cost of ongoing delivery. When the model is brittle to one unknown, design a short, cheap validation experiment and define the decision rule before authorizing more spend.

   Expected outputs
5. 5

   Present an authorization-ready investment recommendation

   Provide a one-page executive decision memo with recommended option, evidence, economic range, principal risks, strongest rejected alternative and staged investment conditions. Specify a limited next tranche or experiment when uncertainty is high; state explicit stop/reconsider thresholds. Do not make commitments, budget approvals or vendor purchases on behalf of the sponsor. If a decision meeting occurs, capture its actual outcome rather than inventing approval. Gate: management can choose Go, Pilot, Wait or Stop without reopening the entire investigation.

   V2 practical protocol:
   Required memo format (1–2 pages plus optional model): (1) decision requested and owner; (2) recommendation: Go / Pilot / Wait / Stop; (3) competing options and reason for selection; (4) investment amount, horizon, expected financial range and payback/NPV only if computable; (5) three key assumptions with source/confidence; (6) top risks and structural blockers; (7) conditions, owners and evidence to release the next tranche. Put exact calculations in an annex. Distinguish modeled returns from observed results and proposed from approved investment. Quality check: a reviewer can recalculate headline figures from the driver ledger, identify the worst-case downside and find what evidence would change the recommendation.

   Expected outputs